TÜV RHEINLAND ENERGIE UND UMWELT GMBH

Transcription

1 TÜV RHEINLAND ENERGIE UND UMWELT GMBH Report on type approval testing of the F measuring system with PM 2.5 -pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM 2.5 TÜV-report: 936/ /A Cologne, 17 th March The department of Environmental Protection of TÜV Rheinland Energie und Umwelt GmbH is accredited for the following work areas: - Determination of air quality and emissions of air pollution and odour substances; - Inspection of correct installation, function and calibration of continuously operating emission measuring instruments, including data evaluation and remote emission monitoring systems; - Combustion chamber measurements; - Type approval testing of measuring systems for continuous monitoring of emissions and ambient air, and of electronic data evaluation and remote emission monitoring systems; - Determination of stack height and air quality projections for hazardous and odour substances; - Determination of noise and vibration emissions and pollution, determination of sound power levels and execution of sound measurements at wind energy plants according to EN ISO/IEC The accreditation is valid up to DAkkS-register number: D-PL Reproduction of extracts from this test report is subject to written consent. TÜV Rheinland Energie und Umwelt GmbH D Cologne, Am Grauen Stein, Tel: , Fax:

2 Page 2 of 269 Report on type approval testing of the F measuring system with PM 2.5-pre-separator manufactured by DURAG GmbH for the component suspended Blank page

3 Page 3 of 269 Report on type approval testing of the F measuring system with PM 2.5 -pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM 2.5 Instrument tested: F with PM 2.5 -pre-separator Manufacturer: DURAG GmbH Kollaustraße Hamburg Germany Test period: December 2012 until March 2014 Date of report: 17th March 2014 Report number: 936/ /A Editor: Dipl.-Ing. Karsten Pletscher Tel.: karsten.pletscher@de.tuv.com Scope of report: Report: 159 pages Annex Page 160 pp. Manual Page 191 pp. Manual of 76 Pages Total 269 Pages

4 Page 4 of 269 Report on type approval testing of the F measuring system with PM 2.5- Blank page

5 Page 5 of 269 Contents 1. GENERAL AND CERTIFICATION PROPOSAL General Certification proposal Summary of test results TASK DEFINITION Nature of test Objective DESCRIPTION OF THE AMS TESTED Measuring principle Principle of operation AMS scope and setup TEST PROGRAMME General Laboratory test Field test REFERENCE METHOD TEST RESULTS Measured value display Easy maintenance Functional check Setup times and warm-up times Instrument design Unintended adjustment Data output...68

6 Page 6 of General Certification ranges Measuring range Negative output signals Failure in the mains voltage Operating states Switch-over Maintenance interval Availability Instrument software General Repeatability standard deviation at zero point Repeatability standard deviation at reference point Linearity (lack of fit) Sensitivity coefficient of sample gas pressure Sensitivity coefficient of sample gas temperature Sensitivity coefficient of surrounding temperature Sensitivity coefficient of supply voltage Cross-sensitivity Averaging effect Standard deviation from paired measurements Long-term drift Short-term drift Response time Difference between sample and calibration port Converter efficiency Increase of NO 2 concentration due to residence in the AMS

7 Page 7 of Overall uncertainty General Equivalency of the sampling system Reproducibility of the sampling systems Calibration Cross sensitivity Averaging effect Constancy of sample volumetric flow Tightness of the measuring system Methodology of the equivalence check (modules ) Determination of uncertainty between systems under test u bs Calculation of expanded uncertainty between systems under test Application of correction factors and terms Requirements on multiple-component measuring systems RECOMMENDATIONS FOR PRACTICAL USE LITERATURE ANNEX

8 Page 8 of 269 Tables Table 1: Description of test sites...15 Table 2: Technical specifications F (manufacturer s information)...42 Table 3: Overview on software versions during the type approval test...44 Table 4: Field test sites...48 Table 5: Ambient conditions at the field test sites, daily mean values...53 Table 6: Used filter materials...55 Table 7: Results of the internal checks of zero- and reference point...63 Table 8: Certification ranges...71 Table 9: Operation status F Table 10: Error status F Table 11: Determination of availability (without test-related downtimes)...80 Table 12: Determination of availability (incl. test-related downtimes)...80 Table 13: Detection limit PM Table 14: Dependence of zero point on ambient temperature, deviations in µg/m³, mean value of three measurements...93 Table 15: Dependence of sensitivity (foil value) on ambient temperature, deviation in %, mean value of three measurements...93 Table 16: Dependence of measured value on supply voltage, deviation in %...95 Table 17: Concentration mean values, standard deviation, uncertainty range, and reproducibility in the field, measured component PM Table 18: Zero point drift SN & SN , PM 2.5, with zero filter Table 19: Sensitivity drift SN & SN , PM Table 20: Results of the calibration function and analytical function, measured component PM Table 21: Deviation between reference measurement and candidate on days with a relative humidity of > 70 %, measured component PM Table 22: Comparison of the candidates / with the reference device, rel. humidity > 70 %, all test sites, measured component PM Table 23: Results of flow rate checks Table 24: Parameters for total flow measurement (24 h mean), SN & SN Table 25: Results from leakage testing during the field tests Table 26: Uncertainty between candidates u bs for the devices SN and SN , measured component PM Table 27: Overview of equivalence test of F for PM Table 28: Uncertainty between reference devices u ref for PM Table 29: Summary of the results of the equivalence test, SN & SN , measured component PM 2.5 after correction of slope / intercept Table 30: Stability of standard weight Table 31: Stability of the control filters

9 Page 9 of 269 Figures Figure 1: Schematic overview Set-up of the measuring system F , here without actively ventilated sampling inlet tube...31 Figure 2: Overview F Figure 3: Sampling inlet PM 2, Figure 4: Active ventilation of the sampling system during field test...34 Figure 5: F during field test...34 Figure 6: Zero filter for provision of particulate-free air, during field test...35 Figure 7: Reference foil...36 Figure 8: Main menu F Figure 9: Flow scheme menu structure for parameters...39 Figure 10: Overview print out of parameters F Figure 11: Course of PM 2.5 concentrations (reference) at test site Bonn, road junction, winter...49 Figure 12: Course of PM 2.5 concentrations (reference) at test site Bornheim, summer...49 Figure 13: Course of PM 2.5 concentrations (reference) at test site Cologne, autumn...50 Figure 14: Course of PM 2.5 concentrations (reference) at test site Cologne, winter...50 Figure 15: Field test site Bonn, road junction, winter...51 Figure 16: Field test site Bornheim, summer...51 Figure 17: Field test site Cologne, autumn & winter...52 Figure 18: Measurement display for concentration values...58 Figure 19: Data display measured values as table...58 Figure 20: Data display measured values as bar chart...59 Figure 21: Rear side of the F Figure 22: Display of software version here 3.10 in the menu Parameter/Service/SW...82 Figure 23: Zero point drift SN , measured component PM Figure 24: Zero point drift SN , measured component PM Figure 25: Drift of the measured value SN , measured component PM Figure 26: Drift of the measured value SN , measured component PM Figure 27: Flow rate of device SN Figure 28: Flow rate of device SN Figure 29: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, all test sites Figure 30: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Bonn, road junction, winter Figure 31: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Bornheim, summer Figure 32: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Cologne, autumn Figure 33: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Cologne, winter Figure 34: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, all test sites, values 18 µg/m³ Figure 35: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, all test sites, values < 18 µg/m³ Figure 36: Figure 37: Reference device vs. candidate, SN , measured component PM 2.5, all test sites Reference device vs. candidate, SN , measured component PM 2.5, all test sites

10 Page 10 of 269 Figure 38: Reference device vs. candidate, SN , measured component PM 2.5, Bonn, road junction, winter Figure 39: Reference device vs. candidate, SN , measured component PM 2.5, Bonn, road junction, winter Figure 40: Reference device vs. candidate, SN , measured component PM 2.5, Bornheim, summer Figure 41: Reference device vs. candidate, SN , measured component PM 2.5, Bornheim, summer Figure 42: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, autumn Figure 43: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, autumn Figure 44: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, winter Figure 45: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, winter Figure 46: Reference device vs. candidate, SN , measured component PM 2.5, values 18 µg/m³ Figure 47: Reference device vs. candidate, SN , measured component PM 2.5, values 18 µg/m³ Figure 48: Stability of standard weight Figure 49: Stability of the control filters

11 Page 11 of General and certification proposal 1.1 General According to Directive 2008/50/EC dated 21 st May 2008 (replaces air quality framework directive 96/62/EC dated 27 th September 1996 including the related daughter directives 1999/30/EC, 2000/69/EC, 2002/3/EC as well as the Council decision 97/101/EC) on ambient air quality and cleaner air for Europe, the reference method for measuring the PM 10 concentration as per Air quality Determination of the PM 10 fraction of suspended particulate matter Reference method and field test procedure to demonstrate reference equivalence of measurement methods of equality given in EN and the reference method for measuring the PM 2,5 concentration as per Ambient air quality Standard gravimetric measurement method for the determination of the PM 2.5 mass fraction of suspended particulate matter given in EN shall be used. A Member State can, in the case of particulate matter, use any other method which the Member State concerned can demonstrate displays a consistent relationship to the reference method. In that event the results achieved by that method must be corrected to produce results equivalent to those that would have been achieved by using the reference method (2008/50/EC, Annex VI, B). The Guide Demonstration of Equivalence of Ambient Air Monitoring Methods [4] which was developed by an ad-hoc EC working group in January 2010 (Source: describes a method for testing for equivalence of non-standardised measurement methods. The requirements set out in the Guide for equivalence testing have been included in the last revision of the VDI Standards 4202, Sheet 1 and VDI 4203, Sheet 3. In this type approval testing the following limit values were applied: Daily limit DL (24 h) Annual limit AL (1 a) PM 2.5 Not defined 25 µg/m³* as well as for the calculations according to the Guide [4] Limit value PM µg/m³

12 Page 12 of 269 The 2002 VDI guideline 4202, Sheet 1 describes the Minimum requirements for suitability tests for ambient air quality systems. General parameters for the related tests are set out in VDI Standard 4203, Sheet 1 Testing of automated measuring systems General concepts of October 2001 and further specified in VDI 4203, Sheet 3 Testing of automated measuring systems Test procedures for point-related ambient air measuring systems for gaseous and particulate air pollutants of August VDI Standards 4202, Sheet 1 and 4203, Sheet 3 underwent extensive revision and were newly published in September Unfortunately, after this revision there are some ambiguities and contradictions in relation to the type approval testing of particulate measuring systems as far as minimum requirements on the hand and the general relevance of test items on the other hand are concerned. The following test items require clarification: Repeatability standard deviation at zero point no minimum requirement defined Repeatability standard deviation at reference point not relevant to particulate measuring systems Linearity (lack of fit) not relevant to particulate measuring systems Sensitivity coefficient of surrounding temperature no minimum requirement defined Sensitivity coefficient of supply voltage no minimum requirement defined Standard deviation from paired measurements no minimum requirement defined Long-term drift no minimum requirement defined Short-term drift not relevant to particulate measuring systems Overall uncertainty not relevant to particulate measuring systems, covered by In order to determine a concerted procedure for dealing with the inconsistencies in the guidelines, an official enquiry was directed to the competent body in Germany.

13 Page 13 of 269 The following procedure was suggested: As before, the test items 5.3.2, 5.3.7, 5.3.8, , and are evaluated based on the minimum requirements set out in VDI 4202, Sheet 1 of 2002 (i.e. using the reference values B 0, B 1, and B 2 ). The test items 5.3.3, 5.3.4, , and are omitted as they are not relevant to particulate measuring systems. The competent body in Germany approved of the suggested procedure by decisions of 27 June 2011 and 7 October The reference values which shall be used according to the applied guidelines explicitly refer to the measured component PM 10. Therefore, the following reference values are suggested for the measured component PM 2.5 : PM 2,5 B 0 2 µg/m³ 2 µg/m³ B 1 25 µg/m³ 40 µg/m³ B µg/m³ 200 µg/m³ PM 10 (for comparison) B 1 shall merely be adjusted to the level of the limit value for the annual mean.

14 Page 14 of 269 DURAG GmbH has commissioned TÜV Rheinland Energie und Umwelt GmbH to carry out a type approval test of the F measuring system for the component suspended particulate matter PM 2.5. VDI Standard 4202, Sheet 1, Performance criteria for type approval tests of automated ambient air measuring systems Point-related measurement methods for gaseous and particulate air pollutants, September 2010/June 2002 [1] VDI Standard 4203, Sheet 3, Testing of automated measuring systems Test procedures for point-related ambient air measuring systems for gaseous and particulate air pollutants, September 2010/August 2004 [2] Standard EN 14907, Ambient air quality Standard gravimetric measurement method for the determination of the PM 2.5 mass fraction of suspended particulate matter, German version EN 14907: 2005 [3] Guidance document Demonstration of Equivalence of Ambient Air Monitoring Methods, English version of January 2010 [4] The measuring system F operates according to the principle of radiometric measurement. By means of a pump ambient air is sucked in via a PM 2.5 -sampling inlet (sample flow l/min). The dust-laden sampling air is then sucked onto a filter tape. The determination of the collected dust mass on the filter tape is carried out after each sample by the radiometric measuring principle of beta attenuation. The tests were performed in the laboratory and during a field test that lasted several months.

15 Page 15 of 269 The field test which lasted several months was performed at the test sites given in Table 1. Table 1: Description of test sites Bonn, street crossing, winter Bornheim, motorway parking area, summer Cologne, parking lot, autumn Cologne, parking lot, winter Period 02/ / / / / / / /2014 No. of paired values: candidates Characteristics Influenced by traffic Rural structure + motorway Level of ambient air pollution Average to high Urban background Urban background Low to average Average Average The minimum requirements were fulfilled during type approval testing. TÜV Rheinland Energie und Umwelt GmbH therefore suggests its approval as a type approval tested measuring system for continuous monitoring of ambient air pollution for suspended particulate matter PM 2.5.

16 Page 16 of Certification proposal Due to the positive results achieved, the following recommendation is put forward for the notification of the AMS as a performance-tested measuring system: AMS designation: F with PM 2.5 -pre-separator for suspended particulate matter PM 2.5 Manufacturer: DURAG GmbH, Hamburg Field of application: Continuous measurement of the PM 2.5 fraction in ambient air (stationary operation). Measuring ranges during type approval testing: Component Certification range Unit PM 2, µg/m³ Software version: 3.10 Restrictions: None Notes: 1. The requirements according to the guide Demonstration of Equivalence of Ambient Air Monitoring Methods are met for the measured component PM During the type approval testing, the cycle time was 1 h and the sample count was 24, i.e. an automatic filter change was carried out every hour, at which each filter spot has been sampled upon up to maximum 24times. 3. The measuring system shall be operated with an actively ventilated sampling system without sample tube heater. 4. The measuring system shall be operated in a lockable measuring cabinet. 5. The measuring system shall be calibrated on site with the gravimetric PM 2.5 reference method as per EN on a regular basis. 6. This report on the type approval testing can be viewed on the internet at Test report: TÜV Rheinland Energie und Umwelt GmbH, Cologne Report no.: 936/ /A of 17th March 2014

17 Page 17 of Summary of test results Performance criterion Specification Test result Fulfilled Page 4 Requirements on instrument design 4.1 General requirements Measured value display Shall be available. The measuring system has got a measured value display Easy maintenance Necessary maintenance of the measuring systems should be possible without larger effort, if possible from outside Functional check If the operation or the functional check of the measuring system requires particular instruments, they shall be considered as part of the measuring system and be applied in the corresponding sub-tests and included in the assessment Setup times and warm-up times Shall be specified in the instruction manual Instrument design Shall be specified in the instruction manual Unintended adjustment It shall be possible to secure the adjustment of the measuring system against illicit or unintended adjustment during operation Data output The output signals shall be provided digitally and/or as analogue signals Maintenance work can be carried out from the outside with commonly available tools and reasonable time and effort. All functions described in the operator s manual are available, can be activated, and work properly. Setup and warm-up times were determined. The instrument design specifications listed in the operator s manual are complete and correct. The measuring system is secured against illicit or unintentional adjustments of instrument parameters. In addition, the AMS shall be locked up in a measuring cabinet. The test signals are provided analogue (in ma) and digitally (via RS232). yes 57 yes 60 yes 63 yes 65 yes 66 yes 67 yes 68

18 Page 18 of 269 Performance criterion Requirement Test result Fulfilled Page 5. Performance criteria 5.1 General The manufacturer s specifications in the instruction manual shall not contradict the results of the type approval test. No differences between the instrument design and the descriptions given in the manuals were found. yes General requirements Certification ranges Shall comply with the requirements of Table 1 of VDI Standard 4202, Sheet 1. Assessment of AMS in the range of the relevant limit values is possible. yes Measuring range The upper limit of measurement of the measuring systems shall be greater or equal to the upper limit of the certification range. A measuring range of µg/m³ is set as standard. Other measuring ranges are possible. The upper limit of the measuring range of the measuring system is greater than the respective upper limit of the certification range. yes Negative output signals Failure in the mains voltage Negative output signals or measured values may not be suppressed (life zero). Uncontrolled emission of operation and calibration gas shall be avoided. The instrument parameters shall be secured by buffering against loss caused by failure in the mains voltage. When mains voltage returns, the instrument shall automatically reach the operation mode and start the measurement according to the operating instructions. Negative output signals are directly displayed by the AMS and are output correctly via corresponding data outputs. All parameters are secured against loss by buffering. When mains voltage returns the AMS goes back to failurefree operation mode and automatically resumes measuring after reaching the start point of the next measurement cycle (during the type approval test after reaching the next full hour). yes 73 yes Operating states The measuring system shall allow the control of important operating states by telemetrically transmitted status signals Switch-over Switch-over between measurement and functional check and/or calibration shall be possible telemetrically by computer control or manual intervention. The measuring systems can be monitored and operated from an external PC via modem or router. The switch-over between measurement and functional check (here: internal zero point and reference point measurement) can be initiated telemetrically or manually. yes 75 yes Maintenance interval If possible 3 months, minimum 2 weeks. The maintenance interval is determined by the necessary maintenance work and is 4 weeks. yes 78

19 Page 19 of 269 Performance criterion Specification Test result Fulfilled Page Availability Minimum 95 %. The availability was 100 % for SN and 99.7 % for SN without test-related downtimes. Including test-related downtimes it was 92.0 % for SN and 92.0 % for SN yes Instrument software The version of the instrument software to be tested shall be displayed during switch-on of the measuring system. The test institute shall be informed on changes in the instrument software, which have influence on the performance of the measuring system. The version of the instrument software is shown on the display. The test institute is informed on any changes in the instrument software. yes Requirements on measuring systems for gaseous air pollutants General Minimum requirement according to VDI 4202, Sheet 1. The test was carried out on the basis of the performance criteria stated in VDI Standard 4202, Sheet 1 (September 2010). However, the test items 5.3.2, 5.3.7, 5.3.8, , and were evaluated on the basis of the performance criteria stated in the 2002 version of VDI Standard 4202, Sheet 1 (i.e. applying the reference values B 0, B 1, and B 2). The test items 5.3.3, 5.3.4, , and were omitted as they are irrelevant to particulate measuring devices. yes Repeatability standard deviation at zero point The repeatability standard deviation at zero point shall not exceed the requirements of Table 2 in the certification range according to Table 1 of VDI Standard 4202, Sheet 1 (September 2010). For PM: Max. B 0. The tests resulted in detection limits of 0.66 µg/m³ for System 1 (SN ) and 0.75 µg/m³ for System 2 (SN ). yes Repeatability standard deviation at reference point The repeatability standard deviation at reference point shall not exceed the requirements of Table 2 in the certification range according to Table 1 of VDI Standard 4202, Sheet 1 (September 2010). Not applicable. - 87

20 Page 20 of 269 Performance criterion Specification Test result Fulfilled Page Linearity (lack of fit) The analytical function describing the relationship between the output signal and the value of the air quality characteristic shall be linear. Particulate measuring systems for PM 2.5 shall be tested according to performance criterion Calculation of expanded uncertainty between systems under test Sensitivity coefficient of sample gas pressure Sensitivity coefficient of sample gas temperature The sensitivity coefficient of the sample gas temperature at reference point shall not exceed the specifications of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). The sensitivity coefficient of the surrounding temperature at zero and reference point shall not exceed the specifications of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). Not applicable Not applicable Sensitivity coefficient of surrounding temperature The sensitivity coefficient of the surrounding temperature at zero and reference point shall not exceed the specifications of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). For PM: Zero point value for Tu of 15 K between +5 C and +20 C or 20 K between +20 C and +40 C shall not exceed B 0. The measurement value in the range of B 1 shall not exceed ± 5 % for Tu of 15 K between +5 C and +20 C or for 20 K between +20 C and +40 C The maximum dependence of ambient temperature in the range of +5 C to +40 C at zero was -1.0 µg/m³. At reference point, no deviations > 3.0 % in relation to the default value at the temperature of 20 C were observed. yes 91

21 Page 21 of 269 Performance criterion Specification Test result Fulfilled Page Sensitivity coefficient of supply voltage The sensitivity coefficient of the electric voltage at reference point shall not exceed the specifications made in Table 2 of VDI Standard 4202, Sheet 1 (September 2010). For PM: Change in measured value at B 1 maximum B 0 within the voltage interval ( /-20) V. No deviations > -1.4 % for PM 2.5 in relation to the default value at the mains voltage 230 V due to changes in supply voltage were detected. yes Cross-sensitivity The change in the measured value caused by interfering components in the sample gas shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010) at zero and reference point Averaging effect For gaseous components the measuring system shall allow the formation of hourly averages. The averaging effect shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). Not applicable Not applicable Standard deviation from paired measurements The standard deviation from paired measurements under field conditions shall be determined with two identical measuring systems by paired measurements in the field test. It shall not exceed the specifications stated in Table 2 of VDI Standard 4202, Sheet 1 (September 2010). For PM: RD 10 related to B 1. In the field test, the reproducibility for the full dataset was 21 for PM 2.5. yes 98

22 Page 22 of 269 Performance criterion Specification Test result Fulfilled Page Long-term drift The long-term drift at zero point and reference point shall not exceed the requirements of Table 2 in the field test of VDI Standard 4202, Sheet 1 (September 2010) in the field test. For PM: Zero point: within 24 h and within the maintenance interval a maximum of B0. As reference point: within 24 h and within the maintenance interval a maximum 5 % of B1. The maximum deviation at zero point was 1.5 µg/m³ in relation to the previous value and 1.7 µg/m³ in relation to the start value. The sensitivity drift values that were determined during testing are max % for PM 2.5 in relation to the respective start value. yes Short-term drift The short-term drift at zero point and reference point shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010) within 12 h (for benzene 24 h) in the laboratory test and within 24 h in the field test Response time The response time (rise) of the measuring systems shall not exceed 180 s. The response time (fall) of the measuring systems shall not exceed 180 s. The difference between the response time (rise) and response time (fall) of the measuring system shall not exceed 10 % of response time (rise) or 10 s, whatever value is larger. Not applicable Not applicable

23 Page 23 of 269 Performance criterion Specification Test result Fulfilled Page Difference between sample and calibration port Converter efficiency Increase of NO2 concentration due to residence in the AMS The difference between the measured values obtained by feeding gas at the sample and calibration port shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). In the case of measuring systems with a converter, the efficiency of the converter shall be at least 98 %. In case of NO x measuring systems, the increase of NO 2 concentration due to residence in the measuring system shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). Not applicable Not applicable Not applicable Overall uncertainty The expanded uncertainty of the measuring system shall be determined. The value determined shall not exceed the corresponding data quality objectives in the applicable EU Directives on air quality listed in Annex A, Table A1 of VDI Standard 4202, Sheet 1 (September 2010). By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. Please refer to module Requirements on measuring systems for particulate air pollutants General Test according to the minimum requirement stated in Table 5 of VDI Standard 4202, Sheet 1. Furthermore, the particle mass concentration shall be related to a defined volume. The test was carried out according to the minimum requirements set out in Table 5 of VDI Standard 4202, Sheet 1 (September 2010). The determined mass is related to a defined and actively controlled sample volume and thus the particulate mass concentration is determined. yes 112

24 Page 24 of 269 Performance criterion Specification Test result Fulfilled Page Equivalency of the sampling system Reproducibility of the sampling systems The equivalency to the reference method according to EN [T2] shall be demonstrated. This shall be demonstrated in the field test for two identical systems according to EN [T2]. Not applicable to PM 2.5 sampling systems. Please refer to module in this report. Not applicable to PM 2.5 sampling systems. Please refer to module in this report Calibration The systems under test shall be calibrated in the field test by comparison measurements with the reference method according to EN and EN Here, the relationship between the output signal and the gravimetrically determined reference concentration shall be determined as a steady function Cross sensitivity Shall not exceed 10 % of the limit value Averaging effect The measuring system shall allow the formation of 24 h mean values. The time of the sum of all filter changes within 24 h shall not exceed 1 % of this averaging time. A statistical correlation between the reference measuring method and the output signal could be demonstrated. No deviation of the measured signal from the nominal value > -0.6 µg/m³ caused by interference due to moisture in the sample could be observed for PM 2.5. No negative influence on the measured values at varying relative humidity was detected during the field test. The comparability of the candidates with the reference method according to the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods [4] is ensured even for days with a relative humidity of > 70 %. With the described instrument configuration and a measurement cycle of 1 h with a sample count of 24, the measuring system allows the formation of daily mean values based on 24 measurement cycles. yes 115 yes 117 yes 120

25 Page 25 of 269 Performance criterion Specification Test result Fulfilled Page Constancy of sample volumetric flow Tightness of the measuring system Determination of uncertainty between systems under test ubs Calculation of expanded uncertainty between systems under test Application of correction factors and terms 5.5 Requirements on multiplecomponent measuring systems ± 3 % of the rated value during sampling; instantaneous values ± 5 % of the rated value during sampling. Leakage shall not exceed 1 % of the sample volume sucked. Shall be determined according to chapter of the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods in the field test for at two identical systems. Determination of the expanded uncertainty of the candidates according to chapters ff of the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods. If the maximum expanded uncertainty of the systems under test exceeds the data quality objectives according to the European Directive on ambient air quality [7], the application of correction factors and terms is allowed. Values corrected shall meet the requirements of chapter ff. of the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods. Shall comply with the requirements set for each component also in the case of simultaneous operation of all measuring channels. All determined daily mean values deviate less than ± 3 % from the rated value and all instantaneous values deviate less than ± 5 %. The criterion for passing the leakage test, which has been specified by the manufacturer, (flow at blocked inlet max. 10 l/h) proved to be an appropriate parameter for monitoring instrument tightness. The detected maximum leak rate of 1 l/h is less than 1 % of the nominal flow rate of 1000 l/h (16.67 l/min). The uncertainty between the candidates u bs with a maximum of 0.84 µg/m³ for PM 2.5 does not exceed the required value of 2.5 µg/m³. The determined uncertainties WCM for all datasets under consideration lie below the defined expanded relative uncertainty Wdqo of 25 % for suspended particulate matter without application of correction factors. The candidates meet the requirements on data quality of ambient air quality measurements already without application of correction factors. A correction of the slope and the intercept nevertheless leads to a further significant improvement of the expanded measurement uncertainties of the full data set. yes 122 yes 125 yes 128 yes 135 yes 149 Not applicable

26 Page 26 of Task definition 2.1 Nature of test DURAG GmbH has commissioned TÜV Rheinland Energie und Umwelt GmbH to carry out type approval testing of the F measuring system with PM 2.5 -pre-separator. The test was performed as a complete type approval test. 2.2 Objective The measuring system shall determine the concentration of suspended particulate matter PM 2.5 within a concentration range of 0 to µg/m³. The type approval test was carried out in accordance with the current standards for type approval tests and with regard to the most recent developments. The testing was performed with respect to the following guidelines: VDI Standard 4202, Sheet 1, Performance criteria for type approval tests of automated ambient air measuring systems Point-related measurement methods for gaseous and particulate air pollutants, September 2010/June 2002 [1] VDI Standard 4203, Sheet 3, Testing of automated measuring systems Test procedures for point-related ambient air measuring systems for gaseous and particulate air pollutants, September 2010/August 2004 [2] European Standard EN 14907, Ambient air quality Standard gravimetric measurement method for the determination of the PM 2,5 mass fraction of suspended particulate matter, German version EN 14907: 2005 [3] Guidance document Demonstration of Equivalence of Ambient Air Monitoring Methods, English Version: January 2010 [4]

27 Page 27 of Description of the AMS tested 3.1 Measuring principle The ambient air measuring system F is based on the measuring principle of beta attenuation. The air volume is sucked through a glass fiber tape (GF), which separates out the dust particles onto the filter. The volumetric flow is controlled and recorded by a microcontroller. After the intake time, the mass collected on the filter is measured radiometrically. A measuring setup consisting of a Beta radiation source (C-14) and a Geiger Mueller counter tube (GM) is used to do this. The measuring principle for determining the dust mass is based on the fact that Beta rays are weakened as they pass through a material. The intensity of the radiation (pulses/ measuring time) is initially assessed after the rays pass through the unused filter paper. Once it has collected the dust, the intensity of the radiation is measured again. The ratio of the two intensity values is a measure of the quantity of dust collected on the filter spot (assuming homogeneous distribution on the filter surface) and, with a constant cross-sectional area of the loaded filter spot, a measure of the absolute dust mass. The absolute dust mass divided by the quantity of air taken in then gives the dust concentration. The radiometric measuring method is universally applicable, as it determines the mass of the dust within wide limits irrespective of the chemical and physical properties of the dust and the carrier gas. With homogeneous distribution of dust precipitation of the mass m on a filter area A F, up to 5 mg/cm² the relationship is approximately linear: where: d = m/a F µ/ρ ln (n 0 /n) = (µ/ρ) * d in µg/cm² is the dust surface density with dust precipitation in µg on the constant precipitation area in cm² in cm² /g is the mass attenuation coefficient µ in cm -1 is the linear attenuation coefficient of the Beta radiation used ρ in g/cm³ is the density of the absorber material n 0 and n are the Beta particles detected by the counter per minute, without or with the dust, registered electronically as voltage pulses. The pulse rate is a measure of the radiation intensity. The mass attenuation coefficient µ/ρ of the Beta radiation used depends on the electron density of the absorber and is thus proportional to the ratio (Z/A). where: Z Chemical atomic number A Mass number However, as the ratio (Z/A) = is approximately constant for most dusts that occur, in practical terms the Beta radiation attenuation is independent of the chemical composition and particle size distribution of the dust.

28 Page 28 of 269 If the filter area remains constant, because (µ/ρ) = constant the dust mass precipitated on the filter A can be calculated from the radiation attenuation using the following equation: m = A F * (ρ/µ) * ln (n 0 /n) where: m Absolute dust mass in g A F Filter area in cm² As the mass attenuation coefficient (µ/ρ) increases as the maximum Beta energy decreases, determination of mass by Beta absorption measurement is more sensitive the weaker the energy of the Beta radiation used. The dust concentration is calculated from the absolute mass divided by the air volume taken in: c = m / Q where: c Dust concentration in g/m³ Q Intake air volume in m³

29 Page 29 of Principle of operation The ambient air measuring system F is based on the measuring principle of beta attenuation. The particulate sample passes the PM 2.5 sampling inlet with a sample flow rate of 1 m³/h (=16.67 l/min) and reaches the measuring device F via the sampling tube. Within the scope of the type approval test, the measuring systems were operated with an actively ventilated sample inlet tube and without sample tube heating. When using the actively ventilated sampling inlet tube, ambient air is carried continuously through the outer sheath tube via a fan unit, so that the sample inlet tube itself inside is kept at the ambient air temperature level up to the measuring part in the device. The power supply of the fan unit is carried out by the device itself. An additional passive insulation of the outer tube can be helpful in case of extreme differences between ambient temperature and temperature at the installation site of the device. The measuring system itself is constructed in a compact way (refer to Figure 1). Except for the sampling tube (sample inlet tube, sampling head), the meteorological sensor for the measurement of the ambient pressure and the ambient temperature and the set-up of the active ventilation of the sample inlet tube, all components are installed in one housing. The measuring device is controlled by a microcontroller board. The filter tape is transported from the supply reel to the take-up reel by a stepper motor. The Geiger Mueller counter tube uses the attenuation of the radiation intensity emitted by the C-14 radiation source to determine the increase in mass on the filter paper. The air is taken in by the pump, and the volumetric flow is measured by the volume sensor and regulated to a constant 1000 l/h by adjusting the bypass valve to the appropriate position. Electronics control the measuring processes, allowing user-friendly operation on a graphical touchscreen, and store the measured values. As an option, the device can protect the measured dust samples against smearing and loss with a cover film to allow subsequent laboratory investigation of dust constituents. This option was not included in the type approval test. During normal measurement, an unloaded filter spot is transported between the C-14 source and the GM counter tube at the beginning of the measurement. The beam intensity is then measured for 300 seconds, i.e. the pulses generated by the GM counter tube are evaluated as a measure of the detected beta radiation. The filter adapter is then opened and the filter tape is transported until this evaluated filter area is in the extraction position. The filter adapter is then closed again. The extraction process begins. The extraction time corresponds to the programmed cycle time (minus the measuring times). After completion of sampling, the filter adapter is opened again and the filter paper is moved back to its original position below the counter tube. The filter adapter closes and the beam intensity is measured for another 300 seconds. The measured count rates are used to determine the dust mass collected on the filter. The dust concentration is calculated using the amount of air extracted, as shown in chapter 3.1.

30 Page 30 of 269 The sampling time corresponds to the respective programmed cycle time and the programmed sample count minus the measuring time respectively the time for filter tape movements. With the help of the sample count, a multiple sampling on a filter spot can be set. It can be set between 1 (=for each cycle a new filter spot) and 24 (=one filter spot is sampled upon for 24times). The sampling time is thus: For cycle time 60 min and sample count 1: 60 min (2 x 300 s measuring time s for filter tape movements) = 48 min In case of a sample count >1, the measurement after sampling serves for the calculation of the measured value of the finished cycle as well as a start measurement for the following cycle, i.e. only one radiometric measurement is necessary per cycle. During the type approval test a cycle time of 60 min with a sample count of 24 was set. The sampling time is then: Cycle 1: 60 min (2 x 300 s measuring time s for filter tape movements) = 48 min Cycle 2-24: 60 min (1 x 300 s measuring time s for filter tape movements) = 53 min The measuring results are shown in the display and are also available as a 4-20mA analogue output signal and via the RS232 data interface (e.g. using Bavaria-Hessen protocol / Gesytec). All measured values from the last 9 months are stored and can be viewed on the display or via the serial interface. A download of the measured values, the error messages as well as the system parameters via this interface is easily possible with a terminal software As an option, you can print on the filter paper to identify the collected dust sample (e.g. for dust content analysis). The latter option was not part in the type approval test.

31 Page 31 of 269 Figure 1: Schematic overview Set-up of the measuring system F , here without actively ventilated sampling inlet tube

32 Page 32 of AMS scope and setup The measuring system consists of a PM 2.5 -sampling inlet, the meteorological sensor, the sample inlet tube with active ventilation, the respective measuring F incl. the glass fiber filter tape, the corresponding lines and cables as well as adapters, the roof penetration incl. flanges as well as the manual in German / English language. Figure 2 gives an overview on the measuring system F Sampling head (example) Radiation shield with sensor for measurement of temperature and relative humidity Double-walled sample tube Active ventilation of the sample tube Sheath air out Drain hose Measurement device with touchscreen Figure 2: Overview F

33 Page 33 of 269 As a sampling inlet, a PM 2,5 sampling inlet (manufacturer: Comde-Derenda GmbH, Stahnsdorf, Germany) is used. Alternatively also TSP- or PM 10 -sampling inlets can be used. Figure 3: Sampling inlet PM 2,5 The connection between sampling inlet and measuring device is carried out with an actively ventilated sampling inlet tube. When using the actively ventilated sampling inlet tube, ambient air is carried continuously through the outer sheath tube via a fan unit, so that the sample inlet tube itself inside is kept at the ambient air temperature level up to the measuring part in the device. The length of the sampling inlet tube during the type approval test was 2 m, alternatively also standard lengths of 1 m and 3 m are available.

34 Page 34 of 269 Figure 4: Active ventilation of the sampling system during field test Figure 5: F during field test

35 Page 35 of 269 For external check of the zero point of the measuring system, a zero filter is mounted at the device inlet. The use of that filter allows the provision of particulate-free air. Figure 6: Zero filter for provision of particulate-free air, during field test

36 Page 36 of 269 For external check of the radiometric measurement, the instrument manufacturer provides a reference foil. Figure 7: Reference foil

37 Page 37 of 269 The operating of the measuring system is carried out on the front of the measuring device via a touchscreen-display. The user can obtain measured data and system information, change parameters as well as perform tests for check on proper functioning of the measuring system. Additional information Action Remaining time for actions Measuring result Current count rate of GM tube or Volumetric flow Buttons Figure 8: Main menu F After switching on the instrument, the main menu appears. Different information is displayed here, as Current instrument status (measurement; n-th measurement (cycle number in case of multiple sampling); standby) Additional information e.g. status during current cycle (0-Rate )

38 Page 38 of 269 Via three keys, the following actions are available: Switch between measuring, data display and setup mode. Go to a submenu Open maintenance menu Via the key Measuring mode: it can be switched between the following menus during normal operation: Dust measurement and display of results, display of activities, perfor mance of service actions. Data display mode: Display the measured values in graphs or tables Setup mode: Display and edit the parameters In the setup mode, all parameters can be viewed and modified after password entry. Figure 9 gives an overview on the menu structure of the parameters. Single parameters are described in details in the instrument manual in chapter

39 Page 39 of 269 Password Measurement Parameter Sub parameter Adjust Interface Date/Time Service Password entry Display the measured / saved values Display and enter the main parameters Display and enter the secondary parameters Correct the input and output signals Set the interface parameters Set the date and time Basic operating functions, troubleshooting F Password Measurement Parameter Sub Parameter Mass [µg] 0-Rate [1/min] M-Rate [1/min] Output 1 [ma] Output 2 [ma] Air temperat.[c] Air press. [hpa] rel humidity [%] Filter adapt.[c] Sampling tube[c] Cycle time Sample count Replacem. T [C] Replacem. [hpa] Replacem. rh [%] Mode start FS conc [µg/m3] Mode 20mA Out 1 Mode 20mA Out 2 Language FS reference[µg] Integration Mode temperature Filter adapt. [C] Samp. tube set[c] Mode zeroizer Mode Synch h:00 max. Mass [µg] Adjust Interface Date/Time Service Offset mass [µg] Span mass Span Volume Offset temp. [C] Offset p [hpa] Mode RS232 Gesytec No. Baud Parity RS484 inactive Hour Minute Day Month Year Password ADC Rowvalue Inputs Factory teach-in Options Sensors SW Serial no.device Serial no. CPU Q-Check Figure 9: Flow scheme menu structure for parameters By pushing the key, additional information can be obtained, e.g. during measuring mode additional information like standardized volume flow, ambient temperature, rel. humidity and ambient pressure.

40 Page 40 of 269 By pushing the key, the maintenance menu is activated after entry of the respective password. The following actions can be carried out: Open filter housing Close filter housing Filter tape forwards Filter tape backwards Start reference measurement Start zero measurement Start reference foil measurement In manual mode, also: Start measurement Besides the direct communication via touchscreen-display, it is also possible to communicate with the measuring system via RS232. During the type approval test work, the measuring systems have been accessed especially for download of internally saved data via RS232 and the terminal software HyperTerminal.

41 Page 41 of 269 Figure 10 gives an overview on the parameters of the candidates during the type approval test (with example of device SN ). Figure 10: Overview print out of parameters F

42 Page 42 of 269 Table 2 contains a list of important technical specifications of the ambient air measuring system F Table 2: Technical specifications F (manufacturer s information) Dimensions / Weight Power supply Power consumption Ambient conditions F Measuring device 482 x 530 x 320 mm / 31 kg Sampling tube Single- or double-walled sampling tube, length 1, 2 m or 3 m Sampling inlet According to manufacturer, during type approval test Comde-Derenda GmbH, sampling inlet PM2,5 1,0m³/h for tube connection 16 mm Analyser: 230 V / 50 Hz or 115 V / 60 Hz approx. 400 W Temperature +5 to +40 C (during type approval test) Humidity Non-condensing Sample flow rate (Inlet) Radiometry Source Detector Parameter filter change l/min = 1 m³/h 14 C-surface emitting source, <450 kbq (< 12,5 µci) Geiger-Müller-end window counter tube Filtertape Glass fiber filter, 30 m or 45 m Measuring cycle (cycle time) 15 min - 24 h During type approval test: 1 h (24 changes per day) Sample count (Multiple sampling) 1 24 During type approval test: 24 (max. 24 cycles per filter spot) Max. dust mass per filter spot Dependent on dust constitution, selectable, during type approval test 400 µg Parameter sample conditioning Conditioning of sampling inlet tube Active ventilation Buffer capacity Data (internal) Dependent on size of SD-card, access via menu directly at device on data over the last 9 months

43 Page 43 of 269 F Instrument input and -output Protocols Status signals / Error messages Output: Analogue: 4-20 ma Digital RS232 Communication with PC via RS232 Bayern-Hessen, Gesytec available, overview refer to chapter 6 table 3 (status signals) respectively chapter 11 table 7 (error messages) of the user s manual

44 Page 44 of Test programme 4.1 General The type approval test was carried out with two identical devices with the serial numbers SN and SN The test was started in December 2012 with the software version Within the frame work of the software maintenance process, the software was in the meanwhile developed and optimised up to version The manufacturing site of the instrument manufacturer DURAG GmbH in Hamburg, Germany is part of the regular surveillance according to Standard EN since Since 2013 also the measuring system F is covered in the audit scope. According to the requirements of the Standard EN , the applied modifications of the software have been described, assessed, the type of modification classified and the information submitted to the test house. The following modifications to the software between version 3.07 and 3.10 have been applied in detail: Table 3: Overview on software versions during the type approval test Version Description of modifications Type of modification Assessment test house 3.07 Start version A software dongle was implemented to hinder copying of hardware. The time, shown on the display, can also be shown with seconds. For the parameter Cycle time, the value 1/2 h was not permanently saved (all other settings worked flawless). Instead the cycle time was 1 h after switching on. The error was corrected. Renaming parameter Modus Start h:00 to Modus Synch h:00 Parameter Steps sens./tube and Pitch dust spot moved from menu point Sub Parameter to Service/Options. Functional extension Functional extension Bug elimination Cosmetic Cosmetic No impact No impact No impact No impact No impact

45 Page 45 of The parameter Q-Check is not delivered anymore when reading out the parameters via the serial interface During the switch-on -message on the display, the date of the software version is not displayed anymore (it could cause confusion with the current date) Cosmetic Cosmetic No impact No impact The applied modifications are for functional extensions, bug elimination and for cosmetic maintenance. They don t have any impact on the measuring system. With the start of the third of in total four comparison campaigns, the software version was upgraded from 3.07 to 3.10 and the instruments were operated with this software for the comparison campaigns Cologne, autumn and Cologne, winter. The test comprised of a laboratory test for the assessment of performance characteristics as well as a field test, conducted over several months and at various field sites. All obtained concentrations are given in µg/m³ (operating conditions). In the following report, the performance criteria according to the considered standards [1, 2, 3, 4] are stated in the caption of each test item with number and wording.

46 Page 46 of Laboratory test The laboratory test was carried out with two identical devices of the type F with the serial numbers SN and SN In conformity with the applicable standards [1, 2], the following performance criteria were tested in the laboratory: Description of device functions Determination of detection limit Dependence of zero point / sensitivity on ambient temperature Dependence of sensitivity on mains voltage In the laboratory test, the following devices were used for the determination of performance characteristics climatic chamber (temperature range from -20 C to +50 C, accuracy better than 1 C) Isolation transformer 1 mass flow meter Model 4043 (Manufacturer: TSI) Zero filter for external zero point control Reference foils The recording of measurement values was performed within the device. The recorded data were read out via RS232 interface with the help of HyperTerminal. The results of the laboratory tests are summarised in chapter 6.

47 Page 47 of Field test The field test was carried out with two identical measuring systems: System 1: SN System 2: SN The following performance criteria were tested in the field: Comparability of the systems under test according to the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods Comparability of the systems under test with the reference method according to the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods Calibration capability, analytical function Reproducibility Drift of zero point and of sensitivity Leak tightness of the sampling system Dependence of the measured values on sample humidity Maintenance interval Availability Consistency of sample volume flow Total uncertainty of tested systems The following auxiliary devices were used during the field test: TÜV Rheinland measuring cabinet, air conditioned to approx. 20 C Weather station (WS 500 of ELV Elektronik AG) for the detection of meteorological parameters such as ambient temperature, atmospheric pressure, humidity, wind velocity, wind direction and amount of precipitation. 2 reference measuring systems LVS3 for PM 2.5 as per item 5 1 gas meter, dry 1 mass flow meter Model 4043 (Manufacturer: TSI) Power consumption measuring device type Metratester 5 (manufactured by Gossen Metrawatt) Zero filter for external zero point checks Reference foils During the field test, two candidates and two reference systems for PM 2.5 were operated simultaneously for a period of 24 hours. The reference system operates discontinuously, that is to say the filter needs to be changed manually after sampling.

48 Page 48 of 269 During the testing, the impaction plates of the PM 2.5 sampling inlets of the reference systems were cleaned and lubricated with silicone grease approx. every 2 weeks in order to ensure a safe separation and deposition of particulates. The PM 2.5 sampling inlets of the candidates were cleaned approx. every 4 weeks. The sampling inlet shall always be cleaned in accord with the instructions provided by the manufacturer. Local concentrations of suspended particulate matter shall also be considered in this procedure. Before and after each change of test site, the flow rate was tested on each candidate as well as on each reference system with a dry gas meter and a mass flow meter, which connects to the system inlet via hose line. Measuring sites and AMS placement For the field test, the measuring systems were set up in such a way that only the sampling heads were installed on the outside of the measuring cabinet above its roof. The central units of both candidates were placed within the air-conditioned measuring cabinet. The entire reference equipment (LVS3) was installed outdoors on the roof of the cabinet. The field test was carried out at the following test sites: Table 4: Field test sites No. Test site Period Characterisation 1 Bonn, road junction, 02/ /2013 Influence of traffic winter 2 Bornheim, summer 05/ /2013 Rural structure + influence of traffic 3 Cologne, autumn 09/ /2013 Urban background 4 Cologne, winter 01/ /2014 Urban background

49 Page 49 of 269 Figure 11 to Figure 14 show the course of PM concentrations at the measuring locations in the field as recorded by the reference measuring systems. 80 Conc. PM2.5 (Ref.) in µg/m³ /02/ /03/ /03/ /03/ /03/ /04/ /04/ /04/ /04/ /05/2013 Figure 11: Course of PM 2.5 concentrations (reference) at test site Bonn, road junction, winter /05/ /05/ /05/ /06/ /06/ /06/ /06/2013 Conc. PM2.5 (Ref.) in µg/m³ 02/07/ /07/ /07/ /07/2013 Figure 12: Course of PM 2.5 concentrations (reference) at test site Bornheim, summer

50 Page 50 of Conc. PM2.5 (Ref.) in µg/m³ /09/ /09/ /09/ /09/ /10/ /10/ /10/ /10/ /10/ /11/ /11/ /11/ /11/ /12/ /12/2013 Figure 13: Course of PM 2.5 concentrations (reference) at test site Cologne, autumn /01/ /01/ /01/ /02/ /02/2014 Conc. PM2.5 (Ref.) in µg/m³ 22/02/ /03/2014 Figure 14: Course of PM 2.5 concentrations (reference) at test site Cologne, winter

51 Page 51 of 269 The following figures show the measuring cabinet at the field test sites Bonn (road junction), Bornheim (motorway parking lot) and Cologne (parking lot). Figure 15: Field test site Bonn, road junction, winter Figure 16: Field test site Bornheim, summer

52 Page 52 of 269 Figure 17: Field test site Cologne, autumn & winter In addition to the measuring systems for the measurement of ambient air pollution through suspended particulate matter, a data acquisition system for meteorological parameters was installed on the cabinet/at the test site where the measurement was carried out. Ambient temperature, ambient pressure, humidity, wind velocity, wind direction, and the amount of precipitation were monitored continuously. 30-minutes mean values were stored. The cabinet setup and the arrangement of the sample probes had the following dimensions: Height of cabinet roof: 2.50 m Sampling height for tested system 0.50 m / 0.51 m above cabinet roof Sampling height for reference system 3.00 / 3.01 m above ground Height of wind vane: 4.5 m above ground The following Table 5 therefore contains an overview of the most important meteorological parameters that have been obtained during the measurements at the 4 field test sites as well as an overview of the concentrations of suspended particulate matter during the test period. All single values are provided in annexes 5 and 6.

53 Page 53 of 269 Table 5: Ambient conditions at the field test sites, daily mean values Bonn, winter Bornheim, summer Cologne, autumn Cologne, winter Number of value pairs Reference PM 2.5 PM 2.5 ratio in PM 10 [%] Range Mean value Ambient temperature [ C] Range Mean value Ambient pressure [hpa] Range Mean value Rel. humidity [%] Range Mean value Wind velocity [m/s] Range Mean value Amount of precipitation [mm/d] Range Mean value

54 Page 54 of 269 Sampling duration According to Standard EN 14907, the sampling time shall be 24 h ± 1 h. During the field test, a sampling time of 24 h was set for all devices (10:00 10:00 (Cologne) and 7:00 7:00 (Bonn, Bornheim)). Data handling Before the respective analyses for each test site were carried out, the paired reference values determined during the field test were subject to a statistical outlier test according to Grubbs (99 %) in order to prevent any effects of evidently implausible data on the test results. Value pairs identified as significant outliers may be discarded from the pool of values as long as the critical value of test statistic does not fall below the target. According to the Guide [4] of January 2010, not more than 2.5 % of data pairs shall be determined as outliers and discarded. As far as candidates are concerned, the measured values are usually not discarded unless there are proven technical reasons for implausible values. Throughout the testing no values measured by the candidates were discarded. The statistical outlier test according to Grubbs (99 %) did not identify any significant outlier for paired measurements for all test sites. Thus no paired measurements have been discarded for the PM 2.5 reference measurement.

55 Page 55 of 269 Filter handling mass determination The following filters were used in the type approval test: Table 6: Used filter materials Measuring system Filter material, type Manufacturer Reference systems LVS3 Emfab, 47 mm Pall The filters were handled in compliance with Standard EN Details on filter handling and weighing processes are describes in appendix 2 of this report.

56 Page 56 of Reference method In accordance with Standard EN 14907, the following devices were used in the testing: 1. as reference device for PM 2.5 : Small Filter Device Low Volume Sampler LVS3 Manufacturer: Ingenieurbüro Sven Leckel, Leberstraße 63, Berlin, Deutschland Date of construction: 2007 and 2010 PM 2.5 sampling head During the testing, two reference systems for PM 2.5 were operated simultaneously with a flow rate of 2.3 m³/h. Under real operating conditions the volume flow control accuracy is < 1 % of the nominal flow rate. The sampling head of the small filter device LVS3 sucks in the sample air via a rotary vane vacuum pump. The sample volume flow is then measured by means of a measuring orifice between filter and vacuum pump. The suctioned air then streams out of the pump via a separator for the abrasion of the rotary vanes and towards the air outlet. As soon as the sampling is complete the electronic measurement equipment displays the sucked-in sample air volume in standard or operating m³. The PM 2.5 concentrations were determined by dividing the amount of suspended particulate matter on each filter that had been determined gravimetrically in the laboratory by the respective sampling volume in operating m³.

57 Page 57 of Test results Measured value display The AMS shall have a means to display the measured values. 6.2 Equipment Additional equipment is not required. 6.3 Method It was checked whether the AMS has a means to display the measured values. 6.4 Evaluation The measuring system has got a measured value display. It shows the respective concentration value of the last cycle. Furthermore in the data display mode, the stored measured values can be easily displayed in either a table display or a bar chart display. 6.5 Assessment The measuring system has got a measured value display. Performance criterion met? yes

58 Page 58 of Detailed presentation of test results Figure 18 shows the display with the current concentration value of the respective last cycle. Figure 19 and Figure 20 show the display of the stored measured values as table and as bar chart. Figure 18: Measurement display for concentration values Figure 19: Data display measured values as table

59 Page 59 of 269 Figure 20: Data display measured values as bar chart

60 Page 60 of Easy maintenance Necessary maintenance of the measuring systems should be possible without larger effort, if possible from outside. 6.2 Equipment Additional equipment is not required. 6.3 Method Necessary regular maintenance work was carried out according to the instructions given in the manual. 6.4 Evaluation The operator shall carry out the following maintenance work: 1. Check of system status: The system status can be monitored and controlled directly on the device itself or online via RS232-interface and Gesytec-Protocol. 2. The sampling inlet shall be cleaned according to the instructions provided by the manufacturer. Local concentrations of suspended particulate matter shall be taken into account (during type approval testing approx. every 4 weeks). 3. Check of the filter tape stock A filter tape with 45 m length is lasting theoretically for 30,000 measurement cycles in case of a cycle time of 1 h and a sample count of 24 (setting during type approval test), which equals to 1,250 days. As - because of a possible exceedance of the maximum allowed mass per filter spot of 400 µg, which is dependent on the PM concentration level under practical conditions - a new filter spot might be used earlier than by reaching the 24times sampling, the time for consumption of the filter tape might be reduced accordingly. In case of a cycle time of 1 h and a minimum sample count of 1 (i.e. for each cycle a new filter spot is used), there are 1,250 measurement cycles possible, which equals 52 days. Thus it is recommended to check the filter tape stock during each visit of the measuring system for regular maintenance (e.g. in the framework of cleaning the sampling inlet). 4. According to the manufacturer. the pump needs maintenance approx. every 6 weeks after one year of operation, i.e. the filters must be blown out and the blade height has to be checked and the blades to be replaced if necessary. 5. According to CEN/TS [8], a check of the sensors for ambient temperature and ambient pressure should be done every 3 months. 6. According to CEN/TS [8], a check of the sampling flow rate should be done every 3 months. 7. Within the framework of the check of the sampling flow rate, the tightness should also be checked every 3 months. 8. The filter adapter, the transport reel and the pressure rollers have to be cleaned every 6 months. 9. The waste air filter and the hose connections have to be checked and if necessary blown out every 6 months. 10. The pump filter and seals shall be replaced once a year.

61 Page 61 of Once per year, the meteorological sensor has to be sent to the manufacturer for recalibration. Furthermore it is recommended to check the radiometric measurement with the help of the reference foil once a year. 12. During the annual maintenance, the cleaning of the sampling tube needs also to be considered. Maintenance work shall be carried out according to the instructions provided in the manual (chapter 5.3 and 10). In general, all work can be carried out with commonly available tools. 6.5 Assessment Maintenance work can be carried out from the outside with commonly available tools and reasonable time and effort. The replacement of the filter tape, the maintenance work at the pump as well as the operations described in item 7ff shall only be performed when the device is on standstill. These operations incur every 6 weeks (pump), every 6 months (cleaning) respectively dependent on the set cycle time and sample count (change of filter tape). In the meantime, maintenance work is limited to the check of contaminations, plausibility and possible status/error messages. Performance criterion met? yes 6.6 Detailed presentation of test results During the testing, work on the devices was carried out on the basis of operations and work processes described in the manuals. By adhering to the described procedures no difficulties were observed. Up to this point, all maintenance could be carried out without difficulty and with conventional tools.

62 Page 62 of Functional check If the operation or the functional check of the measuring system requires particular instruments, they shall be considered as part of the measuring system and be applied in the corresponding sub-tests and included in the assessment. Test gas units included in the measuring system shall indicate their operational readiness to the measuring system by a status signal and shall provide direct as well as remote control via the measuring system. 6.2 Technical equipment Operator s manual, zero filter, reference foil. 6.3 Method The system status is monitored continuously and problems are indicated by a series of different status messages. Parameters, which are important for the correct performance (e.g. flow rate), can be viewed directly on the instrument display and/or recorded continuously during data recording. The measuring system offers for the purpose of a functional check, the possibility of an internal check of zero- and reference point. Both tests can be accessed directly via the maintenance menu. A telemetric control is also possible. During the internal check of the zero point ( zero measurement ), the functionality of the source and the Geiger-Müller-tube is checked. Without filter tape transport, a 0-rate and a M- rate is measured and the corresponding determined mass is calculated. The result shall not exceed a measured value of 10 µg as absolute value. During the internal check of the reference point ( reference measurement ), also a 0-rate and a M-rate is measured without filter tape transport. The M-rate is evaluated 1/6 weaker, which corresponds to a mass of 444 µg (if the parameter offset is set to 0 and the parameter span is set to 1 in the device). This measured value should be obtained with a tolerance of ± 20 µg. If the offset- and span parameters differ from 0 respectively 1, this has to be considered accordingly. Furthermore the zero point of the measuring system can also be checked externally by applying a zero filter to the instrument s inlet. The use of this filter allows the provision of particulate-free air. During the testing, the zero point was determined using a zero filter approx. every 4 weeks. For the external check of the radiometric measurement, the instrument manufacturer supplies a reference foil. By using the reference foil, only mass values can be determined. In the course of the testing, an external check of the stability via reference foil was performed regularly over the entire field test period.

63 Page 63 of Evaluation All functions described in the operator s manual are available or can be activated. The current instrument status is continuously monitored and different warning messages are displayed in the case of problems. For functional check, the measuring system offers the possibility of an internal check of zeroand reference point. Both tests can be accessed directly via the maintenance menu or telemetrically. The internal checks of zero and reference point have been explicitly checked again on March 15, 2014 with the following results: Table 7: Results of the internal checks of zero- and reference point SN SN Zero measurement Reference measurement Nominal Actual Nominal Actual 10 µg 1 µg 10 µg 4 µg 444 µg ± 20 µg 446 µg 444 µg ± 20 µg 444 µg For both devices the determined deviations were within the permissible limits stated by the manufacturer. An external check of the zero point is possible at any time with the help of a zero filter. An external check of the radiometric measurement with the help of the reference foil is also possible at any time. 6.5 Assessment All functions described in the operator s manual are available, can be activated, and work properly. The current instrument status is continuously monitored and different warning messages are displayed in the case of problems. For functional check, the measuring system offers the possibility of an internal check of the zero and the reference point. Both tests can be accessed directly via the maintenance menu or telemetrically. The results of the external zero point checks by means of zero filter that were carried out during the field tests as well as checks of the stability of the radiometric measurement, that were carried out periodically are described in Chapter Long-term drift in this report. Performance criterion met? yes 6.6 Detailed presentation of test results See chapter Long-term drift

64 Page 64 of Setup times and warm-up times The AMS setup and warm-up times shall be stated in the manual. 6.2 Equipment A timer was provided additionally. 6.3 Method The measuring systems were activated according to the manufacturer s specifications. The amounts of time required for setup and warm-up were recorded separately. Structural measures taken before installation, like for instance the opening of the cabinet roof, have not been assessed here. 6.4 Evaluation The setup time comprises the time needed for all necessary works from system installation to start-up. The measuring system must be installed not subject to weather conditions, e.g. in an airconditioned cabinet. Additionally the installation of the actively ventilated sampling system through the roof requires extensive structural measures at the installation site. A nonstationary application is therefore only to be undertaken together with the associated peripheral devices. The following steps are required for the installation of the measuring system: Unpacking and Installation of the AMS (in a rack or on a table) Installation of the double-walled, actively ventilated sampling system and the PM sampling inlet Installation of meteorological sensor (in close proximity to the sampling inlet) Connection of all supply and control lines Power connection Optional connection of peripheral recording- and control systems (data logger, PC) to the respective interfaces Power-up of AMS Insertion of filter tape The execution of these operations and thus the set-up time takes approx. 1 h.

65 Page 65 of 269 The warm-up time is the time between the start of operation of the measuring system and the point when it is ready for measurement. After switching on the system, the system is in waiting position until the set start time for the next measurement cycle is reached. The following actions are necessary for the initial start-up: Check of system settings, e.g. cycle time, sample count, start mode, temperature mode, mode Synch h:00 as well as date and time Check and, if required, adjustment of the meteorological sensor and the ambient pressure measurement Leak test Check and, if required, adjustment of the flow rate If necessary, external check of the radiometric mass determination with the reference foil Duration: approx. 1 hour In case of restarting operation after a short downtime, e.g. after a power outage, the above mentioned steps can be omitted with exception of a check of the system parameters, a plausibility test of the sensor values and an inspection of possible status/error messages. If required, potential changes in the basic parameter setting of the measuring instrument can likewise be performed in just a few minutes by staff familiarised with the instruments. 6.5 Assessment Setup and warm-up times were determined. The measuring system can easily be operated at various measuring sites. The setup time amounts to approximately 1 h. The warm-up time amounts to approx. 1 h in case of initial start-up respectively the waiting time until the start for the next measurement cycle in case of re-start after short downtime. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

66 Page 66 of Instrument design The instruction manual shall include specifications of the manufacturer regarding the design of the measuring system. These elements are: Instrument shape (e.g. bench mounting, rack mounting, free mounting) mounting position (e.g. horizontal or vertical mounting) safety requirements dimensions weight power consumption. 6.2 Equipment Additionally, a measuring device for recording the energy consumption and scales were used to test this performance criterion. 6.3 Method The supplied instruments were compared to the descriptions in the manuals. The specified energy consumption is determined over a 24 h-standard operation during the field test. 6.4 Evaluation The measuring system shall be installed in horizontal position (e.g. on a table or in a rack), not subject to weather conditions. The temperature at the installation site must be in the range of 5 C to 40 C. Direct sunlight respectively a direct exposure to heater or air conditioner shall be omitted. Dimensions and weight of the AMS match the information given in the operator s manual. According to the manufacturer, the energy requirements of the AMS with the inserted pump are about 400 W at maximum for the complete system. During a 24 h test the total power demand of the AMS was determined. During this test, the stated value was not exceeded at any time. 6.5 Assessment The instrument design specifications listed in the operator s manual are complete and correct. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

67 Page 67 of Unintended adjustment It shall be possible to secure the adjustment of the measuring system against illicit or unintended adjustment during operation. 6.2 Technical equipment No additional tools are required here. 6.3 Method The measuring system is operated via touch screen display on the front site of the AMS. A change of parameters or the adjustment of sensors is only possible by typing in several key sequences. The measuring system also has a password protection. There are three different passwords available: Password1: For changing parameters ( Parameter mode) Password2: For performing service actions Password3: System password Without knowledge of password1, instrument parameters can be viewed, but not changed. As an outside installation of the measuring device is not possible, additional protection is given by installation at locations, to which unauthorised people have no access (e.g. locked measuring cabinet). 6.4 Evaluation Unintended and unauthorised adjustment of instrument parameters can be avoided by password protection. Moreover, additional protection against unauthorised intervention is given by installing the system in a locked measuring cabinet. 6.5 Assessment The measuring system is secured against illicit or unintentional adjustments of instrument parameters. In addition, the AMS shall be locked up in a measuring cabinet. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

68 Page 68 of Data output The output signals shall be provided digitally (e.g. RS232) and/or as analogue signals (e.g. 4 ma to 20 ma). 6.2 Equipment Data logger Yokogawa (for analogue signal), PC with Hyper-Terminal software 6.3 Method The test was carried out using a data logger of the type Yokogawa (analogue output, only exemplary test in lab) and a PC with Hyper-Terminal software (via RS232-interface). During the type approval test, the measuring system was connected to a PC via RS232 and the data was downloaded on the PC. The AMS also offers the possibility to output analogue signals the functionality was checked exemplarily during the laboratory test. 6.4 Evaluation The measured signals are offered as follows on the rear side of the instrument: Analogue: 4-20 ma concentration range selectable Digital: RS 232-interface 6.5 Assessment The test signals are provided analogue (in ma) and digitally (via RS232). Connection of additional measuring and peripheral devices via the corresponding ports is possible. Performance criterion met? yes

69 Page 69 of Detailed presentation of test results Figure 21 shows the instrument s rear side with the various data outputs. Meteorology sensor input Temperature sensor input Power plug for heater Mains infeed Data outputs Device fan 24 VDC for external fan exhaust air Figure 21: Rear side of the F

70 Page 70 of General The manufacturer s information provided in the operator s manual shall not contradict the findings of the type approval test. 6.2 Equipment Not required here. 6.3 Method The test results are compared with the information given in the manual. 6.4 Evaluation Instances where the first draft of the manual deviated from the actual design of the instrument have been corrected. 6.5 Assessment No differences between the instrument design and the descriptions given in the manuals were found. Performance criterion met? yes 6.6 Detailed presentation of test result For this module, refer to item 6.4.

71 Page 71 of Certification ranges The certification range over which the AMS will be tested shall be determined. 6.2 Equipment No additional tools are required here. 6.3 Method The certification range over which the AMS will be tested shall be determined. 6.4 Evaluation VDI Standard 4202, Sheet 1 lists the following minimum requirements for the certification ranges of measuring systems intended for the measurement ambient air pollution through suspended particulate matter: Table 8: Certification ranges Component Minimum value cr Maximum value cr Limit value in µg/m³ in µg/m³ in µg/m³ Assessment period PM 2, Calendar year Certification ranges are related to the limit value with the shortest assessment period and used for the assessment period of the measuring system in the range of the limit value. This assessment of the measuring system in the range of the limit value is performed as part of the determination of the expanded uncertainty of the candidates according to the guide [4]. For this purpose, the following values are used as reference values in accordance with the specifications of the Guide: PM 2.5 : 30 µg/m³ Refer to test item Calculation of expanded uncertainty between systems under test in this report. 6.5 Assessment Assessment of AMS in the range of the relevant limit values is possible. Performance criterion met? yes 6.6 Detailed presentation of test results Refer to test item Calculation of expanded uncertainty between systems under test in this report.

72 Page 72 of Measuring range The upper limit of measurement of the measuring system shall be greater or equal to the upper limit of the certification range. 6.2 Equipment No additional tools are required. 6.3 Method It was examined whether the upper limit of measurement is greater or equal to the upper limit of the certification range. 6.4 Evaluation As a standard, a measuring range of µg/m³ is set on the measuring system. As appropriate default setting of the analogue output for European conditions a measuring range of or µg/m³ is recommended. (Recommended) measuring range: or µg/m³ Upper limit of the certification range: PM 2.5 : 50 µg/m³ 6.5 Assessment A measuring range of µg/m³ is set as standard. Other measuring ranges are possible. The upper limit of the measuring range of the measuring system is greater than the respective upper limit of the certification range. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

73 Page 73 of Negative output signals Negative output signals or measured values may not be suppresses (life zero). 6.2 Equipment No additional tools are required here. 6.3 Method In the field test and during laboratory testing, it was examined whether the AMS has a means to output negative measured values as well. 6.4 Evaluation The AMS can output negative values either via display or via the data outputs. 6.5 Assessment Negative output signals are directly displayed by the AMS and are output correctly via corresponding data outputs. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

74 Page 74 of Failure in the mains voltage In case of malfunction of the measuring system or failure in the mains voltage for a period of up to 72 h, uncontrolled emission of operation and calibration gas shall be avoided. The instrument parameters shall be secured by buffering against loss caused by failure in the mains voltage. When mains voltage returns, the instrument shall automatically reach the operation mode and start the measurement according to the operating instructions. 6.2 Equipment Not required here. 6.3 Method A failure in the mains voltage was simulated and it was tested, whether the AMS remains undamaged and is ready for measurement after the restart of power supply. 6.4 Evaluation The measuring systems do not require operation gas or calibration gas, therefore uncontrolled emission of gases is not possible. When mains voltage returns after a power failure, the AMS automatically re-starts the operation when reaching the start point of the next measurement cycle (during the type approval test after reaching the next full hour). 6.5 Assessment All parameters are secured against loss by buffering. When mains voltage returns the AMS goes back to failure-free operation mode and automatically resumes measuring after reaching the start point of the next measurement cycle (during the type approval test after reaching the next full hour). Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

75 Page 75 of Operating states The measuring system shall allow control of important operating states by telemetrically transmitted status signals. 6.2 Equipment PC for data acquisition. 6.3 Method A PC was connected to the AMS via RS232 to check data transfer and instrument status. The use of corresponding routers or modems easily enables telemonitoring and remote control. 6.4 Evaluation The AMS allows telemetric monitoring and control via RS232-interface. 6.5 Assessment The measuring systems can be monitored and operated from an external PC via modem or router. Performance criterion met? yes 6.6 Detailed presentation of test results Table 9: Operation status F Operating status (Field NO. 6) Bit 0 Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 Bit 7 Meaning 1 - Standby, 0 - Measurement, zero point, reference or film measurement Foil measurement Zero point Reference measurement (reference check) Measurement

76 Page 76 of 269 Table 10: Error status F Error status (Field NO. 7) Bit 0 Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 Bit 7 Description Volume error Vacuum break Volume < 500 litres or 250 litres with 1/2 hour extraction time Replace battery Filter crack

77 Page 77 of Switch-over Switch-over between measurement and functional check and/or calibration shall be possible telemetrically by computer control or manual intervention. 6.2 Equipment Not required here. 6.3 Method The operator can monitor and partially control the AMS directly or via remote control. However, some functions such as performance of a reference foil test in order to check the radiometric measurement can only be carried out at the AMS directly. 6.4 Evaluation The switch-over between measurement and functional check (here: internal zero point and reference point measurement) can be initiated telemetrically or manually. However, some functions such as performance of a reference foil test in order to check the radiometric measurement can only be carried out at the AMS directly. 6.5 Assessment The switch-over between measurement and functional check (here: internal zero point and reference point measurement) can be initiated telemetrically or manually. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

78 Page 78 of Maintenance interval The maintenance interval of the measuring system shall be determined during the field test and specified. The maintenance interval should be three months, if possible, but at least two weeks. 6.2 Equipment Not required here. 6.3 Method The types of maintenance and the maintenance intervals required to ensure proper functioning of the AMS were determined in this performance criterion. In order to determine the maintenance interval, the results of the determination of the drift at zero and at reference point according to chapter Long-term drift have been taken into account. 6.4 Evaluation No ineligible drift effects during the entire field test campaign have been observed for the measuring systems. Thus, the maintenance interval is determined by the regularly appearing maintenance work (see also module 4.1.2). During operating time, maintenance may be limited to contamination checks, plausibility checks and possible status and error messages. 6.5 Assessment The maintenance interval is determined by the necessary maintenance work and is 4 weeks. Performance criterion met? yes 6.6 Detailed presentation of results For necessary maintenance work refer to item (module) in this report or chapter 5.3 respectively 10 in the operator s manual.

79 Page 79 of Availability The availability of the measuring system shall be determined during the field test and shall be at least 95 %. 6.2 Equipment Not required here. 6.3 Method The start and end point of the availability checks are determined by the start and end point at each of the field test sites. For this purpose, all interruptions, for instance those caused by malfunctioning or maintenance work, are recorded as well. 6.4 Evaluation Table 11 and Table 12 provide lists of operation times, time used for maintenance, and malfunction times. The measuring systems were operated over a period of 296 days in total during the field test. This period includes 22 days of zero filter operation, audits and days that were lost due to changing from inlet to zero filter (see also annex 5). Because of investigations to determine the unusual deviations between both candidates on October 15, 2013, the measured value for instrument SN was also discarded (though not affected) because of long downtime due to maintenance. Downtimes caused by external influences which the instrument cannot be blamed for have been recorded between September 05, 2013 and September 12, The measured values for both candidates had to be discarded for the entire period, as the flow rate has been systematically misadjusted before start-up due to a wrong parameter setting in the reference flow meter. Therefore the flow rates have been checked again and correctly adjusted on Sept 12, Because of this, the total operation time is reduced to 288 (SN ) respectively 288 (SN ) days of measurements. The following downtimes due to malfunctions have been recorded: SN : For this system no malfunctions have been observed. SN : For SN it was observed at test site Cologne, autumn, that the measured values have been sometimes significantly higher than the ones from SN and also higher than the first batch of reference values from this site. Thus the instrument was investigated intensely at the measurement site by the manufacturer on October 15, 2013 and it was observed, that the protective sheet of the Geiger-Müller counter tube appeared very ripple and as a consequence deviations can occur. The cause for this damage could not be found. It was decided then, to install a new Geiger-Müller counter tube into the device. No measured values from the past have been discarded, but there was no measurement on October 15, Apart from that no further downtimes were recorded.

80 Page 80 of 269 Downtimes caused by routine maintenance (without zero filter operation), e.g. for cleaning the sampling inlet or for the check of the flow rate / tightness usually lead to a loss of one measurement cycle (i.e. 1 h per day). Daily mean values affected by this have thus not been discarded. 6.5 Assessment The availability was 100 % for SN and 99.7 % for SN without test-related downtimes. Including test-related downtimes it was 92.0 % for SN and 92.0 % for SN Performance criterion met? yes 6.6 Detailed presentation of test results Table 11: Determination of availability (without test-related downtimes) System 1 (SN ) System 2 (SN ) Operating time d Downtime d 0 1 Maintenance d 0 0 Actual operating time d Availability % Table 12: Determination of availability (incl. test-related downtimes) System 1 (SN ) System 2 (SN ) Operating time d Downtime d 0 1 Maintenance incl. zero filter d Actual operating time d Availability %

81 Page 81 of Instrument software The version of the instrument software to be tested shall be displayed during switch-on of the measuring system. The test institute shall be informed on changes in the instrument software, which have influence on the performance of the measuring system. 6.2 Equipment Not required here. 6.3 Method It was checked whether the measuring system has a means of displaying the instrument software. The manufacturer was advised to inform the test institute on any changes in the instrument software. 6.4 Evaluation The current software version can be viewed at any time in the menu Parameter/Service under the item SW. The test was started in December 2012 with the software version Within the frame work of the software maintenance process, the software was in the meanwhile developed and optimised up to version The manufacturing site of the instrument manufacturer DURAG GmbH in Hamburg, Germany is part of the regular surveillance according to Standard EN since Since 2013 also the measuring system F is covered in the audit scope. According to the requirements of the Standard EN , the applied modifications of the software have been described, assessed, the type of modification classified and the information submitted to the test house. An overview on the applied changes can be found in chapter 4.1 General. The applied modifications are for functional extensions, bug elimination and for cosmetic maintenance. They don t have any impact on the measuring system. With the start of the third of in total four comparison campaigns, the software version was upgraded from 3.07 to 3.10 and the instruments were operated with this software for the comparison campaigns Cologne, autumn and Cologne, winter. 6.5 Assessment The version of the instrument software is shown on the display. The test institute is informed on any changes in the instrument software. Performance criterion met? yes

82 Page 82 of Detailed presentation of test results Figure 22: Display of software version here 3.10 in the menu Parameter/Service/SW

83 Page 83 of General The testing is performed on the basis of the minimum requirements stated in VDI Standard 4202, Sheet 1 (September 2010). 6.2 Equipment Not required here. 6.3 Method The testing is performed on the basis of the minimum requirements stated in VDI Standard 4202, Sheet 1 (September 2010). 6.4 Evaluation After extensive revision, the VDI Standards 4202, Sheet 1 and 4203, Sheet 3 has been newly published in September Unfortunately, after this revision there are several ambiguities and inconsistencies in relation to concrete minimum requirements and the general significance of particular test items as far as the testing of particulate measuring systems is concerned. The following test items are in need of clarification: Repeatability standard deviation at zero point no performance criterion defined Repeatability standard deviation at reference point not applicable to particulate measuring devices Linearity (lack of fit) not applicable to particulate measuring devices Sensitivity coefficient of surrounding temperature no performance criterion defined Sensitivity coefficient of supply voltage no performance criterion defined Standard deviation from paired measurements no performance criterion defined Long-term drift no performance criterion defined Short-term drift not applicable to particulate measuring devices Overall uncertainty not applicable to particulate measuring devices For this reason, an official enquiry was made to the competent body in Germany, to define a coordinated procedure for dealing with the inconsistencies in the guideline.

84 Page 84 of 269 The following procedure was suggested: The test items 5.3.2, 5.3.7, 5.3.8, , and are evaluated as before on the basis of the minimum requirements stated in the 2002 version of VDI Standard 4202, Sheet 1 (i.e. applying the reference values B 0, B 1, and B 2 ). The test items 5.3.3, 5.3.4, , and are omitted as they are irrelevant to particulate measuring devices. The competent body in Germany agreed with the suggested procedure by decisions of 27 June 2011 and 07 October Assessment The test was carried out on the basis of the performance criteria stated in VDI Standard 4202, Sheet 1 (September 2010). However, the test items 5.3.2, 5.3.7, 5.3.8, , and were evaluated on the basis of the performance criteria stated in the 2002 version of VDI Standard 4202, Sheet 1 (i.e. applying the reference values B 0, B 1, and B 2 ). The test items 5.3.3, 5.3.4, , and were omitted as they are irrelevant to particulate measuring devices. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

85 Page 85 of Repeatability standard deviation at zero point The repeatability standard deviation at zero point shall not exceed the requirements of Table 2 in VDI Standard 4202, Sheet 1 (September 2010) in the certification range according to Table 1 in VDI Standard 4202, Sheet 1 (September 2010). In case of deviating certification ranges, the repeatability standard deviation at zero point shall not exceed 2 % of the upper limit of this certification range. Note: With regard to dust measuring devices, this test item cannot be evaluated on the basis of the current version of VDI Standards 4202, Sheet 1 (September 2010) and 4203, Sheet 3 (September 2010). By resolution of the competent body in Germany (see module 5.3.1), reference is made to the following minimum requirement in the previous version of this guideline (VDI Standard 4202, Sheet 1; June 2002): The detection limit of the measuring system shall not exceed the reference value B 0. The detection limit shall be determined during the field test. 6.2 Equipment Zero filter for testing the zero point. 6.3 Method The detection limits of the candidates, SN and SN , were determined by means of zero filters which were installed at the inlets of instruments. Over a period of 15 days and 24 h/day, particulate-free sample air was fed into the systems. The detection limit was determined in the laboratory test. 6.4 Evaluation The detection limit X is calculated from the standard deviation s x0 from the measured values when particulate-free sample air is sucked in by the two candidates. It corresponds to the standard deviation from the mean value s x0 of the measured values x 0i for each candidate multiplied by the Student s factor: 1 X = t n-1;0.95 s x 0 with s x 0 = n 1 Reference value: B 0 = 2 µg/m³ i= 1, n (x 0 i x 0 ) 2

86 Page 86 of Assessment The tests resulted in detection limits of 0.66 µg/m³ for System 1 (SN ) and 0.75 µg/m³ for System 2 (SN ). Performance criterion met? yes 6.6 Detailed presentation of test results Table 13: Detection limit PM 2.5 Device SN Device SN Number of values n Average of the zero values µg/m³ x 0 s x0 s x0 Standard deviation of the values µg/m³ Student-Factor t n-1;0, Detection limit x µg/m³ The single measured values used in the determination of the detection limit are given in Annex 1 of this report.

87 Page 87 of Repeatability standard deviation at reference point The repeatability standard deviation at reference point shall not exceed the requirements of Table 2 in VDI Standard 4202, Sheet 1 (September 2010) in the certification range according to Table 1 in VDI Standard 4202, Sheet 1 (September 2010). The limit value or the alert threshold shall be used as reference point. In case of deviating certification ranges, the repeatability standard deviation at reference point shall not exceed 2 % of the upper limit of this certification range. In this case a value c t at 70 % to 80 % of the upper limit of this certification range shall be used as reference point. Note: By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

88 Page 88 of Linearity (lack of fit) The analytical function describing the relationship between the output signal and the value of the air quality characteristic shall be linear. Reliable linearity is given, if deviations of the group averages of measured values about the calibration function meet the requirements of Table 2 in VDI Standard 4202, Sheet 1 (September 2010) in the certification range according to Table 1 in VDI Standard 4202, Sheet 1 (September 2010). For all other certification ranges the group averages of measured values about the calibration function shall not exceed 5 % of the upper limit of the corresponding certification range. Note: By resolution of the competent body in Germany (refer to module 5.3.1), this test item is irrelevant to particulate measuring systems. Particulate measuring systems for PM 2.5 shall be tested according to performance criterion Calculation of expanded uncertainty between systems under test. 6.2 Equipment Refer to module (PM 2.5 ) 6.3 Method Particulate measuring systems for PM 2.5 shall be tested according to performance criterion Calculation of expanded uncertainty between systems under test. 6.4 Evaluation Refer to module (PM 2.5 ) 6.5 Assessment Particulate measuring systems for PM 2.5 shall be tested according to performance criterion Calculation of expanded uncertainty between systems under test. Performance criterion met? Detailed presentation of test results Refer to module (PM 2.5 )

89 Page 89 of Sensitivity coefficient of sample gas pressure The sensitivity coefficient of sample gas pressure at reference point shall not exceed the requirements of Table 2 in VDI Standard 4202, Sheet 1 (September 2010). A value c t at 70 % to 80 % of the upper limit of the certification range shall be used as reference point. Note: This test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

90 Page 90 of Sensitivity coefficient of sample gas temperature The sensitivity coefficient of sample gas temperature at reference point shall not exceed the requirements of Table 2 in VDI Standard 4202, Sheet 1 (September 2010). A value c t at 70 % to 80 % of the upper limit of the certification range shall be used as reference point. Note: This test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

91 Page 91 of Sensitivity coefficient of surrounding temperature The sensitivity coefficient of surrounding temperature at zero and reference point shall not exceed the requirements of Table 2 in VDI Standard 4202, Sheet 1 (September 2010). A value c t at 70 % to 80 % of the upper limit of the certification range shall be used at reference point. Note: In relation to particulate measuring systems, this test item cannot be evaluated according to the current versions of VDI Standards 4202, Sheet 1 (September 2010) and 4203, Sheet 3 (September 2010), because the minimum requirements are not defined. By resolution of the competent body in Germany (see module 5.3.1), reference is made to the following requirements stated in the earlier version of VDI Standard 4202, Sheet 1 (June 2002): If the surrounding temperature changes by 15 K in the range +5 C to +20 C or by 20 K in the range +20 C to +40 C, the temperature dependence of the measured value at zero point shall not exceed the reference value B 0. The temperature dependence of the measured value in the range of the reference value B 1 shall not be greater than ± 5 % of the measured value when a change in temperature by 15 K in the range of +5 C to +20 C or +20 C to +40 C occurs. 6.2 Equipment Climatic chamber for a temperature range of +5 to +40 C, zero filter for testing the zero point, reference foil for testing the reference point. 6.3 Method In order to test the dependence of zero point and measured values on the surrounding temperature, the complete measuring systems were operated within a climatic chamber. For the zero point test particle free sampling air was applied to both measuring systems SN and SN by means of zero filters installed at the instrument inlets. The reference point test comprised a check of the stability of the sensitivity of the reference foil value for both candidates SN and SN The ambient temperature within the climatic chamber was altered in the sequence 20 C 5 C 20 C 40 C 20 C. The measured values at zero point (3 times 24 h per temperature level) and the measured values at reference point (3 times per temperature level) were recorded after an equilibration period of 24 h per temperature level.

92 Page 92 of Evaluation Zero point: The measured concentration values obtained in the individual 24-hour measurements were collected and evaluated. The absolute deviation in µg/m³ per temperature level in relation to the default temperature of 20 C is considered. Reference value: B 0 = 2 µg/m³ Reference point: The measured value s change in percentage for each temperature level in relation to the initial temperature of 20 C is checked. It should be noted that concentration values could not be simulated by checking the measured value of the reference foil. It was therefore not possible to examine the range of B Assessment The maximum dependence of ambient temperature in the range of +5 C to +40 C at zero was -1.0 µg/m³. At reference point, no deviations > 3.0 % in relation to the default value at the temperature of 20 C were observed. Performance criterion met? yes

93 Page 93 of Detailed presentation of test results Table 14: Dependence of zero point on ambient temperature, deviations in µg/m³, mean value of three measurements Ambient temperature Deviation Start temperature End temperature SN SN C C µg/m³ µg/m³ Table 15: Dependence of sensitivity (foil value) on ambient temperature, deviation in %, mean value of three measurements Ambient temperature Deviation Start temperature End temperature SN SN C C [%] [%] For the respective results of the 3 individual measurements refer to annex 2 and annex 3.

94 Page 94 of Sensitivity coefficient of supply voltage The sensitivity coefficient of supply voltage shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). A value c t at 70 % to 80 % of the upper limit of the certification range shall be used as reference point. Note: In relation to particulate measuring systems, this test item cannot be evaluated according to the current versions of VDI Standards 4202, Sheet 1 (September 2010) and 4203, Sheet 3 (September 2010), because the minimum requirements are not defined. By resolution of the competent body in Germany (see module 5.3.1), reference is made to the following requirements stated in the earlier version of VDI Standard 4202, Sheet 1 (June 2002): Change in the measured value at reference value B 1 caused by the common changes in the mains voltage in the interval ( /-20) V shall not exceed B Equipment Isolation transformer, reference foil for testing the reference point. 6.3 Method In order to examine the dependence of measured signal on supply voltage, the latter was reduced from 230 V to 210 V and then increased over an intermediate stage of 230 V to 245 V. The reference point test comprised a check of the stability of the sensitivity of the reference foil value for both candidates SN and SN As the AMS is not designed for mobile use, separate testing of the dependence of measurement signal on mains frequency was abstained from. 6.4 Evaluation At reference point, the changes in percentage of the determined measured values were examined for each voltage step in relation to the default voltage of 230 V. It should be noted that concentration values could not be simulated by checking the measured value of the reference foil. It was therefore not possible to examine the range of B 1.

95 Page 95 of Assessment No deviations > -1.4 % for PM 2.5 in relation to the default value at the mains voltage 230 V due to changes in supply voltage were detected. Performance criterion met? yes 6.6 Detailed presentation of test results Table 16 presents a summary of test results. Table 16: Dependence of measured value on supply voltage, deviation in % Mains voltage Deviation Start voltage End voltage SN SN V V [%] [%] For the individual results refer to annex 4 in this report.

96 Page 96 of Cross-sensitivity The change in the measured value caused by interfering components in the sample gas shall not exceed the requirements of Table 2 (VDI Standard 4202, Sheet 1; September 2010) at zero and reference point. Note: This test item is irrelevant to particulate measuring systems. As minimum requirement applies in this case, the test results are stated in module Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

97 Page 97 of Averaging effect For gaseous components the measuring system shall allow the formation of hourly averages. The averaging effect shall not exceed the requirements of Table 2 (VDI Standard 4202 Sheet 1; September 2010). Note: This test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

98 Page 98 of Standard deviation from paired measurements The standard deviation from paired measurements under field conditions shall be determined with two identical measuring systems by paired measurements in the field test. It shall not exceed the requirements of Table 2 (VDI Standard 4202, Sheet 1; September 2010). Note: In relation to particulate measuring systems, this test item cannot be evaluated according to the current versions of VDI Standards 4202, Sheet 1 (September 2010) and 4203, Sheet 3 (September 2010), because the minimum requirements are not defined. By resolution of the competent body in Germany (see module 5.3.1), reference is made to the following requirements stated in the earlier version of VDI Standard 4202, Sheet 1 (June 2002): The Reproduzierbarkeit [reproducibility] R D of the measuring system shall be determined by parallel measurements with two identical measuring systems and shall be at least equal to 10. B 1 shall be used as reference value. 6.2 Equipment For the determination of reproducibility, the additional measuring systems described in chapter 5 were used. 6.3 Method Reproducibility is defined as the maximum difference between two randomly chosen single values that have been obtained under equal conditions. Reproducibility was determined using two identical measuring systems that were operated simultaneously during the field test. For this purpose, all measurement data obtained during the entire field test was evaluated. 6.4 Evaluation The reproducibility is calculated as follows: n 2 D (x1i x 2i ) i= 1 R = B 1 10 U with U = ± s D t (n;0,95 ) and 1 s = 2n R = Reproducibility at B 1 U = Uncertainty B 1 = 25 µg/m³ for PM 2.5 s D = Standard deviation from paired measurements n = No. of paired measurements t (n;0.95) = Student s factor at confidence level of 95 % x 1i = Measured signal of system 1 (e.g. SN ) at i th concentration x 2i = Measured signal of system 2 (e.g. SN ) at i th concentration

99 Page 99 of Assessment In the field test, the reproducibility for the full dataset was 21 for PM 2.5. Performance criterion met? yes 6.6 Detailed presentation of test results The test results are summarised in Table 17. The graphical representation for PM 2.5 is given in Figure 29 to Figure 33. Note: The determined uncertainties are related to reference value B 1 for each site: Table 17: Concentration mean values, standard deviation, uncertainty range, and reproducibility in the field, measured component PM 2.5 Site Number c (SN ) c (SN ) c ges s D t U R µg/m³ µg/m³ µg/m³ µg/m³ µg/m³ Bonn, winter Bornheim, summer Cologne, autumn Cologne, winter All sites c (SN ): Mean value of concentrations for System SN c (SN ): Mean value of concentrations for System SN c ges : Mean value of concentrations for Systems SN & SN For individual values refer to annex 5 of the appendix.

100 Page 100 of Long-term drift The long-term drift at zero point and reference point shall not exceed the requirements of Table 2 (VDI Standard 4202, Sheet 1; September 2010) in the field test. A value c t at 70 % to 80 % of the upper limit of the certification range shall be used as reference point. Note: In relation to particulate measuring systems, this test item cannot be evaluated according to the current versions of VDI Standards 4202, Sheet 1 (September 2010) and 4203, Sheet 3 (September 2010), because the minimum requirements are not defined. By resolution of the competent body in Germany (see module 5.3.1), reference is made to the following requirements stated in the earlier version of VDI Standard 4202, Sheet 1 (June 2002): The temporal change in the measured value at zero concentration shall not exceed the reference value B 0 in 24 h and in the maintenance interval. The temporal change in the measured value in the range of the reference value B 1 shall not be greater than ± 5 % of B 1 in 24 h and in the maintenance interval. 6.2 Equipment Zero filter for testing the zero point, reference foil for testing the reference point. 6.3 Method The test was carried out as part of the field test over a period of about 12 months altogether. In the context of the regular monthly checks carried (including those at the beginning and end of tests at each field test site), both measuring systems were operated with zero filters applied to their inlets for at least 24 h. The measured zero values were then evaluated. Furthermore, the stability of the measured value for the reference foil was checked and evaluated regularly during the entire field test campaign in order to check the reference point.

101 Page 101 of Evaluation An evaluation of the drift of zero point and measured value in 24 h is possible for this system via the internal check procedure for zero- and reference point, but is uncommon for particulate measuring systems in practice. The evaluation at zero point is made on the basis of the measurement results of the regular external zero point measurement by comparing the respective values with the corresponding measured values of the previous test and the measured value of the first test. The evaluation at reference point is made on the basis of the measurement results for the reference foil by comparing the respective values with the corresponding measured values of the previous test and the measured value of the first test. It should be noted that concentration values could not be simulated by checking the measured value of the reference foil. It was therefore not possible to examine the range of B Assessment The maximum deviation at zero point was 1.5 µg/m³ in relation to the previous value and 1.7 µg/m³ in relation to the start value. The sensitivity drift values that were determined during testing are max % for PM 2.5 in relation to the respective start value. Performance criterion met? yes

102 Page 102 of Detailed presentation of test results Table 18 provides the obtained measured values for zero point as well as the calculated deviations in relation to the previous and the starting value in µg/m³. Figure 23 to Figure 24 provide a graphic representation of zero point drift over the course of testing. The deviations of the measured values from the corresponding previous value in % are listed in Table 19. Figure 25 and Figure 26 present graphical representations of the drift of measured values (in relation to the previous values). Table 18: Zero point drift SN & SN , PM 2.5, with zero filter SN SN Measured Deviation from Deviation from Measured Deviation from Deviation from Date Value previous value start value Date Value previous value start value µg/m³ µg/m³ µg/m³ µg/m³ µg/m³ µg/m³ 2/28/ /28/ /30/ /30/ /31/ /31/ /1/ /1/ /26/ /26/ /27/ /27/ /28/ /28/ /14/ /14/ /15/ /15/ /22/ /22/ /23/ /23/ /26/ /26/ /4/ /4/ /16/ /16/ /8/ /8/ /9/ /9/ /10/ /10/ /14/ /14/ /15/ /15/ /13/ /13/ /7/ /7/ /8/ /8/ /9/ /9/ /10/ /10/

103 Page 103 of 269 Measured value [µg/m³] Zero point drift PM2,5 SN Time Figure 23: Zero point drift SN , measured component PM 2.5 Measured value [µg/m³] Zero point drift PM2,5 SN Time Figure 24: Zero point drift SN , measured component PM 2.5

104 Page 104 of 269 Table 19: Sensitivity drift SN & SN , PM 2.5 Date Measured Value SN SN Deviation from Deviation from Measured Deviation from Deviation from previous value start value Date Value previous value start value % % % % 2/27/ /27/ /2/ /2/ /3/ /3/ /12/ /12/ /11/ /11/ Dev. from previous value [%] Span point drift PM2,5 SN Time Figure 25: Drift of the measured value SN , measured component PM 2.5

105 Page 105 of 269 Dev. from previous value [%] Span point drift PM2,5 SN Time Figure 26: Drift of the measured value SN , measured component PM 2.5

106 Page 106 of Short-term drift The short-term drift at zero point and reference point shall not exceed the requirements of Table 2 (VDI Standard 4202, Sheet 1; September 2010) within 12 h (for benzene 24 h) in the laboratory test and within 24 h in the field test. A value c t at 70 % to 80 % of the upper limit of the certification range shall be used as reference point. Note: By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

107 Page 107 of Response time The response time (rise) of the measuring system shall not exceed 180 s. The response time (fall) of the measuring system shall not exceed 180 s. The difference between the response time (rise) and the response time (fall) of the measuring system shall not exceed 10 % of response time (rise) or 10 s, whatever value is larger. Note: This test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

108 Page 108 of Difference between sample and calibration port The difference between the measured values obtained by feeding gas at the sample and calibration port shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). A value c t at 70 % to 80 % of the upper limit of the certification range shall be used as reference point. Note: This test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

109 Page 109 of Converter efficiency In case of measuring systems with a converter, the converter efficiency shall be at least 98 %. Note: This test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

110 Page 110 of Increase of NO 2 concentration due to residence in the AMS In case of NO x measuring systems the increase of NO 2 due to residence in the measuring system shall not exceed the requirements of Table 2 of VDI Standard 4202, Sheet 1 (September 2010). The requirements of Table 2 of VDI Standard 4202, Sheet 1 apply to certification ranges according to Table 1 of VDI Standard 4202, Sheet 1 (September 2010). For deviating certification ranges the requirements shall be proportionally converted. Note: This test item is irrelevant to particulate measuring systems. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

111 Page 111 of Overall uncertainty The expanded uncertainty of the measuring system shall be determined. The value determined shall not exceed the corresponding data quality objectives in the applicable EU Directives on air quality listed in Annex A, Table A 1 of VDI Standard 4202, Sheet 1 (September 2010). Note: By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. Please refer to module Equipment By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. Please refer to module Method By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. Please refer to module Evaluation By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. Please refer to module Assessment By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. Please refer to module Performance criterion met? Detailed presentation of test results By resolution of the competent body in Germany (see module 5.3.1), this test item is irrelevant to particulate measuring systems. Please refer to module

112 Page 112 of General The testing of particulate measuring systems shall be carried out according to the minimum requirements set out in Table 5 of VDI Standard 4202, Sheet 1 (September 2010). Particle mass concentrations shall be related to a defined volume. The relation to volume with respect to pressure and temperature shall be comprehensively described. 6.2 Equipment No equipment is necessary to test this performance criterion. 6.3 Method The test was carried out according to the minimum requirements set out in Table 5 of VDI Standard 4202, Sheet 1 (September 2010). To determine whether the measured particle mass concentrations are related to a defined volume was the objective of the test. 6.4 Evaluation The test was carried out according to the minimum requirements set out in Table 5 of VDI Standard 4202, Sheet 1 (September 2010). The measuring system F is a radiometric measuring system. The mass, which is separated at the filter tape, is determined by a radiometric measurement. The determined mass is related to a defined and actively controlled sample volume and thus the particulate mass concentration is determined. 6.5 Assessment The test was carried out according to the minimum requirements set out in Table 5 of VDI Standard 4202, Sheet 1 (September 2010). The determined mass is related to a defined and actively controlled sample volume and thus the particulate mass concentration is determined. Performance criterion met? yes 6.6 Detailed presentation of test results No equipment is necessary to test this performance criterion.

113 Page 113 of Equivalency of the sampling system The equivalency between the PM 10 sampling system and the reference method according to Standard EN [T5] shall be demonstrated. Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.2 Equipment Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.3 Method Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.4 Evaluation Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.5 Assessment Not applicable to PM 2.5 sampling systems. Please refer to module in this report. Performance criterion met? Detailed presentation of test results Not applicable to PM 2.5 sampling systems. Please refer to module in this report.

114 Page 114 of Reproducibility of the sampling systems The PM 10 sampling systems of two identical systems under test shall be reproducible among themselves according to Standard EN [T5]. This shall be demonstrated in the field test. Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.2 Equipment Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.3 Method Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.4 Evaluation Not applicable to PM 2.5 sampling systems. Please refer to module in this report. 6.5 Assessment Not applicable to PM 2.5 sampling systems. Please refer to module in this report. Performance criterion met? Detailed presentation of test results Not applicable to PM 2.5 sampling systems. Please refer to module in this report.

115 Page 115 of Calibration The systems under test shall be calibrated in the field test by comparison measurements with the reference method according to Standard EN respectively EN Here, the relationship between the output signal and the gravimetrically determined reference concentration shall be determined as a steady function. 6.2 Equipment Refer to module Method For PM 2.5 : The reproducibility of the measuring systems as per module was proven during testing. In order to determine the calibration function and the analytical function, the complete dataset was used (213 valid data pairs (SN ) and 213 valid data pairs (SN )). The quantities of the calibration function y = m * x +b were determined by means of orthogonal regression. The analytical function is the inverse of the calibration function. It is: x = 1/m * y b/m The slope m of the regression line describes the sensitivity of the measuring system, the y- intercept b describes the zero point. 6.4 Evaluation The resulting quantities are given in Table 20. Table 20: Results of the calibration function and analytical function, measured component PM 2.5 Device no. Calibration function Analytical function System 1 (SN ) System 2 (SN ) Y = m * x + b x = 1/m * y - b/m m b 1/m b/m µg/m³ / µg/m³ µg/m³ µg/m³ / µg/m³ µg/m³

116 Page 116 of Assessment A statistical correlation between the reference measuring method and the output signal could be demonstrated. Performance criterion met? yes 6.6 Detailed presentation of test results Refer to module

117 Page 117 of Cross sensitivity The interference caused by moisture in the sample may not exceed 10 % of the limit value in the range of the limit value. 6.2 Equipment Not required here. 6.3 Method The interference caused by moisture in the sample was determined under field conditions. Using the data from field test days with a relative humidity of > 70 % the difference between the obtained reference value (= nominal value) and the measured values of each candidate was calculated and the mean difference was applied as a conservative estimate for the interference caused by moisture in the sample. In addition to that, reference/equivalence functions were determined for both devices using the data from field test days with a relative humidity of > 70 %. 6.4 Evaluation Using the data from field test days with a relative humidity of > 70 %, the mean difference between the calculated reference value (= nominal value) and the measured value of the respective candidate was calculated and the relative deviation from the mean concentration was determined. Annual limit value PM 2.5 = 25 µg/m³ 10 % of the annual limit value = 2.5 µg/m³ It was also examined whether the reproducibility of the measuring systems under test using the reference method according to Guide Demonstration of Equivalence of Ambient Air Monitoring Methods [4] can be ensured even if the measured values were obtained on days with a relative humidity of > 70 %.

118 Page 118 of Assessment No deviation of the measured signal from the nominal value > -0.6 µg/m³ caused by interference due to moisture in the sample could be observed for PM 2.5. No negative influence on the measured values at varying relative humidity was detected during the field test. The comparability of the candidates with the reference method according to the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods [4] is ensured even for days with a relative humidity of > 70 %. Performance criterion met? yes 6.6 Detailed presentation of test results Table 21 provides a summary of the results. Table 21: Deviation between reference measurement and candidate on days with a relative humidity of > 70 %, measured component PM 2.5 Field test, days with rh >70 % Reference SN SN Mean µg/m³ Deviation to mean value of reference in µg/m³ µg/m³ Deviation in % of mean value of % reference Deviation in % of ALV % Single values are provided in annexes 5 and 6. The measurement uncertainties W CM on days with a relative humidity of > 70 % are presented in Table 22. Single values are provided in annexes 5 and 6.

119 Page 119 of 269 Table 22: Comparison of the candidates / with the reference device, rel. humidity > 70 %, all test sites, measured component PM 2.5 Comparison candidate with reference according to Guide "Demonstration of Equivalence Of Ambient Air Monitoring Methods", January 2010 Candidate F SN SN / SN Limit value 30 µg/m³ Status of measured values Raw data Allowed uncertainty 25 % All test sites, rh>70% Uncertainty between Reference 0.57 µg/m³ Uncertainty between Candidates 0.61 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty W CM % %

120 Page 120 of Averaging effect The measuring system shall allow the formation of 24 h mean values. The time of the sum of all filter changes within 24 h shall not exceed 1 % of this averaging time. 6.2 Equipment Additionally a timer was used. 6.3 Method It was tested, whether the AMS allows the formation of daily mean values. 6.4 Evaluation The measuring system operates with measurement cycles between 15 min and 24 h. The sampling time corresponds to the respective programmed cycle time and the programmed sample count minus the measuring time respectively the time for filter tape movements. With the help of the sample count, a multiple sampling on a filter spot can be set. It can be set between 1 (=for each cycle a new filter spot) and 24 (=one filter spot is sampled upon for 24times). The sampling time is thus: For cycle time 60 min and sample count 1: 60 min (2 x 300 s measuring time s for filter tape movements) = 48 min In case of a sample count >1, the measurement after sampling serves for the calculation of the measured value of the finished cycle as well as a start measurement for the following cycle, i.e. only one radiometric measurement is necessary per cycle. During the type approval test a cycle time of 60 min with a sample count of 24 was set. The sampling time is then: Cycle 1: 60 min (2 x 300 s measuring time s for filter tape movements) = 48 min Cycle 2-24: 60 min (1 x 300 s measuring time s for filter tape movements) = 53 min Thus the available sampling time per measurement cycle is between 80 % and 88.3 % of the total cycle time. The results from the field investigations according to chapter Calculation of expanded uncertainty between systems under test in this report show, that for this instrument configuration the comparability of the candidates with the reference method could be demonstrated and thus, the formation of daily mean values can be guaranteed.

121 Page 121 of Assessment With the described instrument configuration and a measurement cycle of 1 h with a sample count of 24, the measuring system allows the formation of daily mean values based on 24 measurement cycles. Performance criterion met? yes 6.6 Detailed presentation of test results Not required here.

122 Page 122 of Constancy of sample volumetric flow The sample volumetric flow averaged over the sampling time shall be constant within ± 3 % of the rated value. All instantaneous values of the sample volumetric flow shall be within a range of ± 5 % of the rated value during sampling. 6.2 Equipment As indicated in chapter 4, a flow meter was used in the testing of this performance criterion. 6.3 Method The sample volumetric flow was calibrated before testing at the first field test site. Before testing at the other field test sites it was checked for correctness with a mass flow meter and readjusted if necessary. In order to determine the constancy of sample volumetric flow, the flow rate was recorded over 24 h by means of a mass flow meter and evaluated according to the relevant upcoming test item Constancy of sample flow rate of Technical Specification EN/TS (August 2013) [8]. 6.4 Evaluation The obtained measured values for the flow rate were used to calculate mean value, standard deviation as well as maximum and minimum value.

123 Page 123 of Assessment The results of the flow rate checks carried out at each field test site are given in Table 23. Table 23: Flow rate check before testing at Test site: Results of flow rate checks [l/min] SN SN Deviation from nominal value [%] [l/min] Deviation from nominal value [%] Bonn, winter 16.67* * - Bornheim, summer Cologne, autumn Cologne, winter * adjusted on February 27, 2013 The graphical representations of flow rate constancy show that none of the values obtained during sampling deviates from the respective nominal value by more than ±5 %. The 24 h mean values for the total flow rate of l/min also deviate significantly less than the permissible ±3 % from the nominal value. All determined daily mean values deviate less than ± 3 % from the rated value and all instantaneous values deviate less than ± 5 %. Performance criterion met? yes 6.6 Detailed presentation of test results Table 24 shows the parameters determined for the flow. Figure 27 and Figure 28 present a graphic representation of the flow measurements of the two candidates SN and SN Table 24: Parameters for total flow measurement (24 h mean), SN & SN Device Mean [l/min] Deviation from nominal [%] Std. Dev. [l/min] Max [l/min] Min [l/min] SN SN

124 Page 124 of Constancy of sample flow rate SN Flowrate in l/min Ambient temperatur e in C Nominal value l/min Max. permissible dev. Flow rate Ambient temperature [ C] /12/2014 9:36 3/12/ :24 3/12/ :12 3/13/2014 0:00 3/13/2014 4:48 3/13/2014 9:36 3/13/ :24 Figure 27: Flow rate of device SN Constancy of sample flow rate SN Flow rate in l/min Ambient temperature in C Nominal value l/min Max. permissible dev. Flow rate Ambient temperature [ C] /13/2014 9:36 3/13/ :24 3/13/ :12 3/14/2014 0:00 3/14/2014 4:48 3/14/2014 9:36 3/14/ :24 Figure 28: Flow rate of device SN

125 Page 125 of Tightness of the measuring system The complete measuring system shall be checked for tightness. Leakage shall not exceed 1 % of the sample volume sucked. 6.2 Equipment Plug for blocking the sampling tube 6.3 Method The flow meter of the measuring system F is located directly upstream the pump. To determine the leak rate of the AMS, the instrument is started up according to chapter of the instrument s manual and after reaching the nominal flow rate of 1000 l/h, the inlet of the sampling tube is sealed by e.g. a thumb or a plug. According to the manufacturer, the flow rate, measured by the device, has to drop lower than 10 l/h, ideally 0 l/h. This procedure was carried out every time the AMS was installed at a new field test site. It is recommended to check the tightness of the measuring system by means of the aforementioned procedure every three months right before the regular flow rate check. 6.4 Evaluation Leakage testing was performed right after the AMS was installed at a new field test site. The criterion for passing the leakage test, which has been proposed by the manufacturer (maximum flow at blocked inlet 10 l/h), proved to be an appropriate parameter for monitoring instrument tightness. The detected maximum leak rate of 1 l/h is less than 1 % of the nominal flow rate of 1000 l/h (16.67 l/min).

126 Page 126 of Assessment The criterion for passing the leakage test, which has been specified by the manufacturer, (flow at blocked inlet max. 10 l/h) proved to be an appropriate parameter for monitoring instrument tightness. The detected maximum leak rate of 1 l/h is less than 1 % of the nominal flow rate of 1000 l/h (16.67 l/min). Performance criterion met? yes 6.6 Detailed presentation of test results Table 25 lists the values obtained in leakage testing. Table 25: Results from leakage testing during the field tests Leak test before SN SN Test site: Nominal [l/h] Actual [l/h] Nominal [l/h] Actual [l/h] Bonn, winter < 10 0 < 10 0 Bornheim, summer < 10 1 < 10 0 Cologne, autumn < 10 0 < 10 0 Cologne, winter < 10 0 < 10 0

127 Page 127 of Methodology of the equivalence check (modules ) According to the January 2010 version of the Guide [4], the following 5 criteria shall be met in order to prove equivalence: 1. At least 20 % of the concentration values from the complete dataset (determined by means of reference method) shall exceed the upper assessment threshold for annual limit values determined in 2008/50/EC [7], i.e. 28 µg/m³ for PM 10 and 17 µg/m³ for PM The uncertainty between the candidates must be less than 2.5 µg/m³ for all data and for two sub datasets corresponding to all the data split greater than or equal to and lower than 30 µg/m³ or 18 µg/m³ for PM 10 and PM 2.5 respectively. 3. The uncertainty between the reference devices must be less than 2.0 µg/m³. 4. The expanded uncertainty (W CM ) is calculated at 50 µg/m³ for PM 10 and 30 µg/m³ for PM 2.5 for each candidate against the mean value of the reference method. In each of the following cases, the expanded uncertainty shall not exceed 25 %: Full dataset; Dataset with PM concentrations greater/equal 30 µg/m³ for PM 10 or greater/equal 18 µg/m³ for PM 2.5, provided that the dataset contains 40 or more valid data pairs; Datasets for each field test site. 5. For the complete dataset to be accepted it is required that the slope b differs insignificantly from 1: b 1 2 u(b) and that the intercept a differs insignificantly from 0: a 2 u(a). Should these requirements not be met, the candidates may be calibrated using the values for slope and/or intercept from the complete dataset. In the following 5 chapters, compliance with the 5 criteria is tested: In chapter Determination of uncertainty between systems under test u bs criteria 1 and 2 will be checked. In chapter Calculation of expanded uncertainty between systems under test criteria 3, 4, and 5 will be checked. In chapter Application of correction factors and terms there is an exemplary evaluation for the event that criterion 5 cannot be met without application of correction factors or terms.

128 Page 128 of Determination of uncertainty between systems under test u bs For the test of PM 2.5 measuring systems the uncertainty between the systems under test shall be determined according to chapter of the Guide Demonstration of equivalence of Ambient Air Monitoring Methods in the field test at least at four sampling sites representative of the future application. 6.2 Equipment No equipment is necessary to test this performance criterion. 6.3 Method The test was carried out at four different comparisons during the field test. Different seasons and varying concentrations for PM 2.5 were taken into consideration. At least 20 % of the concentration values from the complete dataset determined with the reference method shall exceed the upper assessment threshold according to 2008/50/EC [7]. The upper assessment threshold is 17 µg/m³ for PM 2.5. At least 40 valid data pairs were determined per comparison. Out of the complete dataset (4 test sites, 213 valid data pairs for SN and 213 valid data pairs for SN ), 27.2 % of the measured values exceed the upper assessment threshold of 17 µg/m for PM 2.5. The measured concentrations are related to ambient conditions. 6.4 Evaluation According to chapter of the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods the following applies: The uncertainty between the candidates u bs shall be 2.5 µg/m³. If the uncertainty between the candidates exceeds 2.5 µg/m³, one or both systems might not be working properly. In such a case, equivalence cannot be declared. Uncertainty is determined for: All test sites/comparisons together (full dataset) 1 dataset with measured values 18 µg/m³ for PM 2.5 (basis: mean values of reference measurement)

129 Page 129 of 269 In addition to that, this report provides an evaluation of the following datasets: Each test site/comparison separately 1 dataset with measured values < 18 µg/m³ for PM 2.5 (basis: mean values of reference measurement) The uncertainty between the candidates u bs is calculated from the differences of all daily mean values (24 h values) of the simultaneously operated candidates by means of the following equation: u 2 bs = n i= 1 (y i,1 y 2n i,2 ) 2 with y i,1 and y i,2 = results of the parallel measurements of individual 24 h values i n = number of 24 h values 6.5 Assessment The uncertainty between the candidates u bs with a maximum of 0.84 µg/m³ for PM 2.5 does not exceed the required value of 2.5 µg/m³. Performance criterion met? yes

130 Page 130 of Detailed presentation of test results Table 26 lists the calculated values for the uncertainty between candidates u bs. Graphical representations of the results are provided in Figure 29 to Figure 35. Table 26: Uncertainty between candidates u bs for the devices SN and SN , measured component PM 2.5 Device Test site No. of values SN / / / / / / / Uncertainty u bs µg/m³ All test sites Single test sites Bonn, winter Bornheim, summer Cologne, autumn Cologne, winter Classification over reference value Values 18 µg/m³ Values < 18 µg/m³

131 Page 131 of 269 F , SN / SN , All comparisons, Raw data Candidate 2 [µg/m³] Measured values Regression line y=x y = x R² = Candidate 1 [µg/m³] Figure 29: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, all test sites 130 F , SN / SN , Bonn, Raw data Candidate 2 [µg/m³] Measured values Regression line y=x y = x R² = Candidate 1 [µg/m³] Figure 30: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Bonn, road junction, winter

132 Page 132 of F , SN / SN , Bornheim, Raw data Candidate 2 [µg/m³] Measured values Regression line y=x y = x R² = Candidate 1 [µg/m³] Figure 31: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Bornheim, summer F , SN / SN , Cologne, Autumn, Raw data Candidate 2 [µg/m³] Measured values Regression line y=x y = x R² = Candidate 1 [µg/m³] Figure 32: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Cologne, autumn

133 Page 133 of 269 F , SN / SN , Cologne, Winter, Raw data Candidate 2 [µg/m³] Measured values Regression line y=x y = x R² = Candidate 1 [µg/m³] Figure 33: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, test site Cologne, winter F , SN / SN , All comparisons, 18 µg/m³, Raw data Candidate 2 [µg/m³] Measured values Regression line y=x y = x R² = Candidate 1 [µg/m³] Figure 34: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, all test sites, values 18 µg/m³

134 Page 134 of 269 F , SN / SN , All comparisons, <18 µg/m³, Raw data Candidate 2 [µg/m³] Measured values Regression line y=x y = x R² = Candidate 1 [µg/m³] Figure 35: Results of the parallel measurements with the devices SN / SN , measured component PM 2.5, all test sites, values < 18 µg/m³

135 Page 135 of Calculation of expanded uncertainty between systems under test For the test of PM 2.5 measuring systems the equivalency with reference method shall be demonstrated according to chapter to 9.6 of the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods in the field test at least at four sampling sites representative of the future application. The maximum expanded uncertainty of the systems under test shall be compared with data quality objectives to Annex A of VDI Standard 4202, Sheet 1 (September 2010). 6.2 Equipment Additional instruments according to item 5 of this report were used in the testing of this performance criterion. 6.3 Method The test was carried out at four different comparisons during the field test. Different seasons and varying concentrations for PM 2.5 were taken into consideration. At least 20 % of the concentration values from the complete dataset determined with the reference method shall exceed the upper assessment threshold according to 2008/50/EC [7]. The upper assessment threshold is 17 µg/m³ for PM 2.5. At least 40 valid data pairs were determined per comparison. Out of the complete dataset (4 test sites, 213 valid data pairs for SN and 213 valid data pairs for SN ), 27.2 % of the measured values exceed the upper assessment threshold of 17 µg/m for PM 2.5. The measured concentrations are related to ambient conditions. 6.4 Evaluation [Item ] The calculation of expanded uncertainty is preceded by an uncertainty check between the two simultaneously operated reference devices u ref. The uncertainty between the simultaneously operated reference devices is determined analogous to the uncertainty between the candidates and shall be 2 µg/m³. The evaluated results are given in 7.6 of this test item. In order to evaluate the comparability between the candidates y and the reference method x, a linear correlation y i = a + bx i between the measured results obtained from both methods is assumed. The correlation between the mean values of the reference devices and the candidates, which shall be assessed individually, is established by means of orthogonal regression. Regression is calculated for: All test sites/comparisons together Each test site/comparison separately 1 dataset with measured values 18 µg/m³ for PM 2.5 (basis: mean values of reference measurement)

136 Page 136 of 269 For further evaluation, the results of the uncertainty u c_s of the candidates compared with the reference method is described in the following equation, which describes u CR as a function of the OM concentration x i. 2 RSS 2 u [ ] 2 CR (y i ) = u (x i ) + a + (b 1) x i (n 2) With RSS = Sum of the (relative) residuals from orthogonal regression u(x i ) = random uncertainty of the reference procedure, if the value u bs, which is calculated for using the candidates, can be used in this test (refer to item Determination of uncertainty between systems under test ubs) Algorithms for the calculation of intercept a as well as slope b and its variances by means of orthogonal regression are specified in Annex B of [4]. The sum of the (relative) residuals RSS is calculated using the following equation: RSS = n i= 1 (y a bx 2 i i ) Uncertainty u CR is calculated for: All test sites/comparisons together Each test site/comparison separately 1 dataset with measured values 18 µg/m³ for PM 2.5 (basis: mean values of reference measurement) According to the Guide, preconditions for acceptance of the full dataset are that: the slope b differs insignificantly from 1: b 1 2 u(b) and that the intercept a differs insignificantly from 0: a 2 u(a) with u(b) and u(a) being the standard uncertainties of slope and intercept, each calculated as the square root of their variances. If these preconditions are not met, the candidates may be calibrated according to item 9.7 of the guideline (refer to Application of correction factors and terms. The calibration shall only be applied to the full dataset.

137 Page 137 of 269 [Item 9.5.4] The combined uncertainty of the candidates w c,cm is calculated for each dataset by combining the contributions from and according to the following equation: u w (y ) = 2 c,cm i (y ) 2 CR i 2 y i For each dataset, the uncertainty w c,cm is calculated at the level of y i = 30 µg/m³ for PM 2.5. [Item 9.5.5] The expanded relative uncertainty of the results of the candidates is calculated for each dataset by multiplying w c,cm with a coverage factor k according to the following equation: WCM = k w CM In praxis k=2 for large n [Item 9.6] The highest resulting uncertainty W CM is compared with the requirements on data quality of ambient air measurements according to EU Standard [7] and assessed. There are two possible results: 1. W CM W dqo Candidate method is considered equivalent to the reference method 2. W CM > W dqo Candidate method is considered not equivalent to the reference method The specified expanded relative uncertainty W dqo for particulate matter is 25 % [7]. 6.5 Assessment The determined uncertainties W CM for all datasets under consideration lie below the defined expanded relative uncertainty W dqo of 25 % for suspended particulate matter without application of correction factors. Performance criterion met? yes Because of the significance of the slope and the intercept for the complete data set, correction factors are applied according to chapter Application of correction factors and terms. Table 27 provides an overview of all results from the equivalence test of the F for PM 2.5. In the event that a criterion has not been met, the respective cell is marked in red.

138 Page 138 of 269 Table 27: Overview of equivalence test of F for PM 2.5 Comparison candidate with reference according to Guide "Demonstration of Equivalence Of Ambient Air Monitoring Methods", January 2010 Candidate F SN SN / SN Limit value 30 µg/m³ Status of measured values Raw data Allowed uncertainty 25 % All comparisons Uncertainty between Reference 0.58 µg/m³ Uncertainty between Candidates 0.61 µg/m³ SN / SN Number of data pairs 213 Slope b significant Uncertainty of b Ordinate intercept a significant Uncertainty of a Expanded meas. uncertainty W CM % All comparisons, 18 µg/m³ Uncertainty between Reference 0.70 µg/m³ Uncertainty between Candidates 0.84 µg/m³ SN / SN Number of data pairs 53 Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty W CM % All comparisons, <18 µg/m³ Uncertainty between Reference 0.53 µg/m³ Uncertainty between Candidates 0.50 µg/m³ SN / SN Number of data pairs 160 Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty W CM %

139 Page 139 of 269 Comparison candidate with reference according to Guide "Demonstration of Equivalence Of Ambient Air Monitoring Methods", January 2010 Candidate F SN SN / SN Limit value 30 µg/m³ Status of measured values Raw data Allowed uncertainty 25 % Uncertainty between Reference 0.62 µg/m³ Uncertainty between Candidates 0.62 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM % % Uncertainty between Reference 0.52 µg/m³ Uncertainty between Candidates 0.45 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM 6.70 % 8.07 % Uncertainty between Reference 0.65 µg/m³ Uncertainty between Candidates 0.81 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM % % Uncertainty between Reference 0.49 µg/m³ Uncertainty between Candidates 0.33 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM % % Uncertainty between Reference 0.70 µg/m³ Uncertainty between Candidates 0.84 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM % % All comparisons, <18 µg/m³ Uncertainty between Reference 0.53 µg/m³ Uncertainty between Candidates 0.50 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM % % Bonn Bornheim Cologne, Autumn Cologne, Winter All comparisons, 18 µg/m³ All comparisons Uncertainty between Reference 0.58 µg/m³ Uncertainty between Candidates 0.61 µg/m³ SN SN Number of data pairs Slope b significant significant Uncertainty of b Ordinate intercept a significant significant Uncertainty of a Expanded meas. uncertainty WCM % %

140 Page 140 of 269 The results of the check of the five criteria given in chapter 6.1 Methodology of the equivalence check (modules ) are as follows: Criterion 1: More than 20 % of the data are greater than 17 µg/m³. Criterion 2: The uncertainty between the candidates is less than 2.5 µg/m³. Criterion 3: The uncertainty between the reference devices is less than 2.0 µg/m³. Criterion 4: All of the expanded uncertainties are below 25 %. Criterion 5: The slopes and the intercepts obtained from the evaluation of the full dataset are significantly greater than the permissible values for SN and SN Other: The evaluation of the full data set for both candidates together shows, that the measuring system has got a very good correlation with the reference method with a slope of and an intercept of at an expanded overall uncertainty of 14.6 %. The January 2010 version of the Guide is ambiguous with respect to which slope and which intercept should be used to correct a candidate should it fail the test of equivalence. After consultation with the convenor (Mr Theo Hafkenscheid) of the EC working group responsible for setting up the Guide, it was decided that the requirements of the November 2005 version of the Guide are still valid, and that the slope and intercept from the orthogonal regression of all the paired data be used. These are stated additionally under Other in the above. The 2006 UK Equivalence Report [9] has highlighted this was a flaw in the mathematics required for equivalence as per the November 2005 version of the Guide as it penalised instruments that were more accurate (Annex E Section 4.2 therein). This same flaw is copied in the January 2010 version. Hence, the F measuring system for PM 2.5 is indeed being penalised by the mathematics for being accurate. It is proposed that the same pragmatic approach is taken here that was previously undertaken in earlier studies. Therefore, according to Table 27, the slope and intercept should be corrected for PM 2.5 due to its determined significance. Nonetheless it should be noted that, even without application of correction factors, the determined uncertainties W CM for PM 2.5 lie below the specified expanded relative uncertainty W dqo of 25 % for particulate matter for all datasets considered.

141 Page 141 of 269 The slope for the complete dataset is The intercept for the complete dataset is Thus, an additional evaluation applying the respective calibration factors to the datasets is made in chapter Application of correction factors and terms. The revised January 2010 version of the Guide requires that, in order to monitor the processes in compliance with the Directive, random checks shall be performed on a number of systems within a measuring network and that the number of measuring sites shall depend on the expanded uncertainty of the system. Either the network operator or the responsible authority of the member state is responsible for the appropriate realisation of the requirement mentioned above. However, TÜV Rheinland recommends that the expanded uncertainty for the full dataset (here: uncorrected raw data) shall be referred to, i.e % for PM2.5, which would require annual checks at 3 sites (Guide [4], Chapter 9.9.2, Table 6). Due to the necessary application of the corresponding calibration factors, this assessment should be made on the basis of the evaluation of the corrected datasets (refer to chapter Application of correction factors and terms).

142 Page 142 of Detailed presentation of test results Table 28 presents an overview of the uncertainties between the reference devices u ref obtained in the field tests. Table 28: Uncertainty between reference devices u ref for PM 2.5 Reference devices No. Test site No. of values Uncertainty u bs µg/m³ 1 / 2 Bonn, winter / 2 Bornheim, summer / 2 Cologne, autumn / 2 Cologne, winter / 2 All test sites The uncertainty between the reference devices u ref is < 2 µg/m³ for all test sites.

143 Page 143 of 269 Reference vs. F , SN , PM2.5, All comparisons, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 36: Reference device vs. candidate, SN , measured component PM 2.5, all test sites Reference vs. F , SN , PM2.5, All comparisons, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 37: Reference device vs. candidate, SN , measured component PM 2.5, all test sites

144 Page 144 of 269 Reference vs. F , SN , PM2.5, Bonn, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 38: Reference device vs. candidate, SN , measured component PM 2.5, Bonn, road junction, winter Reference vs. F , SN , PM2.5, Bonn, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 39: Reference device vs. candidate, SN , measured component PM 2.5, Bonn, road junction, winter

145 Page 145 of 269 Reference vs. F , SN , PM2.5, Bornheim, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 40: Reference device vs. candidate, SN , measured component PM 2.5, Bornheim, summer Reference vs. F , SN , PM2.5, Bornheim, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 41: Reference device vs. candidate, SN , measured component PM 2.5, Bornheim, summer

146 Page 146 of 269 Reference vs. F , SN , PM2.5, Cologne, Autumn, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 42: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, autumn Reference vs. F , SN , PM2.5, Cologne, Autumn, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 43: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, autumn

147 Page 147 of 269 Reference vs. F , SN , PM2.5, Cologne, Winter, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 44: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, winter Reference vs. F , SN , PM2.5, Cologne, Winter, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 45: Reference device vs. candidate, SN , measured component PM 2.5, Cologne, winter

148 Page 148 of 269 Reference vs. F , SN , PM2.5, All comparisons, 18 µg/m³, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 46: Reference device vs. candidate, SN , measured component PM 2.5, values 18 µg/m³ Reference vs. F , SN , PM2.5, All comparisons, 18 µg/m³, Raw data Candidate [µg/m³] Reference [µg/m³] Figure 47: Reference device vs. candidate, SN , measured component PM 2.5, values 18 µg/m³

149 Page 149 of Application of correction factors and terms If the maximum expanded uncertainty of the systems under test exceeds the data quality objectives according to Annex B of Standard VDI 4202, Sheet 1 (September 2010) for the test of PM 2.5 measuring systems, the application of factors and terms is allowed. Values corrected shall meet the requirements of chapter ff of the Guide Demonstration of Equivalence of Ambient Air Monitoring Methods. The tests were also carried out for the component PM Equipment No equipment is necessary to test this performance criterion. 6.3 Method Refer to module Evaluation If evaluation of the raw data according to module leads to a case where W CM > W dqo, which means that the candidate systems is not regarded equivalent to the reference method, it is permitted to apply a correction factor or term resulting from the regression equation obtained from the full dataset. The corrected values shall satisfy the requirements for all datasets or subsets (refer to module ). Moreover, a correction factor may be applied even for W CM W dqo in order to improve the accuracy of the candidate systems. Three different cases may occur: a) Slope b not significantly different from 1: b 1 2u(b), intercept a significantly different from 0: a > 2u(a) b) Slope b significantly different from 1: b 1 > 2u(b), intercept a not significantly different from 0: a 2u(a) c) Slope b significantly different from 1: b 1 > 2u(b) intercept a significantly different from 0: a > 2u(a) With respect to a) The value of the intercept a may be used as a correction term to correct all input values y i according to the following equation. y = y a i,corr i

150 Page 150 of 269 The resulting values of y i,corr may then be used to calculate the following new terms by linear regression: y = c + dx and i,corr 2 RSS 2 uc _ s (y i,corr ) = u (x i ) + i + (n 2) i 2 2 [ c + (d 1)x ] u (a) with u(a) = uncertainty of the original intercept a, the value of which has been used to obtain y i,corr. Algorithms for the calculation of intercepts as well as slopes and their variances by orthogonal regression are described in detail in annex B of [4]. RSS is determined analogue to the calculation in module With respect to b) The value of the slope b may be used as a term to correct all input values y i according to the following equation. y i y i,corr = b The resulting values of y i,corr may then be used to calculate the following new terms by linear regression: y = c + dx and u (y i,corr RSS 2 ) = u (x i ) + (n 2) i [ c + (d 1)x ] x u (b) 2 c _ s i,corr i + i with u(b) = uncertainty of the original slope b, the value of which has been used to obtain y i,corr. Algorithms for the calculation of intercepts as well as slopes and their variances by orthogonal regression are described in detail in annex B of [4]. RSS is determined analogue to the calculation in module With respect to c) The values of the slope b and of the intercept a may be used as correction terms to correct all input values y i according to the following equation. y i a y i,corr = b The resulting values of y i,corr may then be used to calculate the following new terms by linear regression: y = c + dx i,corr i

151 Page 151 of 269 and 2 RSS 2 uc _ s (y i,corr ) = u (x i ) + i i + (n 2) [ c + (d 1)x ] + x u (b) u (a) with u(b) = uncertainty of the original slope b, the value of which has been used to obtain y i,corr and with u(a) = uncertainty of the original intercept a, the value of which has been used to obtain y i,corr. Algorithms for the calculation of intercepts as well as slopes and their variances by orthogonal regression are described in detail in Annex B of [4]. RSS is determined analogue to the calculation in module The values for u c_s,corr are used for the calculation of the combined relative uncertainty of the candidate systems after correction according to the following equation: u w (y ) = 2 c,cm,corr i 2 c _ s,corr y 2 i (y ) i For the corrected dataset, uncertainty w c,cm,corr is calculated at the daily limit value by taking y i as the concentration at the limit value. The expanded relative uncertainty W CM,corr is calculated according to the following equation: W = k w CM,corr In practice: k=2 for large number of available experimental results The highest resulting uncertainty W CM,corr is compared and assessed with the requirements on data quality of ambient air measurements according to EU Standard [7]. Two results are possible: 1. W CM W dqo Candidate method is accepted as equivalent to the standard method. 2. W CM > W dqo Candidate method is not accepted as equivalent to the standard method. The specified expanded relative uncertainty W dqo for particulate matter is 25 % [7]. 6.5 Assessment The candidates meet the requirements on data quality of ambient air quality measurements already without application of correction factors. A correction of the slope and the intercept nevertheless leads to a further significant improvement of the expanded measurement uncertainties of the full data set. Performance criterion met? yes CM,corr

152 Page 152 of 269 The evaluation of the full dataset for both candidates shows a significant slope and a significant intercept for the measuring component PM 2.5. The slope for the full dataset is The intercept for the full dataset is (refer to Table 27). Slope and intercept were corrected for the complete dataset. All datasets were then reevaluated using the corrected values. After correction, all datasets fulfil the requirements on data quality and the measurement uncertainties improve significantly at some sites. Only the test site Bornheim, summer gets worse significantly by the correction, but the expanded uncertainty is still below the permissible 25 %. The January 2010 version of the Guide requires that the systems are tested annually at a number of sites corresponding to the highest expanded uncertainty found during equivalence testing, if the AMS is operated within a network. The corresponding criterion for determining the number of test sites is divided into 5 % steps (Guide [4], chapter 9.9.2, Table 6). It should be noted that the highest expanded uncertainty determined for PM 2.5 lies in the range of <10% after correction, whereas it is in the range 10 % to 15 % before the correction. The network operator or the responsible authority of the member state is responsible for the appropriate realisation of the required regular checks in networks mentioned above. However, TÜV Rheinland recommends to use the expanded uncertainty for the full dataset, i.e. for PM % (uncorrected dataset) and 8.5 % (dataset after slope/offset correction), which would require an annual test at 3 measurement sites (uncorrected) or 2 measurement sites (corrected).

153 Page 153 of Detailed presentation of test results Table 29 presents the results of the evaluations of the equivalence test after application of the correction factors for slope and intercept on the full dataset. Table 29: Summary of the results of the equivalence test, SN & SN , measured component PM 2.5 after correction of slope / intercept Comparison candidate with reference according to Guide "Demonstration of Equivalence Of Ambient Air Monitoring Methods", January 2010 Candidate F SN SN / SN Limit value 30 µg/m³ Status of measured values Slope and offset corrected Allowed uncertainty 25 % All comparisons Uncertainty between Reference 0.58 µg/m³ Uncertainty between Candidates 0.67 µg/m³ SN / SN Number of data pairs 213 Slope b not significant Uncertainty of b Ordinate intercept a not significant Uncertainty of a Expanded meas. uncertainty W CM 8.46 % All comparisons, 18 µg/m³ Uncertainty between Reference 0.70 µg/m³ Uncertainty between Candidates 0.92 µg/m³ SN / SN Number of data pairs 53 Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty W CM % All comparisons, <18 µg/m³ Uncertainty between Reference 0.53 µg/m³ Uncertainty between Candidates 0.54 µg/m³ SN / SN Number of data pairs 160 Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty W CM 8.36 %

154 Page 154 of 269 Comparison candidate with reference according to Guide "Demonstration of Equivalence Of Ambient Air Monitoring Methods", January 2010 Candidate F SN SN / SN Limit value 30 µg/m³ Status of measured values Slope and offset corrected Allowed uncertainty 25 % Uncertainty between Reference 0.62 µg/m³ Uncertainty between Candidates 0.67 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM 9.79 % % Uncertainty between Reference 0.52 µg/m³ Uncertainty between Candidates 0.49 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM % % Uncertainty between Reference 0.65 µg/m³ Uncertainty between Candidates 0.89 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM 7.54 % % Uncertainty between Reference 0.49 µg/m³ Uncertainty between Candidates 0.36 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM % % Uncertainty between Reference 0.70 µg/m³ Uncertainty between Candidates 0.92 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM 9.81 % % All comparisons, <18 µg/m³ Uncertainty between Reference 0.53 µg/m³ Uncertainty between Candidates 0.54 µg/m³ SN SN Number of data pairs Slope b Uncertainty of b Ordinate intercept a Uncertainty of a Expanded meas. uncertainty WCM 8.15 % % Bonn Bornheim Cologne, Autumn Cologne, Winter All comparisons, 18 µg/m³ All comparisons Uncertainty between Reference 0.58 µg/m³ Uncertainty between Candidates 0.67 µg/m³ SN SN Number of data pairs Slope b not significant not significant Uncertainty of b Ordinate intercept a not significant not significant Uncertainty of a Expanded meas. uncertainty WCM 8.33 % 9.64 %

155 Page 155 of Requirements on multiple-component measuring systems Multiple-component measuring systems shall comply with the requirements set for each component, also in the case of simultaneous operation of all measuring channels. 6.2 Equipment Not applicable. 6.3 Method Not applicable. 6.4 Evaluation Not applicable. 6.5 Assessment Not applicable. Performance criterion met? Detailed presentation of test results Not applicable.

156 Page 156 of Recommendations for practical use Works in the maintenance interval (4 weeks) The following procedures are required to be carried out at regular intervals: Regular visual inspection / telemetrical monitoring Check of instrument status No error messages No contamination Check of instrument functions according to manufacturer Check of filter tape stock Maintenance of sampling inlet according to manufacturer As for the rest, the instructions and recommendations provided by the manufacturer shall be followed. Further maintenance work In addition to the regular maintenance work in the maintenance interval, the following procedures are necessary: Check of the filter tape stock A filter tape with 45 m length is lasting theoretically for 30,000 measurement cycles in case of a cycle time of 1 h and a sample count of 24 (setting during type approval test), which equals to 1,250 days. As - because of a possible exceedance of the maximum allowed mass per filter spot of 400 µg, which is dependent on the PM concentration level under practical conditions - a new filter spot might be used earlier than by reaching the 24times sampling, the time for consumption of the filter tape might be reduced accordingly. In case of a cycle time of 1 h and a minimum sample count of 1 (i.e. for each cycle a new filter spot is used), there are 1,250 measurement cycles possible, which equals 52 days. Thus it is recommended to check the filter tape stock during each visit of the measuring system for regular maintenance (e.g. in the framework of cleaning the sampling inlet). According to the manufacturer. the pump needs maintenance approx. every 6 weeks after one year of operation, i.e. the filters must be blown out and the blade height has to be checked and the blades to be replaced if necessary. According to CEN/TS [8], a check of the sensors for ambient temperature and ambient pressure should be done every 3 months. According to CEN/TS [8], a check of the sampling flow rate should be done every 3 months. Within the framework of the check of the sampling flow rate, the tightness should also be checked every 3 months. The filter adapter, the transport reel and the pressure rollers have to be cleaned every 6 months.

157 Page 157 of 269 The waste air filter and the hose connections have to be checked and if necessary blown out every 6 months. The pump filter and seals shall be replaced once a year. Once per year, the meteorological sensor has to be sent to the manufacturer for recalibration. Furthermore it is recommended to check the radiometric measurement with the help of the reference foil once a year. During the annual maintenance, the cleaning of the sampling tube needs also to be considered. Further details are provided in the user manual. Department of Environmental Protection/ Dipl.-Ing. Karsten Pletscher Dipl.-Ing. Guido Baum Cologne, 17th March / /A

158 Page 158 of Literature [1] VDI Standard 4202, Part 1, Performance criteria for type approval tests of automated ambient air measuring systems Point-related measurement methods for gaseous and particulate air pollutants, June 2002 & September 2010 [2] VDI Standard 4203, Part 3, Testing of automated measuring systems Test procedures for point-related ambient air measuring systems for gaseous and particulate air pollutants, August 2004 & September 2010 [3] Standard EN 14907, Ambient air quality Standard gravimetric measurement method for the determination of the PM 2.5 mass fraction of suspended particulate matter, German version EN 14907: 2005 [4] Guidance document Demonstration of Equivalence of Ambient Air Monitoring Methods, English version of January 2010 [5] Operator s manual F Rev. 04 von 03 / 2014 [6] Operator s manual LVS3, Stand 2000 [7] Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe [8] Technical Specification CEN/TS 16450, Ambient air Automated measuring systems for the measurement of the concentration of particulate matter (PM10; PM2.5) ; German version, August 2013 [9] Report UK Equivalence Programme for Monitoring of Particulate Matter, Report No.: BV/AQ/AD202209/DH/2396 of 5 June 2006

159 Page 159 of Annex Appendix 1 Annex 1: Annex 2: Annex 3: Annex 4: Annex 5: Annex 6: Measured and calculated values Detection limit Temperature dependence of zero point Temperature dependence of the sensitivity Dependence on supply voltage Measured values at the field test sites Ambient conditions at the field test sites Appendix 2 Filter weighing procedure Appendix 3 Manuals

160 Page 160 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 1 Detection limit Page 1 of 1 Manufacturer DURAG GmbH Type F Standards ZP Measured values with zero filter Serial-No. SN / SN No. Date Measured values [µg/m³] Date Measured values [µg/m³] SN SN No. of values 15 No. of values 15 Mean 0.28 Mean 0.34 Standard deviation s x Standard deviation s x Detection limit x 0.66 Detection limit x s xo = ( ) (x 0i x 0 ) n 1 i= 1,n 2

161 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 161 of 269 Annex 2 Dependence of zero point on ambient temperature Page 1 of 1 Manufacturer Type DURAG GmbH F Standards ZP Zero filter Serial-No. SN / SN Cycle 1 Cycle 2 Cycle 3 SN Temperature Measured value Dev. Measured value Dev. Measured value Dev. No. [ C] [µg/m³] [µg/m³] [µg/m³] [µg/m³] [µg/m³] [µg/m³] ZP SN Temperature Measured value Dev. Measured value Dev. Measured value Dev. No. [ C] [µg/m³] [µg/m³] [µg/m³] [µg/m³] [µg/m³] [µg/m³] ZP

162 Page 162 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 3 Dependence of measured value on ambient temperature Page 1 of 1 Manufacturer Type DURAG GmbH F Standards Reference foil Serial-No. SN / SN Cycle 1 Cycle 2 Cycle 3 SN Temperature Measured value Dev. Measured value Dev. Measured value Dev. No. [ C] [%] [%] [%] RP SN Temperature Measured value Dev. Measured value Dev. Measured value Dev. No. [ C] [%] [%] [%] RP

163 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 163 of 269 Annex 4 Dependence of measured value on mains voltage Page 1 of 1 Manufacturer Type DURAG GmbH F Standards Reference foil Serial-No. SN / SN Cycle 1 Cycle 2 Cycle 3 SN Mains voltage Measured value Dev. Measured value Dev. Measured value Dev. No. [V] [%] [%] [%] RP SN Mains voltage Measured value Dev. Measured value Dev. Measured value Dev. No. [V] [%] [%] [%] RP

164 Page 164 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 5 Measured values from field test sites, related to actual conditions Page 1 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM2,5 PM2,5 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 1 2/28/2013 Zero point Bonn, Winter 2 3/1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/ /12/ /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/

165 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 165 of 269 Annex 5 Measured values from field test sites, related to actual conditions Page 2 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 31 3/30/2013 Zero point Bonn, Winter 32 3/31/2013 Zero point 33 4/1/2013 Zero point 34 4/2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/ /12/ /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/2013 Zero point 59 4/27/2013 Zero point 60 4/28/2013 Zero point

166 Page 166 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 5 Measured values from field test sites, related to actual conditions Page 3 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 61 4/29/ Bonn, Winter 62 4/30/ /1/ /2/ /3/ /4/ /5/ /6/ /14/2013 Zero point Bornheim, Summer 70 5/15/2013 Zero point 71 5/16/ /17/ /18/ /19/ /20/ /21/ Power loss Ref. PM2,5 G#1 77 5/22/ /23/ /24/ Power loss Ref. PM2,5 80 5/25/ /26/ Power loss Ref. PM2,5 G#2 82 5/27/ /28/ /29/ /30/ /31/ /1/ /2/ /3/ /4/

167 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 167 of 269 Annex 5 Measured values from field test sites, related to actual conditions Page 4 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 91 6/5/ Bornheim, Summer 92 6/6/ /7/ /8/ /9/ /10/ /11/ /12/ /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ Zero point 109 6/23/ Zero point 110 6/24/ /25/ /26/ /27/ /28/ /29/ /30/ /1/ /2/ /3/ /4/

168 Page 168 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 5 Measured values from field test sites, related to actual conditions Page 5 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 121 7/5/ Bornheim, Summer 122 7/6/ /7/ /8/ /9/ /10/ /11/ /12/ Outlier Ref. PM /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ Audits 141 7/25/ Switch to zero filter 142 9/4/2013 Zero point Cologne, Autumn 143 9/5/2013 discarded due to faulty flow cal /6/ discarded due to faulty flow cal /7/2013 discarded due to faulty flow cal /8/ discarded due to faulty flow cal /9/ discarded due to faulty flow cal /10/ discarded due to faulty flow cal /11/ discarded due to faulty flow cal /12/ discarded due to faulty flow cal.

169 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 169 of 269 Annex 5 Measured values from field test sites, related to actual conditions Page 6 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 151 9/13/ Cologne, Autumn 152 9/14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/ /12/

170 Page 170 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 5 Measured values from field test sites, related to actual conditions Page 7 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] /13/ Cologne, Autumn /14/ /15/2013 Maintenance SN /16/2013 Zero point /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /31/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/2013 Zero point /9/2013 Zero point /10/2013 Zero point /11/

171 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 171 of 269 Annex 5 Measured values from field test sites, related to actual conditions Page 8 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] /12/ Cologne, Autumn /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/

172 Page 172 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 5 Measured values from field test sites, related to actual conditions Page 9 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 241 1/13/ Zero point Cologne, Winter 242 1/14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /31/ /1/ /2/ /3/ /4/ /5/ /6/ /7/2014 Zero point 267 2/8/2014 Zero point 268 2/9/2014 Zero point 269 2/10/ /11/

173 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 173 of 269 Annex 5 Measured values from field test sites, related to actual conditions Page 10 of 10 Manufacturer DURAG PM2,5 Type of instrument F Measured values in µg/m³ (ACT) Serial-No. SN / SN No. Date Ref. 1 Ref. 2 Ref. 1 Ref 2. Ratio SN SN Remark Test site PM2,5 PM2,5 PM10 PM10 PM2,5/PM10 PM10 PM10 [µg/m³] [µg/m³] [µg/m³] [µg/m³] [%] [µg/m³] [µg/m³] 271 2/12/ Cologne, Winter 272 2/13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/

174 Page 174 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 6 Ambient conditions at the field test sites Page 1 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 1 2/28/2013 Bonn, Winter /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/ /12/ /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/

175 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 175 of 269 Annex 6 Ambient conditions at the field test sites Page 2 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 31 3/30/2013 Bonn, Winter /31/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/ /12/ /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/

176 Page 176 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 6 Ambient conditions at the field test sites Page 3 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 61 4/29/2013 Bonn, Winter /30/ /1/ /2/ /3/ /4/ /5/ /6/ /14/2013 Bornheim, Summer 70 5/15/2013 no weather data available 71 5/16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /31/ /1/ /2/ /3/ /4/

177 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 177 of 269 Annex 6 Ambient conditions at the field test sites Page 4 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 91 6/5/2013 Bornheim, Summer /6/ /7/ /8/ /9/ /10/ /11/ /12/ /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /1/ /2/ /3/ /4/

178 Page 178 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 6 Ambient conditions at the field test sites Page 5 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 121 7/5/2013 Bornheim, Summer /6/ /7/ /8/ /9/ /10/ /11/ /12/ /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /4/2013 Cologne, Autumn /5/ /6/ /7/ /8/ /9/ /10/ /11/ /12/

179 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 179 of 269 Annex 6 Ambient conditions at the field test sites Page 6 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 151 9/13/2013 Cologne, Autumn /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /1/ /2/ /3/ /4/ /5/2013 no weather data available /6/ /7/ /8/ /9/ /10/ /11/ /12/

180 Page 180 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 6 Ambient conditions at the field test sites Page 7 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] /13/2013 Cologne, Autumn /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /31/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/

181 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 181 of 269 Annex 6 Ambient conditions at the field test sites Page 8 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] /12/2013 Cologne, Autumn /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /1/ /2/2013 no weather data available /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/

182 Page 182 of 269 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Annex 6 Ambient conditions at the field test sites Page 9 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 241 1/13/2014 Cologne, Winter /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/ /27/ /28/ /29/ /30/ /31/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/ /10/ /11/

183 Report on type approval testing of the F measuring system with PM2.5-pre-separator manufactured by DURAG GmbH for the component suspended particulate matter PM2.5, Report no.: 936/ /A Page 183 of 269 Annex 6 Ambient conditions at the field test sites Page 10 of 10 No. Date Test site Ambient temperature (AVG) Ambient pressure Rel. air humidity Wind velocity Wind direction Precipitation [ C] [hpa] [%] [m/s] [ ] [mm] 271 2/12/2014 Cologne, Winter /13/ /14/ /15/ /16/ /17/ /18/ /19/ /20/ /21/ /22/ /23/ /24/ /25/ /26/2014 no weather data available 286 2/27/ /28/ /1/ /2/ /3/ /4/ /5/ /6/ /7/ /8/ /9/

184 Page 184 of 269 Appendix 2 Filter weighing procedure A.1 Carrying out the weighing All weighings are done in an air-conditioned weighing room. Ambient conditions are 20 C ±1 C and 50 % ±5 % relative humidity, which conforms to the requirements of Standard EN The filters used in the field test are weighed manually. In order to condition the filters (including control filters), they are placed on sieves to avoid overlap. The specifications for pre- and post-weighing are specified beforehand and conform to the Standard. Before sampling = pre-weighing Conditioning 48 h + 2 h Filter weighing Re-conditioning 24 h +2 h Filter weighing and immediate packaging After sampling = post-weighing Conditioning 48 h + 2 h Filter weighing Re-conditioning 24 h + 2 h Filter weighing The balance is always ready for use. An internal calibration process is started prior to each weighing series. The standard weight of 200 mg is weighed as reference and the boundary conditions are noted down if nothing out of the ordinary results from the calibration process. Deviations of prior weighings conform to the Standard and do not exceed 20 µg (refer to Figure 48). All six control filters are weighed afterwards and a warning is displayed for control filters with deviations > 40 µg during evaluation. These control filters are not used for postweighing. Instead, the first three acceptable control filters are used while the others remain in the protective jar in order to replace a defective or deviating filter, if necessary. Figure 49 shows an exemplary process over a period of more than four months. All filters which display a difference of more than 40 µg between the first and second weighing are excluded from the pre-weighing process. Filters exhibiting deviations of more than 60 µg are not considered for evaluation after post-weighing, as conforming to standards.

185 Page 185 of 269 Weighed filters are packed in separate polystyrene jars for transport and storage. These jars remain closed until the filter is inserted. Virgin filters can be stored in the weighing room for up to 28 days before sampling. Another pre-weighing is carried out if this period is exceeded. Sampled filters can be stored for up to 15 days at a temperature of 23 C or less. The filters are stored at 7 C in a refrigerator.

186 Page 186 of 269 A2 Filter evaluation The filters are evaluated with the help of a corrective term in order to minimise relative mass changes caused by the weighing room conditions. Equation: Dust = MF post ( M Tara x ( MKon post / MKon pre ) ) (F1) MKon pre = mean mass of the 3 control filters after 48 h and 72 h pre-weighing MKon post = mean mass of the 3 control filters after 48 h and 72 h post-weighing M Tara = mean mass of the filter after 48 h and 72 h pre-weighing MF post = mean mass of the loaded filter after 48 h and 72 h post-weighing Dust = corrected dust mass of the filter This shows that the method becomes independent from weighing room conditions due to the corrective calculation. Influence due to the water content of the filter mass between virgin and loaded filter can be controlled and do not change the dust content of sampled filters. Hence, point of EN is fulfilled. The example of the standard weight between November 2008 and February 2009 shows that the permissible difference of max. 20 µg from the previous measurement is not exceeded.

187 Page 187 of 269 Stability of standard weight between Nov 08 and Feb Weight [g] Standard weight Mean standard weight No. of weighing Figure 48: Stability of standard weight

188 Page 188 of 269 Table 30: Stability of standard weight Marked in yellow = average value Marked in green = lowest value Marked in blue = highest value Date Weighing No. Standard weight Difference to the previous weighing g µg

189 Page 189 of 269 Stability control filter Emfab Weight [g] No. of weighing TM1 TM2 TM3 Mean TM1 Mean TM2 Mean TM3 Figure 49: Stability of the control filters

190 Page 190 of 269 Table 31: Stability of the control filters Control filter no. Weighing no. TM1 TM2 TM Mean value Standard deviation E E E-05 Rel. standard deviation Median Lowest value Highest value Marked in yellow = average value Marked in green = lowest value Marked in blue = highest value

191 Page 191 of 269 Appendix 3 Manual

192 F Rev /2014 Beta dust meter F Technical manual Valid from operating software version 3.07 GmbH Kollaustraße Hamburg

193 Page 2 F Rev. 4

194 F Rev. 04 Page 3 Contents 1. Application Functional description F device version Functional overview of F Device dimensions Sampling Ambient temperature measurement (option) Air pressure measurement (option) Sample probe heater (option) Constituent analysis (option) Measuring device Sampling system (overview) Measuring device (overview) Front view Rear view Assembly plate module (front) Assembly plate module (rear) Device control and measured value calculation module Volumetric flow measurement and regulation module Installation and commissioning Checking the package supplied Choosing the measuring location Assembling and startup Sampling Sensors Leakage test Volumetric flow Electrical connection Operating the F Turning on Using the touchscreen Basic functions Measuring mode Data display mode Setup mode Maintenance menu / mode Data memory Viewing the data memory on the display Exporting the data using the RS232 interface Measured value output / Time responses / Diagram The GESYTEC protocol for transferring numerical data Basic requirements Setting the F parameters Protocol types and format details Getting started Examples Measuring error analysis / Setting cycle times... 55

195 Page 4 F Rev Maintenance Device maintenance Sample inlet maintenance Instructions for replacing a filter reel Error messages / Troubleshooting Calculating the filter tape and cover foil consumption Technical data Service documents Parameter list Vacuum pump specification type VTE Sample tube heater (option) Figures Fig. 1: Function diagram for F Fig. 2: F device dimensions Fig. 3: Measuring head types Fig. 4: Installation instructions for heated sampling system Fig. 5: Front internal view of measuring device Fig. 6: Rear internal view of measuring device Fig. 7: Double-walled sample system Fig. 8: F device interface / ma outputs / status signals ("Data ) Fig. 9: Serial RS232 connection between PC and F ("RS232") Fig. 10: Interface for optional temperature inputs on F ("Temperature") Fig. 11: Measuring menu after starting Fig. 12: Parameter menu structure Fig. 13: Time lapse for single filter spot load Fig. 14: Time lapse for multiple filter spot load Fig. 15: Time lapse for "zero point check" Fig. 16: Upper section of filter housing (from below) Fig. 17: Lower section of filter housing (top view) Fig. 18: Transport reel Equations (EQ1) Determination of dust mass... 7 (EQ2) Mass = f(a, ρ/µ, ln(n0/n))... 8 (EQ3) Dust concentration = f(m, Q)... 8 (EQ4) Recalculation of Span Volume (EQ5) Determination of absolute measuring error... 55

196 F Rev. 04 Page 5 Tables (Table 1) (Z/A) ratio for certain elements... 8 (Table 2) Examples of filter tape label (Table 3) Signal description for the 50-pin Sub-D connector (Table 4) Control parameters for communication using Gesytec protocol (Table 5) Maintenance overview (Table 6) Maintenance: Display check (Table 7) Error messages / Troubleshooting (Table 8) Technical data... 70

197 Page 6 F Rev. 4

198 F Rev. 04 Page 7 1. Application The F dust measuring device (Beta dust meter) is used for continuous monitoring and recording of the dust content in the ambient air. It is used in environmental measuring networks to monitor the air quality. The device contains a fully automatic gas sampling system. The extracted dust particles are collected on a filter tape and assessed. Identification of each dust sample with date, time and dust mass also allows subsequent analysis of the particle composition in the laboratory. This enables the influence of particular events, e.g. accidents or failures on adjacent industrial systems, to be traced. The device has 4/20mA signal outputs that transfer the measuring result to a connected data process system. Alternatively, the device has serial interfaces that can be used to transfer data and to control the device and/or to connect a log printer. The measurement system has been designed, manufactured and tested so that it is not suitable for use in explosive environments. 2. Functional description The Beta dust meter measures the dust concentration in µg dust per cubic meter of ambient air in µg/m³. The air volume is sucked through a glass fibre tape (GF), which separates out the dust particles onto the filter. The volumetric flow is controlled and recorded by a microcontroller. After the intake time, the mass collected on the filter is measured radiometrically. A measuring setup consisting of a Beta radiation source (C-14) and a Geiger Mueller counter tube (GM) is used to do this. The measuring principle for determining the dust mass is based on the fact that Beta rays are weakened as they pass through a material. The intensity of the radiation (pulses/ measuring time) is initially assessed after the rays pass through the unused filter paper. Once it has collected the dust, the intensity of the radiation is measured again. The ratio of the two intensity values is a measure of the quantity of dust collected on the filter spot (assuming homogeneous distribution on the filter surface) and, with a constant cross-sectional area of the loaded filter spot, a measure of the absolute dust mass. The absolute dust mass divided by the quantity of air taken in then gives the dust concentration. The radiometric measuring method is universally applicable, as it determines the mass of the dust within wide limits irrespective of the chemical and physical properties of the dust and the carrier gas. With homogeneous distribution of dust precipitation of the mass m on a filter area A F, up to 5 mg/cm² the relationship is approximately linear: ln (n 0 / n) = (µ / ρ ) d (EQ1) Determination of dust mass where: d= m/a F µ/ρ in µg/cm² is the dust surface density with dust precipitation in µg on the constant precipitation area in cm² in cm² /g is the mass attenuation coefficient µ in cm -1 is the linear attenuation coefficient of the Beta radiation used ρ n 0 and n in g/cm³ is the density of the absorber material are the Beta particles detected by the counter per minute, without or with the dust, registered electronically as voltage pulses. The pulse rate is a measure of the radiation intensity.

199 Page 8 F Rev. 4 The mass attenuation coefficient µ/ρ of the Beta radiation used depends on the electron density of the absorber and is thus proportional to the ratio (Z/A). where: Z A Chemical atomic number Mass number However, as the ratio (Z/A) = is approximately constant for most dusts that occur, in practical terms the Beta radiation attenuation is independent of the chemical composition and particle size distribution of the dust. Some elements and their (Z/A) ratio are listed below as examples: Element AL C Ca Fe K Mg Z/A Element Na O S Si Z/A (Table 1) (Z/A) ratio for certain elements One exception is hydrogen, for which (Z/A) = 1. However, the proportion of hydrogen in the dust is relatively low, which means that its differing behaviour does not interfere with the measuring principle. If the filter area remains constant, because (µ/ρ) = constant the dust mass precipitated on the filter A can be calculated from the radiation attenuation using the following equation: m = A F (ρ/µ) ln (n 0 /n) where: m A F (EQ2) Mass = f(a, ρ/µ, ln(n0/n)) Absolute dust mass in g Filter area in cm² As the mass attenuation coefficient (µ/ρ) increases as the maximum Beta energy decreases, determination of mass by Beta absorption measurement is more sensitive the weaker the energy of the Beta radiation used. Therefore, in addition to its long half-life of T 1/2 = 5730 years, the carbon isotope C-14 is particularly suitable, with a maximum energy of W max = MeV. The dust concentration is calculated from the absolute mass divided by the air volume taken in: c = m / Q (EQ3) Dust concentration = f(m, Q) where: c Dust concentration in µg/m³ Q Intake air volume in m³

200 F Rev. 04 Page 9 3. F device version 3.1. Functional overview of F The figure below shows the key components of the F system. The measuring device has a compact design. Apart from the sample probe (probe inlet, sampling head), the optional meteorological sensor and the optional sample probe heater, all components are contained in a single housing. Probeneinlass Sample Inlet Messung von T und rel. Feuchte (optional) Measurement of T and rel. Humidity (optional) Rohrbegleitheizung (optional) Sample Probe Heater (optional) Drucker / Linienschreiber Printer / Recorder Auswertung,Steuerung Data processing, Control Abdeckfolie (optional) Cover Foil (optional) Filterbanddrucker (optional) Filter Tape Printer (optional) GM-Zählrohr GM Counter Tube C14-Strahler C-14 Source Filterhalter Filter Adapter PC Lesen/ Schreiben von Daten vom/ zum Instrument Read/ Write Data from/ to Instrument Aufwickelrolle Take-up Reel Volumenstrom-Meter Flow-Meter Schrittmotor für Filtertransport Stepping Motor for Filter Transport Vorratsrolle Supply Reel Vakuumpumpe Vacuum Pump Filter Filter Bypass-Regelventil Bypass-Valve Fig. 1: Function diagram for F The measuring device is controlled by a microcontroller board. The filter tape is transported from the supply reel to the take-up reel (see 4.2.3) by a stepper motor. The Geiger Mueller counter tube uses the attenuation of the radiation intensity emitted by the C-14 radiation source to determine the increase in mass on the filter paper. The air is taken in by the pump, and the volumetric flow is measured by the volume sensor and regulated to a constant 1000 l/h by adjusting the bypass valve to the appropriate position. Electronics control the measuring processes, allowing user-friendly operation on a graphical touchscreen, and store the measured values. As an option, the device can protect the measured dust samples against smearing and loss with a cover film to allow subsequent laboratory investigation of dust constituents.

201 Page 10 F Rev. 4 During normal measurement, an unloaded filter spot is transported between the C-14 source and the GM counter tube at the beginning of the measurement. The beam intensity is then measured for 300 seconds, i.e. the pulses generated by the GM counter tube are evaluated as a measure of the detected beta radiation. The filter adapter is then opened and the filter tape is transported until this evaluated filter area is in the extraction position. The filter adapter is then closed again. The extraction process begins. The extraction time corresponds to the programmed cycle time (minus the measuring times). After completion of sampling, the filter adapter is opened again and the filter paper is moved back to its original position below the counter tube. The filter adapter closes and the beam intensity is measured for another 300 seconds. As shown in equation 2, the measured count rates are used to determine the dust mass collected on the filter. The dust concentration is calculated using the amount of air extracted, as shown in equation 3. The measuring results are shown in the display. They are also available as a 4/20mA current signal and via the RS232 data interface (e.g. using Bavaria-Hessen protocol / Gesytec). All measured values from the last 9 months are stored and can be viewed on the display or via the serial interface. As an option, you can print on the filter paper to identify the collected dust sample (e.g. for dust content analysis).

202 F Rev. 04 Page Device dimensions Frontansicht / Front view Abb 2 Geräteabmessungen_f _ Draufsicht / Top view Raum für Anschlüsse freihalten/ Keep this area clear for connections Gasauslass Gas Outlet Anschluss für Probenentnahmesonde Connection for Sample Probe Gas outlet Gas outlet Handgriff Handle Fig. 2: F device dimensions Handgriff Handle 30 45

203 Page 12 F Rev Sampling Different sampling heads (sample probes) are used depending on the measuring task. The TSP head is used for collection and measurement of the full dust. The PM10 head is used to record and evaluate dust fractions with an aerodynamic diameter 10 µm. Finally, the PM2.5 head is used to record and evaluate dust fractions with an aerodynamic diameter 2.5 µm. TSP-Einlass(VDI 2463) PM2.5-Einlass (EN14907) PM10-Einlass (NAAQS, PM2.5-Impaktor, nur mit äußerlich gleiche Bauart 40 CFR, part 50) PM10-Einlass (NAAQS, PM10-Einlass (EN12341) 40 CFR, part 50) TSP Inlet (VDI 2463) PM2.5 Inlet (EN14907) PM10 Inlet (NAAQS, PM2.5 impactor only externally same style 40 CFR, part 50) together with PM10 Inlet PM10 Inlet (EN12341) (NAAQS,40 CFR, part 50) Fig. 3: Measuring head types In many cases a basic probe tube can be used for PM10 measurements. To avoid condensation in the tube and/ or the sample volume an additional sample probe heater can be employed (see 3.6). For automatic operation the use of a PT100 temperature sensor (see 3.4) and an absolute pressure sensor for the measurement of ambient air pressure (see 3.5) is recommended. This guarantees a constant volumetric flow of 1 m³/h adjusted to ambient conditions, which is a precondition for the size dependent dust fractionation in the sample head. Figure 4 shows the installation of the basic sample system with heater. For PM2.5 measurements an active ventilated sample system with double-walled sample tube is available. For this system an additional tube heater isn t obligatory even at high relative humidity in ambient air (see 4.1). A ventilator unit continuously feeds air through the outer tube, so that the real probe tube inside is kept on ambient air temperature from top down to the sample volume. Power supply for the ventilator unit is provided by the device itself. An additional passive isolation of the outer tube might be helpful if extreme temperature differences between ambient air and installation space exist. For PM2.5 measurements we strongly recommend the installation of sensors for the measurement of ambient air pressure (see 3.5), temperature and relative humidity (see 3.4) to keep the volumetric flow constant and to heat the measurement volume not more than necessary. Thereby, not only the wanted dust size fraction is separated by the sample inlet, but also loss of volatile dust components is widely reduced.

204 F Rev. 04 Page 13 2, approx. 400 mm Wind cable 5 times around pipe PT with 4 m cable Stick excess cable in here Mounting the PT (siedeview) Wind heater 5 times around pipe and fasten 4-5 times using straps 9-pi Sub-D plug for PTs Strap all cables to pipe PT with 1m cable. Push underneath heater and fasten with strap Power plug for heater Fig. 4: Installation instructions for heated sampling system

205 Page 14 F Rev Ambient temperature measurement (option) An optional temperature sensor (PT100) can be used to measure the ambient air temperature. To do this, the sensor is installed on the sample tube in the radiation protected housing (see Fig. 4). The sensor signal is fed into the device via a 9-pin Sub-D connector labelled with Temperature on the rear and then evaluated and saved. Alternatively, a meteorological sensor with radiation shield can be used to measure temperature and relative humidity of ambient air (see 4.1). The sensor is hooked up to the connector labelled with Meteorology on the rear of the device. In conjunction with the value from the absolute pressure sensor (also optional), the temperature measurement in the external air is used to regulate the volumetric flow independently of the ambient air temperature and to determine the volumetric flow standardised to normal conditions (0 C, 1013 hpa). It is required particularly if the temperature of the measuring volume (in the filter adapter) is to be regulated depending on the external temperature to prevent condensation on the one hand and loss of volatile dust components on the other. It s possible to increase measuring volume temperature control by use of the meteorological sensor because the determination of relative humidity makes a control possible depending on dew point temperature. For PM2.5 measurements under directive EN14907 the use of the meteorological sensor for the simultaneously measurement of temperature and relative humidity is strongly recommended! 3.5. Air pressure measurement (option) As an option, an absolute pressure sensor can be installed in the device to measure the air pressure. The control electronics evaluate and save the measured values from the sensor. In conjunction with the value from the optional ambient temperature sensor, the air pressure measurement is used to regulate the volumetric flow independently of the ambient air pressure (e.g. in mountainous areas or in high or low pressure weather conditions) and to determine the volumetric flow standardized to normal conditions (0 C, 1013 hpa). The sensor is installed and configured in the factory, but can also be retrofitted by appropriately trained personnel. For PM2.5 measurements under directive EN14907 the use of the absolute pressure sensor is strongly recommended! 3.6. Sample probe heater (option) The sample probe heater may be required, depending on the location. This option is used if the conditions at the measuring location mean that condensation of the gas sample is possible. The precipitation of moisture that would occur on the inner wall of the sample tube in this situation would bond dust particles to be extracted in the sample tube before they are precipitated on the filter paper, thus falsifying the measuring results with an artificially low reading. With severe condensation, water droplets would fall from the inside of the sample tube through the filter adapter and onto the filter paper. The additional moisture on the filter spot would falsify the measuring result and may lead to irreparable damage to the volumetric flow probe. If condensation is anticipated, the sample tube heater should be installed as a precaution. A heating tape with adjustable temperature is wound around the probe tube, preventing condensation of moisture in the air. The heating tape length and temperature depend on the location and the local installation conditions. Figure 4 shows the installation of the heating tape with temperature sensor and the ambient air temperature sensor.

206 F Rev. 04 Page 15 Alternatively, a double-wall tube can be employed to keep the temperature in the inner sample tube nearly at the same level as in the ambient air by active ventilation. Using an additional isolation (e.g. commercially available Armaflex tube isolation) for the tube, inside the measurement chamber or container stable conditions will be ensured until a temperature difference of 40 degrees between ambient air and measurement chamber Constituent analysis (option) With the constituent analysis option, a dust spot is identified by a label after the spot on the filter tape and then protected by being covered with a transparent intermediate film layer. The label contains details about the start and end time of dust collection and the collected mass. The constituent analysis option includes: Dot matrix printer Control unit Mount for cover foil Cover foil As the printer is not positioned directly at the sampling point on the filter tape, the data for the dust spots is saved and printing is then carried out with a delay. If it is not included when delivered, the option should only be installed by trained personnel. Select "Active" for Filter printer in the "Service / Options" parameters. No other settings are necessary.

207 Page 16 F Rev. 4 Examples: The dust spot was collected on between 13:00 and 17:53. The calculated absolute mass is 54 µg :00 54 µg :53 Test print during filter tape feed by pressing the button with the "Forwards" symbol : TEST If asterisks are printed before and after the entry, an error has occurred or the user has executed another action during the measuring cycle :00 **** 54 µg **** :53 Effects: - Print position may not be correct. - The printed mass or time at the end of the sampling cycle may be incorrect. Reasons: - Error Filter crack - User activities: Opening/ closing filter housing, moving filter tape backwards, cycle test with reference film, stop by user - Device turned off Some errors can be corrected manually using the information about measured values and messages in the database, or the weight of the dust spots can be calculated. (Table 2) Examples of filter tape label

208 F Rev. 04 Page Measuring device 4.1. Sampling system (overview) Sampling head (example) Radiation shield with sensor for measurement of temperature and relative humidity Double-walled sample tube Active ventilation of the sample tube Sheath air out Drain hose Measurement device with touchscreen

209 Page 18 F Rev Measuring device (overview) Module Assembly plate module, front Front door, hinged on right Fig. 5: Front internal view of measuring device Module Assembly plate module, rear Module Device control and measured value calculation module Module Volumetric flow measurement and regulation module Fig. 6: Rear internal view of measuring device

210 F Rev. 04 Page Front view Front door lock Touch screen Front door, hinged on right Rear view Meteorology sensor input Temperature sensor input Power plug for heater Mains infeed Data outputs Device fan 24 VDC for external fan exhaust air

211 Page 20 F Rev Assembly plate module (front) Air inlet Mains switch Deflection reel Filter adapter, heated Supply reel with GF tape Filter pressure mechanism Filter transport reel Take-up reel C-14 source GM counter Assembly plate module (rear) Distribution card Filter transport stepper motor Filter adapter lifting mechanism Filter tape tension motors and controls

212 F Rev. 04 Page Device control and measured value calculation module Control for filter tape printer Measuring amplifier for GM tube Power supply 24 VDC Device control Absolute pressure measurement Signal converter 0-20mA to 0-1V Mains input filter Volumetric flow measurement and regulation module Bypass control valve Volumetric flow measurement Vacuum switch Vacuum pump Device fan Waste air filter 24 VDC plug Measuring gas outlet

213 Page 22 F Rev Installation and commissioning 5.1. Checking the package supplied After unpacking the device and all accessories, use the delivery note to check that the delivery is complete. Open the front door and remove the visible transport protection fittings. Report any transport damage immediately to the carrier and Durag GmbH Choosing the measuring location When using an F , the technical operating conditions for the measuring device must first be met. The standard version should never be operated outdoors, only in indoor working areas, measuring stations and other suitable locations. However, the sampling system for monitoring breathable air should be positioned indoors or outdoors depending on the measuring task. Installation fittings and roof conduits are available for this purpose. The F must be protected against direct exposure to moisture. The permitted ambient temperature range must be observed. If there are exceptional demands in terms of the operating conditions (outdoors, extended temperature range, unmanned operation etc.) special optional equipment is required. The F can be used as a table-top unit or as an insert in a 19" rack. In both cases, make sure that the device is operated on a flat horizontal surface and that the vertical sample tube can be routed from the top of the device to the sampling point without any obstructions or diversions. Particularly when used as an inserted in a 19" rack with multiple measuring equipment components, we recommend installing the F in the top position. Because of its weight, the device should be positioned on additional sliding brackets attached to the side. Make sure that the heat dissipation from the interior of the device is not prevented by modifications or other equipment. The ventilation slots in the base and rear panel of the device must always be kept clear. The setup of the device and sampling had should meet the European standard EN12341 for PM10 and EN14907 for PM2.5, respectively. When choosing the position of the sampling head, make sure that the measuring area around the head is free of objects and any other structural features and that the measuring location is not exposed to extreme air flows that would falsify the dust collection Assembling and startup Two different sample systems are available, one with a basic and one with a double-walled sample tube. The latter is recommended for PM2.5 measurements Sampling Two sampling systems are available, one with a basic sample tube and one with doublewalled sample tube. The latter is recommended for PM2.5 measurements Basic sampling system (Figure 4) Usually the assembling of device and basic sample tube is easy to do. Preferably, the sample tube is inserted to the measurement/ operating room from above together with the PT100 temperature sensor cable through the delivered roof penetration. After this the sample tube may be screwed to the gas inlet of the device by the connecting nut. Outside the sample head is put on the top of the tube and screwed in place. The PT100 temperature sensor is mounted inside the radiation shield as shown in Figure 4. If larger differences are expected between inside and outside temperature the sample probe heater should be wound around the tube inside the maybe air conditioned operating room (see Figure 4). The second PT100

214 F Rev. 04 Page 23 temperature sensor must be stuck between tube and heater band. Secure the heater and all cable with cable straps Double-walled sample system (Figure 7) Installation with the double-walled sample system is somewhat more complex and therefore described here in more detail. The system will be delivered pre-assembled with shield tube (3) and sample tube (4). For the mounting on the roof first remove the air shield (6) by unscrewing the three hexagon socket head cap screws. Put the shield tube through the roof penetration (1) from bottom-up and bring it close to its final position. Now put the tube with the bend in front through the roof opening and correct the tube length down or up (if necessary), so that the sample tube may be mounted to the F Screw the roof penetration in place. Screw the sample tube (4) to the gas inlet of F without tensioning. Pay attention, that the bend is aligned in direction to the rear panel of the device. Slide the clamping ring (8) on the shield tube (3) from above down to the bushing at the roof penetration and fix it with the hexagon socket head cap screws. Slide protective cover No. 3 (7) down to the clamping ring, so that rain may run off. Slide protective cover No. 1 (9) over the straight end of the exhaust air duct (10). Slide the exhaust air duct (10) from above through the small bushing of the roof penetration as far as the tube ranges downward acceptably. Slide protective cover No. 1 (9) down to the bushing of the roof penetration (1), so that rain may run off. Mount the weather and radiation shield (2) to the shield tube (3), so that the meteorological sensor will be located later on directly below the probe inlet. Now remount the air shield (6) on top of the shield tube (3), by feeding the hexagon socket head cap screws through the corresponding holes in the shield tube (3) and center the sample tube (4) by alternate screwing. The sample tube must stick up from the shield tube by 30 mm. This can be arranged easily, if the probe inlet (reff. to EN14907) is mounted in addition for a short time. If mounted correctly, the pipe socket of the inlet fits directly to the air shield (6). If necessary correct the position of the air shield (6) on the sample tube (4). Pitch the cable for the meteorological sensor with its short connector carefully into the cable duct (5), namely from the bend in direction of the straight end of the tube. Slide protective cover No. 2 (14) over the straight end of cable duct (5) Fig. 7: Double-walled sample system

215 Page 24 F Rev. 4 First insert the cable itself and than following the straight end of cable duct (5) to the last empty bushing of the roof penetration (1) as far as the roof structure requires, to make conduction of the cable to the rear panel of the F easily. Slide protective cover No. 2 (14) down to the bushing of the roof penetration (1), so that rain may run off. Mount the Hygroclip sensor to the cable connector (long). Insert it than into the weather and radiation shield and tighten it. Secure the cable with cable straps at the shield tube (3) and pull the remaining cable to the inner of the operating room. Secure the cable in the duct entrance with the also delivered plug (11). Connect the cable to the corresponding connector at the rear panel of the F labelled with Meteorology. Insert the PVC drain hose (15) into the L-shaped hose coupling at the bottom of shield tube (3) and lock it by little short pulling. Push the hose clamp (12) over the drain hose and close it. Mount the ventilation unit (13) with the clamp above the F in a way that the short hose at its bottom fits to the output of the shield tube bend and can be fixed by a hose clamp. The exhaust air hose needs to be mounted at the second bushing of the ventilator unit (13) with the corresponding hose clamp. Bring the other hose end to the exhaust air duct (10) and fix it there also with a hose clamp. If necessary, shorten the hose. Connect the power cord of the ventilation unit (13) to the corresponding connector at the rear panel of the F It is marked with External Fan. The fan is switched on and off together with the device. At least isolate the shield tube inside the operating room with the additional delivered isolation. Cut it to the necessary length and put it around the tube from the roof to the bend at the tube bottom, better right to the top cover of the device. Connect the device to power supply and switch it on. In the display appears the main menu shown in Figure Sensors Before starting with measurements, please check the parametrisation and function of connected sensors. If differences in measures to the true values (e.g. values from the station) exist, please use the possibilities in the Adjust menu to correct measures for ambient air pressure and ambient air temperature. How this can be done you will find out in chapter 7. If no sensors are connected please set the replacement values in the menu Parameter to the mean expected values for ambient air pressure and ambient air temperature (daily means). This is important to maintain an as accurate as possible volumetric flow of 1 m³/h. Changes in temperature and pressure will not be considered in this case Leakage test Next, please check the leak tightness of your measurement with the following procedure: On top of the sample tube must not be a probe inlet. Have a rubber plug or something near it ready to block the sample tube. Switch the device on. Choose in the menu Parameter the value Automatic for the Mode Start parameter. The device starts immediately with a measurement. Wait until the pump is working and the device shows a volumetric flow of 1000 l/h.

216 F Rev. 04 Page 25 Now block the sample tube with the plug. Monitor the volumetric flow output on the display. The value must be smaller than 10 l/h, ideally 0 l/h. If so, your system is tight. Remove the plug as fast as possible, otherwise the device will output an error message and switch off the pump due to the high underpressure. Wait until the device has tuned the volumetric flow back to 1000 l/h. This may take a few minutes. Complete the test by switching off the device. Mount the sample inlet on the sample tube. In case of a leakage check the connection between sample tube and device. The tube must fit straight to the gas inlet of the device, without bending or mechanical stress. How to get into the menu and to change parameters you will find out in chapter Volumetric flow In the factory the device is adjusted to keep a volumetric flow of 1000 l/h as function of ambient air pressure and temperature. For that air temperature and pressure need to be measured by the sensors mentioned above. Generally, an additional on-site calibration of the volumetric flow is not necessary and may lead to erroneous settings, which may adversely affect measurement accuracy. An on-site inspection can be done with a calibrated flowmeter as follows: For a calibration of volumetric flow sensors for the measurement of ambient air temperature and pressure must be installed. Switch the device off. Remove the sample inlet and mount the flowmeter instead. Switch the device on and set the parameter Mode Start in the menu Parameter to Manual or Automatic. The latter one leads immediately and after each power-on to the start of a measurement cycle and therefore to a switch-on of the pump. In the main menu check temperature and pressure readings. Wait in minimum for 10 minutes until both devices, the F and your calibrated flowmeter, have adapted to ambient conditions. The F should now show a volumetric flow of 1000 ±5 l/h. Compare the output of the F with the measure of your flowmeter. If the deviation exceeds 5 l/h, you may correct the volumetric flow by setting of the parameter Span volume in the menu Adjust. For that calculate the new span value under equation 4. SV new = SV old VF VF M F (EQ4) Recalculation of Span Volume With: SV new SV old VF M VF F new value for parameter Span volume - existing value for parameter Span volume - Volumetric flow measured by a calibrated Flowmeter - Volumetric flow shown by F

217 Page 26 F Rev Electrical connection The connections for the power supply, measurement and status signals and for external control and data recording equipment are located on the rear of the device. The figure below shows the standard output socket. Gerät ein ON IN0 IN1 Potentialfreie Kontakte Potential free Contacts Messwert liegt an Measuring Value present IN2 Remote Start Volumenstromfehler Volume Flow Error V Sammelstörung Fault Nullpunkt, Referenz od. Folie-Wert liegt an Zero Check, Reference or Tape Value present Filterriss Filter Crack Vakuumfehler Vacuum Error V 0 V Option Out V Messbereichsfehler Measuring Range Error f ma_out GNA ma_out GNA Messwert 2 Measuring Value 2 Masse Ground Messwert 1 Measuring Value 1 Masse Ground Fig. 8: F device interface / ma outputs / status signals ("Data )

218 F Rev. 04 Page 27 The 50-pin Sub-D connector is located on the rear of the device housing and is labelled "Data". The signals have the following specific meaning: Name Function Pin no. Device on after "Power on check" 01, 04 Measured value received Zero point, reference or foil value received Switching contact is closed 07, 10 Switching contact is closed 19, 22 Measuring range error Volumetric flow error Filter crack Vacuum break Combined fault If the values are above or below the set concentration measuring range Status notification if the measured volumetric flow leaves the range between 950 and 1050 litres/h for a continuous period of more than 30 seconds. If there is no filter paper between the beta source and the Geiger Mueller tube, the pulse rate increases. If the pulse rate rises above 138,000 pulses per minute, the device goes into standby mode. The filter paper must be replaced. Filter crack is automatically disabled if the pulse rate reaches < 100,000 pulses per minute. If the vacuum behind the filter is more than 0.4 bar during the extraction process, extraction is ended at the current dust spot and the dust quantity collected up to that point is determined. Depending on the measuring regime, dust collection may continue automatically at a new filter spot. "OR" operator, if one of the measuring range, volumetric flow, filter crack or vacuum break errors is present. 03, 49 06, 09 18, 22 24, 27 12, 15 Out 2 Voltage output V (option) 35, 38 Measured value 1 Current output ma for the dust concentration in µg/m³ (standard). The scaling of the measuring range can be set. Other output values are possible. 47, 50

219 Page 28 F Rev. 4 Measured value 2 Current output ma for the mass in µg (standard). The scaling of the measuring range can be set. Apart from the mass, the concentration, the ambient air temperature, the air pressure and other parameters can be output here. 41, 44 IN0 Unassigned 02, 14 IN1 Unassigned 05, 14 IN2 Input for "Start measuring cycle via contact". If this contact is assigned, this function must be set up on the device (siehe b). (Table 3) Signal description for the 50-pin Sub-D connector 08, 14 The following figure shows the pin assignment for the I/O socket for the serial RS232 interface. 1 6 Receive Data RXD Transmit Data TXD Signal Ground GND f Request to Send RTS Clear to Send CTS Fig. 9: Serial RS232 connection between PC and F ("RS232") An interface cable is used to connect a PC running the corresponding communication software. It is then possible to export and evaluate measured data, parameter settings and error status messages. As an option on the F , a PT100 can be connected to pins 1 and 6 of the 9-pin Sub-D connector signed with Temperature at the rear of the device for measurement of the ambient temperature at the sampling head. Temperature measurement is essential for controlling the flow of the internal vacuum pump depending on the ambient temperature. The flow rate at the sampling head is then 1 m³/h regardless of the ambient temperature. A further option is the sample inlet heater. This must be installed if the moisture concentration in the ambient air is very high and no active ventilated sample system is available. The temperature of the heating tape is measured and regulated by a PT100 at pins 2 and 7 of the 9-pin Sub-D connector.

220 F Rev. 04 Page 29 Both options are available with fully assembled connectors. For details of how to install the heating tape and the sensors for the ambient temperature and heating tape temperature, see figure 4. The figure below shows the pin assignment for the input connector. 9-pol. Sub-D Buchse 9 pin Sub-D socket PT 100 Messung der Umgebungsluft PT 100 Measurement of Ambient Air PT 100 Begleitheizung PT 100 Heater f Fig. 10: Interface for optional temperature inputs on F ("Temperature") Alternatively, a sensor for the measurement of temperature and relative humidity in ambient air can be mounted in a radiation shield close to the sample head. The connection cable of the sensor is plugged into the connector with the label Meteorology at the rear of the device. For the use of the sensor measurement values in control of volumetric flow or filter adapter temperature an appropriate parametrization is necessary (see ). If the double-walled probe system is in use, the fan cable must be plugged into the corresponding connector with the label External fan at the rear of the device.

221 Page 30 F Rev Operating the F Turning on After connecting to the mains supply, a start screen appears. Start mode and the screen display depend on the parameter settings. By factory default, the automatic measuring mode is activated and the measuring menu is displayed. Additional information Action Remaining time for actions Measuring result Current count rate of GM tube or Volumetric flow Buttons Fig. 11: Measuring menu after starting 7.2. Using the touchscreen The F has a 19" 7RU housing and has a touchscreen for controlling the device on the front door. The surface of the touchscreen is glare free and scratch resistant. It can be operated using fingers or a pen with slightly pressure on the surface. Different keys are available on the display depending on the menu displayed. Touching the keys on the screen executes the corresponding action. The different keys are described using their symbol in the subsequent section of the manual. The following keys are available at the bottom of the display in most menus: Switch between measuring, data display and setup mode. Go to a submenu Open maintenance menu

222 F Rev. 04 Page Basic functions In normal operation, press the Measuring mode: Data display mode: Setup mode: key to switch to the following menus: Dust measurement and display of results, display of activities, performance of service actions. Display the measured values in graphs or tables Display and edit the parameters Measuring mode The display in measuring mode corresponds to that shown in Fig. 10. The action currently in progress is shown in the top left corner. Possible actions are: Standby Wait for next cycle (see 'Start' parameter) n-th measurement Cycle number for multiple loading selection Measurement measuring cycle Additional information about the current action is shown on the line below, e.g. 0 rate, M rate or volume. (optional) Pressing the key displays additional information in measuring mode: Date Time GM count rate [ 1/min ] Volumetric flow converted to normal conditions [ Nl/h ] Temperature [ C ] Pressure [ hpa ] Relative humidity [ % ] If the pressure, temperature and relative humidity are recorded by sensors, the current measured values are displayed. Otherwise, the asterisk after the figure indicates that a default value has been used. All measured values are also saved in a database. If the start parameter for the measurement has been set to manual, the key is also available to manually start the measurement.

223 Page 32 F Rev Data display mode Measured value display The current measured values are shown on the display and saved in measuring mode. In data display mode, this data can be displayed in the form of a graph or table. Press the or ( oder ) keys to display one measured value at a time, or use the or ( oder ) keys to skip from page to page. More recent values are on the right-hand side (top), older values on the left (bottom). a) Selecting the data display The data for the last 9 months is saved in the memory. Therefore, the month from which you want to display information has to be selected first. This is done using the > cursor in the first column on the left, which can be moved up and down using the arrow keys. Pressing the and keys then displays the data as a table or a histogram, respectively. The key takes you back to the month selection. b) Graphical display as bar chart The scale of the bar chart depends on the setting for the FS conc parameter (see b). The following information is displayed below the diagram for the measured value selected with the or keys: Date Time at which the action/ measurement was started Measuring result Measured volume Actions other than the measurement are also indicated as follows below the relevant bar: N Zero point measurement R Reference point measurement F Reference foil measurement Pressing the histogram key repeatedly displays additional information in sequence.

224 F Rev. 04 Page 33 The first time the histogram key is pressed, the number of loads (SC, see b) is displayed and whether an error (E) has been registered for the selected action. Error code 1 Volume error Error code 2 Vacuum break Error code 0 No error Pressing the histogram key a second time shows the count rates at the beginning (0) and at the end (M) of the action and the absolute dust mass value in µg. Pressing the histogram key again displays information about the air pressure and temperature of the ambient air during the action. c) Table display In each line, the table contains all the key information for an action in the following order: Action code, as described in b) Date Time at which the action/ measurement was started Measuring result Measured volume Error messages (E) with error code 0 (no error), 1 (volume error) or 2 (vacuum break) Number of loads per filter position (SC)

225 Page 34 F Rev Displaying messages Pressing the list key again displays the saved messages in chronological order. In addition to the date and time, the occurence of an error is logged and when an error was resolved. The following messages are available: a) General notifications: Power On---- The device has been turned on Power Off--- The device has been turned off Change battery The battery on the main board needs to be replaced User stop Action has been cancelled by the user User shortcut Action has been shortened by the user Range error The current measured value is outside the set measuring range Change Parameter Parameter change SD card plugged SD card has been inserted again after it was removed during a running action Change Time/Date Date and time have been set Vacuum Break The vacuum behind the filter is too great b) Errors registered when they first occur and after resolution ("gone"/ well ): Adapter blocked The filter adapter could not be properly opened/ closed. Adapter Temp The filter adapter temperature does not correspond to the setting. Air Temp The temperature sensor for ambient air delivers no data. Heater Temp The heater temperature does not correspond to the setting. Filter Crack Insert new filter paper, the registered count rate is too high. GM fault The registered count rate is too low. Volume Error The volumetric flow is outside the specified limits Setup mode All parameters can be displayed in setup mode and, after entering a password, changed. Intermediate results can be displayed and service actions performed. Calling up setup mode does not interfere with the current measurement. You can use the and keys to scroll through the parameters. The selected menu or parameter is indicated by an arrow > to the left of the line. The key takes you into a submenu or a parameter entry. Submenus are indicated by the fact that no value is displayed on the right.

226 F Rev. 04 Page 35 Parameters can only be changed after entering a password. Three different passwords are available: Password1 Password2 Password3 Password 1 is For changing parameters For performing service actions For adjusting the electronics (system password) Access (the password entered) is automatically cancelled 30 minutes after the last password entry Changing parameters The / keys can be used to select a parameter and the key to open it. The parameter value is selected and entered using a selection list (a) or manually using the keypad (b). Some parameters are factory set and cannot be changed. a) Selection list If a selection list is activated, the current selection is indicated by an arrow > after the parameter. The / keys can be used to select the parameter value. When the desired value is displayed, it can be applied and saved by pressing the key. The selected parameter remains unchanged if you exit the data entry using the key. b) Keypad If the data is entered manually using the keypad, the selected parameter is highlighted and is simply overwritten. Once the value has been completely entered, exit the entry by pressing the CR key. The new value is saved. Figures can be deleted individually using the BS key. If all figures have been deleted, pressing the BS key again exits the data entry and the parameter remains unchanged. The values entered using the keypad are immediately checked for validity. Invalid values are not accepted and an error message is displayed along with the permitted data range. The original saved value remains unchanged. Pressing any key takes you back to the parameter display.

227 Page 36 F Rev Parameter menu The parameter menu (Fig. 9, see below) consists of the following submenus: Password Password entry Measurement Display the measured / saved values Parameter Display and enter the main parameters Sub parameter Display and enter the secondary parameters Adjust Correct the input and output signals Interface Set the interface parameters Date/Time Set the date and time Service Basic operating functions, troubleshooting F Password Measurement Parameter Sub Parameter Mass [µg] 0-Rate [1/min] M-Rate [1/min] Output 1 [ma] Output 2 [ma] Air temperat.[c] Air press. [hpa] rel humidity [%] Filter adapt.[c] Sampling tube[c] Cycle time Sample count Replacem. T [C] Replacem. [hpa] Replacem. rh [%] Mode start FS conc [µg/m3] Mode 20mA Out 1 Mode 20mA Out 2 Language FS reference[µg] Integration Mode temperature Filter adapt. [C] Samp. tube set[c] Mode zeroizer Mode Synch h:00 max. Mass [µg] Adjust Interface Date/Time Service Offset mass [µg] Span mass Span Volume Offset temp. [C] Offset p [hpa] Mode RS232 Gesytec No. Baud Parity RS484 inactive Hour Minute Day Month Year Password ADC Rowvalue Inputs Factory teach-in Options Sensors SW Serial no.device Serial no. CPU Q-Check Fig. 12: Parameter menu structure

228 F Rev. 04 Page 37 a) Measurement The following measuring results can be displayed: Mass [µg] 0 Rate [1/min] M Rate [1/min] Output 1 [ma] Output 2 [ma] Air temperat. [C] Air press. [hpa] Rel. humidity [%] Filter adapt. [C] Sampling tube[c] Mass from the last measuring cycle Last measurement 0 Rate Last measurement M Rate The current ma signal for output 1 is displayed The current ma signal for output 2 is displayed Ambient air temperature or default Ambient air pressure or default Relative humidity in ambient air or default Filter adapter temperature Sample tube heater temperature The measured values displayed cannot be changed. b) Parameter This menu is made up of frequently used parameters. Cycle time Defines the sampling time, 15 minutes to 24 hours. A changed cycle time takes effect when a new measuring cycle is started. Sample count Defines a multiple filter load, number of forwards/ backwards cycles. Possible values between 1 and 24. Example: If the sample value is reduced from 5 to 2 while a multiple sample is being taken for the third time, the next measurement is performed at a new filter spot. If the sample value is increased from 3 to 5, the fourth and fifth samples are taken at the same spot on the filter tape. Replacem. T [C] Default value for ambient air temperature if no external sensor is installed. A modified Replacement T value is used for the next standard volume calculation. The range is -20 C to +50 C. Replacem. [hpa] Default value for the air pressure used to calculate the standard volume if no sensor is installed. Mode Start Sets the conditions for starting the cycle: Manual - Start each measuring cycle manually using the start key Auto - Immediate automatic restart (continuous cycles) 60 Min - Start at the beginning of each complete hour 30 Min - Start at the beginning of each half hour 15 Min - Start at the beginning of each quarter hour Relais - Start by closing a contact on the data connector Gesytec - Start by Gesytec remote command Changing the start parameter takes effect when the next measuring cycle is started. FS conc [µg/m3] Mode 20mA Out1 Measuring range for 20 ma output of concentration in µg/m 3. The current output can be scaled for 50 µg/m 3 to 9999 µg/m 3. A change takes effect immediately. The entry also scales the display in the histogram (see section b). Output 1, the concentration by default, which uses the measuring range defined above, can be set to different values for testing purpos-

229 Page 38 F Rev. 4 Mode 20mA Out2 es, e.g. mass or 20mA, see also Output 2. It is reset to the concentration each time the device is turned on! Output 2 for various values. The selected value remains unchanged after turning on until the parameter is modified. The following settings are possible: conc - Concentration, measuring range as defined under FS conc [µg/m 3 ] mass - Mass, measuring range µg 20 ma - Constant 20 ma signal 4 ma - Constant 4 ma signal For testing purposes only: GM Tube - GM tube pulses [#/min] temper. - Air temperature (4/20mA = -20 C/+60 C) pressu. - Air pressure (4/20mA = 800hPa/1300hPa) volume - Volumetric flow (4/20mA = 0l/h / 1600l/h) humidity - Relative humidity (4/20mA = 0/ 100%) c) Sub parameter Language: FS reference[µg] English, German Measuring range for 20 ma output for the mass in µg for measurement with reference foil; used in the next reference cycle or immediately if a reference cycle is running. The range is between 50 µg and 9999 µg. The factory default setting is 1000 µg. Integration Number of values used for the sliding average. The factory setting is 1. Mode temperature Determines the type of temperature control for the filter adapter and sample probe heater: Filter adapt [C] absolute relative humidity - The temperature is kept within +/-0.5 degrees of the value specified in the "Filter adapt. " and "Samp. tube set" parameters. - The temperature is regulated using an offset specified in the "Filter adapt." and "Samp. tube set" parameters to the measured air / ambient temperature or to the default value. - The temperature is regulated using an offset specified in the "Filter adapt." and "Samp. tube set" parameters to the dew point temperature. The dew point temperature is determined from the measured values or the default values for temperature and relative humidity in ambient air. The minimum / maximum temperature for the heater and the filter adapter is independent of the air temperature of 20 C or 50 C, respectively. Filter adapter temperature in C, set depending on the "Mode temperature" parameter. Takes effect immediately. The factory setting is 50 C. Samp. tube set[c] Temperature of the (optional) sample probe heater in C, set depending on the "Mode temperature" parameter. Takes effect immediately. The factory setting is 50 C. Mode zeroizer Mode synchr h:00 Max. mass [µg] If set to "active" negative measuring results are output as zero. h may vary between 0 and 23. It defines the hour of the day at which the measuring cycle is synchronised. A running measurement cycle will be terminated before h.00 o clock. An input of -1 deactivates the synchronisation. Defines the maximum absolute mass on a filter spot. If this mass limit could be exceeded during the next sample cycle on the same spot, a

230 F Rev. 04 Page 39 new filter spot will be initiated and the clean filter is surveyed before the next sample cycle starts. d) Adjust Offset mass [ug] Span mass Span volume Offset temp. [C] Offset p [ hpa] Used to correct the mass determined by the device with the equation y = x* Span + Offset, where x = Measured value from F The factory setting is 0. A modified value is used the next time the mass is calculated. The permitted range is +/-500 ug. Used to correct the mass determined by the device with the equation y = x* Span + Offset, where x = Measured value from F The factory setting is 1. A modified value is used the next time the mass is calculated. The permitted range is 0.1 to 10. The volume signal is multiplied by the specified correction factor and takes effect immediately. The permitted range is The factory setting is 1. Offset value for correcting the air temperature in C. The factory default setting is 0 C. The permitted range is -10 C to 10 C. Offset value for correcting the air pressure in hpa. The factory setting is 0 hpa. The permitted range is +/-200 hpa. e) Interface Mode RS232 Determines the communication method on the RS232 interface. terminal - Communication with the PC is via a terminal program, e.g. HyperTerminal. printer - A serial printer has been connected. Alternatively, selecting this parameter enables all data for the measuring result to be automatically transferred at the end of a measurement when communicating with a PC using a terminal program. print V2 - Output compatible with software version 2.0 Gesytec - The connected PC communicates with the device using the Gesytec protocol. Gesytec No. Range between 0 and 999. The factory setting is 123. Baud Sets the baud rate between 1200bd and 19200bd. The factory default setting is 9600 bd. Parity/Bit Sets the parity and data bits, a stop bit is not used. There is no flow control. The following settings are possible: No 7 / Even 7 / Odd 7 No 8 / Even 8 / Odd 8 The factory default setting is 'No 8'. f) Date/Time Hour Defines the hour Minute Defines the minute Day Defines the day All inputs become active immediately. Month Defines the month Year Defines the year (input as two-digit date) h) Service Password A special password is required to change the calibration factors and factory default parameter settings.

231 Page 40 F Rev. 4 Attention! Changes may only be made by authorised personnel. If parameter values different to those specified here are displayed, this leads to malfunctions and can damage the device. This is particularly applicable for the parameters marked in red. If there are any differences or queries, please contact Durag GmbH. ADC Raw Value Inputs Factory teach-in Displays the current values from the analogue-digital converter (ADC), range , values >16081 indicate an overflow in the A/D converter. Filter adapter Tube heater Temperature Volume Pressure Humidity - Filter adapter temperature - Tube heater temperature - Ambient air temperature - Volumetric flow - Air pressure - relative humidity Displays the current values for digital inputs. 0 = Open 1 = Closed Filter ad. open - Limit position switch: Filter adapter open Filter ad. close - Limit position switch: Filter adapter closed GM-pulses static - Open by default Underpressure - Vacuum switch, normally open, closes with vacuum in gas path Input 1 - Not used, no function Input 2 - Not used, no function Input 3 - Start measuring cycle remotely Position refer. - Not used, no function The electronics are not adjusted using potentiometers but using parameters for adjusting the inputs and outputs. Offs sc1 [µg] - Offset for calculating the mass, from calibration, at multiple filter load only for the first measurement. Offs sc2-n [µg] - Offset for calculating the mass, from calibration, at multiple filter load from the second measurement. Span Service - Span for calculating the mass, from calibration Offs Fi Ad[0.1C] - Enables the filter adapter temperature to be adjused Filter adapt Filter adapter temperature value from ADC prevously imported via a 100 Ohm resistor (corresponds to 0 C) Filter adapt Filter adapter temperature value from ADC prevously imported via a 120 Ohm resistor (corresponds to 52 C) Tube heater Tube heater temperature value from ADC previously imported via a 100 Ohm resistor (corresponds to 0 C) Tube heater Tube heater temperature value from ADC previously imported via a 120 Ohm resistor (corresponds to 52 C) Temperature Ambient air temperature value from ADC previously imported via a 100 Ohm resistor (corresponds to 0 C) Temperature Ambient air temperature value from ADC previously imported via a 120 Ohm resistor ( corresponds to 52 C) Volumesensor 1V- Volumetric flow value from ADC previously imported using a 1 V signal

232 F Rev. 04 Page 41 Factory teach-in (continued) Volumesensor 5V- Volumetric flow value from ADC previously imported using a 5 V signal Pressure 4 ma - Air pressure value from ADC previously imported using a 4 ma signal Pressure 20 ma - Air pressure value from ADC previously imported using a 20 ma signal Reserve 4 ma - Not used, no function Reserve 20 ma - Not used, no function b 20 ma Out1 - Factor for setting analogue output 1 to exactly 20 ma: Exactly 20 ma = Original signal * b+c c 20 ma Out1 - Offset for setting analogue output 1 to exactly 20 ma: Exactly 20 ma = Original signal * b+c b 20 ma Out2 - Factor for setting analogue output 2 to exactly 20 ma: Exactly 20 ma = Original signal * b+c c 20 ma Out2 - Offset for setting analogue output 2 to exactly 20 ma: Exactly 20 ma = Original signal * b+c Options Message - Set to 10 to suppress messages shorter than one debounce minute (default). Set to 1 to receive all messages (testing purposes only). Beta sensor - GM tube Device type - F Filter motor - micro Steps - Distance of the GM tube to the collection position in sens./tube steps of the stepper motor. The factory default setting is 36 mm = 1600 steps. Should not be modified. Pitch dust spot - Distance between filter spots. The factory default setting is 36 mm = 1600 steps. A modified value takes effect when a new spot is moved to the measurement area. Should not be modified. Intell. corr. - Active 1 = A so-called intelligent data correction algorithm is used for the calculation of mass. No = Standard algorithm according to equation 2. ICC value - Percentage R0 and RM may differ in maximum for the use of intelligent correction algorithm (Default = 0.39). Vol. GM source - Air volume in the measurement volume between C14 source and GM tube in mm³. By setting it to 0 the change of air mass in the measurement volume Mode filterprint by air pressure changes will not be considered. - Inactive = No filter tape printer (default) - Active = Device uses a filter tape printer with the "Constituent analysis" option. Sensors Sensor air T - Replace = The default value is used for the air temperature. - PT100 = The external air temperature is measured using a PT meteor. = Ambient air temperature is measured by a sensor which also determines relative humidity. Air temperat. [C] - Displays the current (uncorrected) air temperature or default value.

233 Page 42 F Rev. 4 Sensors Sensor air p - Replace = The default value is used for the air (continued) pressure. - 4/20mA = Absolute pressure sensor (optional). - I2C = Not used, for testing purposes only. Air press. [hpa] - Displays the current (uncorrected) air pressure in hpa or default value. Pressure b - Factor for pressure calculation, where Pressure [hpa]= Pressure b *ma+ Pressure c (Default = 31.25). Pressure c - Offset for pressure calculation, where Pressure [hpa]= Pressure b *ma+ Pressure c (Default = 675). Sensor air rh - Replace = The default value for the relative humidity in ambient air is used. - meteor. = Ambient air relative humidity is measured by a sensor which also determines air temperature. Rel humidity [%] - Displays the current relative humidity in % or default value. SW Serial no. device Serial no. CPU Q-Check For testing purposes only: Volume [mv] - Raw signal from volumetric flow sensor in mv Pressure [ma] - Raw signal from air pressure sensor in ma Humidity [ma] - Raw signal from humidity sensor in ma Software version F serial number Controller serial number Manufacturer's identification to identify the person who tested.

234 F Rev. 04 Page Maintenance menu / mode Pressing the key activates the maintenance menu once the corresponding password has been entered. The filter tape can be moved, the filter adapter opened and closed and functional tests performed. Open filter housing Close filter housing Filter tape forwards Filter tape backwards Start reference measurement Start zero measurement Start reference foil measurement In manual mode, also: Start measurement The desired action does not always start immediately. Probably the actual running action must be completed first, before a new command can be executed. While the current action is being processed, the other keys may be inactive. Once an action in maintenance mode has started, it can be cancelled using the key. Pressing the key reduces the current action to a maximum of 10 seconds and, for a reference foil measurement, confirms the insertion or removal of the foil for the measurement. The reference measurement is a device test that involves measuring a 0 and M rate without transporting the filter. The M rate is assessed only a sixth weaker. This corresponds to a mass of 444 µg if one takes into consideration parameters Offset mass [µg] = Offset sc1 [µg] = Offset sc2-n [µg] = 0 and Span mass = Span service = 1. This measured value should be achieved with a tolerance of +/- 20 µg and is a measure of the stability of the 0 and M rate measurement, making it a test function for testing the source and the GM tube. A current of approx ma is generated at the corresponding measured value output on the data connector (with FS reference parameter = 1000 µg). If the offset parameters differ from 0 and the span parameters differ from 1, this must be taken into account accordingly. The zero measurement provides another device test that can be used to test the function of the source and the GM tube. There is no filter transport. The result should be an absolute measured value 10 µg. The reference foil measurement is a simple, interactive device test for assessing the device stability.

235 Page 44 F Rev. 4 The following work steps should be carried out as shown in the figures below: Move to a new (clean) filter position by pressing the "Filter tape forwards" key. Start reference foil measurement. The 0 rate is measured. Insert foil from right Continue the measurement with. The M rate is measured. Remove foil, continue with. The result of the measurement is displayed and should correspond to the measuring range of the relevant reference foil. The figures show the steps in the reference foil cycle. Comment 1: Up to software version 1.x, a characteristic curve was saved for a reference measurement of 630 µg and for a reference foil measurement between 600 µg and 800 µg. Comment 2: The set points specified above for the reference and reference foil measurement require the parameters for "Offset mass [µg]" and "Span mass" from the "Adjust" menu, as well as the "Offset sc1 [µg]", Offset sc2-n [µg] and "Span Service" parameters from the "Service/Factory teach-in" menu.

236 F Rev. 04 Page Data memory The device saves the measured concentration value with the associated date and time and other information, e.g. volumetric flow, filter adapter temperature or ambient air pressure, temperature and relative humidity, when measured. The memory capacity is 9 months. The data are not lost after switching off the device. The content of the data memory can be viewed on the device display or exported using the RS232 interface on the rear of the device Viewing the data memory on the display Viewing on the device display is outlined in section of this manual Exporting the data using the RS232 interface The output via the RS232 interface can be done via a terminal program. To do this, the settings for the RS232 interface configuration must correspond to those described in e. For easier subsequent processing, communication using a terminal program such as "Hyper Terminal" is recommended. The following key entries can be used to retrieve data from the data logger: P<CR> M<m y><cr> E<m y><cr> - Parameters - Measurements from month (m) and year (y) - Errors and messages from month (m) and year (y) If the F receives an unknown command, it sends a short help text. The command chain must always be completed with a <CR>. Below are some examples for outputting possible lists: Help text P - Parameters M m y - Measurement m-month, y-year E m y - Error / Message M?/E? - Data M/ E available?

237 Page 46 F Rev. 4 Measured values Measurement( ): T Date Time ug/m3 Litre E SC ug 0-Rate M-Rate FiH C C hpa % IC N : N : M : M : M : M : M : M : M : M : F : Explanations: The heading contains the serial number of the device in brackets. In T column: M = Measured value, R = Reference measurement, N = Zero measurement, F = Reference foil measurement Date and Time: Start time of measuring cycle µg/m3: Measured dust concentration per cycle Litre: Gas volume extracted per cycle E: Error messages during operating cycles 0 = no error 1 = volume error 2 = vacuum break SC: With multiple loads, number of loads per filter position, otherwise zero µg: Measured dust mass per cycle 0-Rate: Count rate for unloaded filter position per minute M-Rate: Count rate for filter position after collection cycle per minute FiH C: Filter adapter temperature in C C: Ambient air temperature in C or default value hpa: Ambient pressure in hpa or default value %: Relative humidity in % or default value IC: Check value, only for internal use Error and operation status messages Error/Message( ): : Power On---- Ser-No: :21 User shortcut :26 Vacuum Break :28 User shortcut : Power Off : Power On---- Ser-No: :48 Adapter Temp flt :49 Adapter Temp well :50 User shortcut :29 Change time/date :08 Change parameter :43 User stop :52 Change parameter :53 Change parameter A relevant event is entered in the message database. Error messages (see also section and 11) include the occurrence of the error (flt) and until when the error condition persisted ("gone"/ well ). Additional information such as turning on the device or cancellation of the measuring cycle by an operator is also recorded.

238 F Rev. 04 Page Measured value output / Time responses / Diagram display measuring value read out status signals device ON measuring value is on (display, ma, Gesytec) O-ref. or film value is on measurement 0-rate 0 0-rate sampling measurement M-rate measurement 0-rate sampling measurement M-rate measurement 0-rate S M 0 S M 0 S filter tape spot 1 filter tape spot 2 filter tape spot 3 measurement measurement measurement sampling M-rate 0-rate sampling M-rate 0-rate sampling 0 1. measured value 2. measured value 0 f7 Fig. 13: Time lapse for single filter spot load

239 Page 48 F Rev. 4 display measuring value read out status signals device ON measuring value is on (display, ma, Gesytec) O-ref. or reference value is on measurement 0-rate 0 sampling measurement M-rate sampling measurement M-rate sampling measurement M-rate measurement 0-rate S M S M S M 0 S sampling filter tape spot 1 filter tape spot 2 1st measurement 2nd measurement 3rd measurement 1st measurement sampling M-rate sampling M-rate sampling M-rate 0-rate sampling 0 1. measured value 2. measured value 3. measured value f Fig. 14: Time lapse for multiple filter spot load

240 F Rev. 04 Page 49 display measuring value read out status signals device ON measuring value is on (display, ma, Gesytec) O-ref. or reference value is on measurement 0-rate 0 sampling measurement M-rate measurement 0-rate measurement M-rate measurement 0-rate sampling measurement M-rate measurement 0-rate sampling S M 0 M 0 S M 0 S filter tape spot 1 filter tape spot 2 filter tape spot 3 measurement zero point control measurement measurement 0-rate sampling M-rate 0-rate M-rate 0-rate sampling M-rate 0-rate sampling 0 1. measured value zero point control measuring value 2. measured value 0 f7 Fig. 15: Time lapse for "zero point check"

241 Page 50 F Rev The GESYTEC protocol for transferring numerical data 8.1. Basic requirements Communication is carried out using the RS232 interface on the rear panel of the device. The pin assignment can be seen in Fig. 8. The Gesytec protocol is used (also known as the Bavaria/Hessen protocol). This enables the current measured value with any errors and status messages to be retrieved. As actions, a measuring cycle, zero point or reference check can be started or a measurement in progress can be cancelled Setting the F parameters The corresponding parameters for the Gesytec protocol are as follows (see also ): Menu option Function Parameter/Start Activate Gesytec A measuring cycle is started using a Gesytec start command. Interface/ Mode RS232 Interface/ Gesytec No. Interface/Baud Interface / Parity/Bit Activate Gesytec nnn Define baud rate for interface Make the setting for parity and data bits Set the parameter to activate Gesytec communication Measuring instrument address Possible settings: 1200 Baud to Baud Possible settings: No 7, Even 7, Odd 7, No 8, Even 8, Odd 8 (Table 4) Control parameters for communication using Gesytec protocol 8.3. Protocol types and format details Example transmission: The data query <STX> "DA" "123" <ETX> <BCC1> <BCC2> is output by bytes on the serial interface (PC) as follows: 0x03 0x44 0x41 0x31 0x32 0x33 0x04 0x32 0x33 Note: "0x" indicates a hexadecimal display of the value

242 F Rev. 04 Page 51 Data query with a PC (master) from F (slave) Response does not depend on the measuring instrument address ("Gesystec No.") Field No. Start byte End byte Data format Field description <STX> Start Text DA Protocol ID <ETX> End Text <BCC1> Higher value block check character (BCC) <BCC2> Lower value block check character (BCC) Data query with a PC (master) from F (slave) Response depends on measuring instrument address. (Parameter: "Gesystec No.") Field No. Start byte End byte Data format Field description <STX> Start Text DA Protocol ID nnn Measuring device ID (device address, number between 0001 and 0255), programmable on device <ETX> End Text <BCC1> Higher value block check character (BCC) <BCC2> Lower value block check character (BCC) Data transmission from F (slave) to PC (master) Field No. Start byte End byte Data format Field description <STX> Start Text MD Protocol ID nn# Number of instruments in measuring system nnn Measuring device ID (device address, number between 0001 and 0255), programmable on device ±nnnn±ee# Concentration; "nnnn" should be interpreted in µg with a point after the first figure. Example " " = 0.047*10³ equates to 47 µg/m hh# Operating status (see table below) hh# Error status (see table below) hhh#hhhhhh# Instrument type no., rest not used <ETX> End Text <BCC1> Higher value block check character (BCC) <BCC2> Lower value block check character (BCC)

243 Page 52 F Rev. 4 Assignment for operating and error status messages Operating status (Field NO. 6) Bit 0 Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 Bit 7 Meaning 1 - Standby, 0 - Measurement, zero point, reference or film measurement Foil measurement Zero point Reference measurement (reference check) Measurement Operating status Bit 0 indicates the current status - Standby/Measurement. The operating status bits 1, 2, 3, 7 indicate which type of measurement the measured value is to be assigned to, i.e. a concentration measurement or a mass in a test cycle. Error status (Field NO. 7) Bit 0 Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 Bit 7 Description Volume error Vacuum break Volume < 500 litres or 250 litres with 1/2 hour extraction time Replace battery Filter crack The error status message bits 0, 1, 2 indicate that this error occurred in the concentration measurement transmitted via Gesytec. The error status message bits 5, 6 are currently present and have no reference to the concentration in this transmission. Undefined bits contain "0".

244 F Rev. 04 Page 53 Command transmission from a PC (master) to F (slave) Field No. Start byte End byte Data format Index description <STX> Start Text ST Protocol ID nnn# Measuring device ID (device address, number between 0001 and 0255), programmable on device A Control code for transmission check command to start the measuring cycle, one letter (mnemonic, see below) <ETX> End Text <BBC1> Higher value block check character (BCC) <BBC2> Lower value block check character (BCC) Transmission check (A =) M Measurement K Reference measurement N Zero point S Stop measuring cycle The ST command is only executed in standby mode. To record standby mode, see the operating status of the data (requested using DA command). To stop a measuring cycle, use the transmission check command "S" Getting started A function test on the Gesytec protocol can easily be performed using a terminal program such as "Hyper Terminal". In this case, you can replace the last 3 characters (<ETX>, <BCC1> and <BCC2>) with a <CR>. First try the simplest query: <STX>DA<CR>. You get <STX> by simultaneously pressing the <Ctrl> and <b> keys.

245 Page 54 F Rev Examples Data query Master Slave (protocol ID "DA") Field Format <STX> DA <ETX> <BCC1> <BCC2> Query by master from slave Slave Master (protocol ID "DA") Field Format <STX> MD nn# nnn# ±nnnn±ee# hh# hh# hhh# <ETX> <BCC1> <BCC2> Response from slave Device no. 01 One device Address 070 Number Concentr µg/m 3 or 57µg Function 00 Measurement (dust, zero point, reference or foil) Function 01 Standby Function 02/ 03 Function 04 /05 Function 08 /09 Function 80 /81 Foil measurement/ Result+ Standby Zero point/ Result+ Standby Reference/ Result+ Standby Dust concentr./ Result+ Standby Error 01 Volume error Error 02 Vacuum break Error 04 Volume < 500 l Error 20 Battery Error 40 Filter crack Device type 701 F-701 Data transmission Master Slave (protocol ID "ST") Field Format <STX> ST nnn# A <ETX> <BBC1> <BBC2> Transmission from master Device no. F Control character Control character Control character Control character M K N S Measurement Reference measurement Zero point Cancel measurement, forces F to standby

246 F Rev. 04 Page Measuring error analysis / Setting cycle times The measuring accuracy of the F dust measuring device is ±10 µg absolute dust mass on a filter spot. The measuring accuracy of the concentration measurement depends on the extracted gas volume. To achieve a high resolution with low dust concentrations, extracting a large volume is recommended. This is done by extending the extraction time or the programmed cycle time. The cycle time should be set as follows: 1. The expected normal dust concentration must be estimated. For example, a dust concentration of 100 µg/m³ can be used. 2. An optimum operating point for the F dust measuring device is at 300 µg absolute mass. Dividing 300 µg absolute mass by a dust concentration of 100 µg/m³ gives a volume of 3m³. As the device extracts 1m³ per hour, the cycle time should be 3 hours. 3. An optimum extraction time can also be covered by multiple loading of a filter spot. For example, the extraction time of 3 hours could be achieved by three loads with cycle times of 1 hour. The advantage of this is that intermediate results can also be output, which result in the same accuracy when averaging over the entire period. This results in an increase in the dynamic range. 4. The measuring accuracy of the dust concentration measurement is given by the following calculation: c = ± 10µg / Q where: c Q (EQ5) Determination of absolute measuring error Measuring inaccuracy of dust concentration measurement Total volume extracted

247 Page 56 F Rev Maintenance The device should be checked and serviced at regular intervals, particularly the following parts: Filter adapter, transport reel, pump, hoses and waste air filter, and the sample inlet. The filter tape should be replaced if required. If the constituent analysis option is installed, check the ink ribbon cartridge on the printer and the cover foil and replace them if necessary. All work may only be carried out by appropriately trained and authorised personnel. The following table provides a brief overview. Period Part Action As required Filter tape Replace Filter tape printer ink ribbon Replace cartridge Cover foil Replace At each check Filter adapter temperature Check in display Volumetric flow Check in display Count rates Check in display Sensors for T, p, rh (if exist) Check in display Error data base Chek for error messages 4 weeks, consider local dust concentration 6 weeks or after approx hours operation Sample inlet in line with EN12341 and EN14907 Vacuum pump VTE3 Clean nozzles and impaction plate Grease impaction plate Initial maintenance of the vacuum pump is normally only required after a year. Blow out filter Check blade height and replace if necessary 6 months Filter adapter Clean upper and lower section Transport reel Clean pressure rollers Clean Waste air filter Check and replace if necessary Hoses Check and blow out if necessary 1 year Vacuum pump VTE3 Replace filter and seals Meteorology sensor (if exist) Send to Durag GmbH for recalibration (Table 5) Maintenance overview

248 F Rev. 04 Page Device maintenance The following items should be checked in the display: Variable: Setpoint: See menu: Filter adapter temperature See specified setpoint Parameters menu / Measured values Volumetric flow 1000 l/h ±50 l Measuring mode Count rate Temperature, air pressure, relative humidity Error-/ message data base Between approx pulses/ minute If sensors are connected, the actual measures are displayed (without asterisk!) Must not include error messages which have not been resolved or which are monitored several times. (Table 6) Maintenance: Display check Measuring mode, (additional information) Measuring mode, (additional information) Displaying messages, if necessary, please contact service department of Durag GmbH Attention! Before starting maintenance work which require direct intervention to the device, switch off the device and disconnect from power. The filter adapter should be cleaned at least every 6 months. This is done by removing the upper section of the filter housing. First, check whether the cover foil of the GM tube and the source are still undamaged. Any damage is a malfunction of the device and the damaged part must be replaced. The interior is cleaned carefully! with a dry cloth or with particle free compressed air (spray can). If not all glass fiber disposals can be removed this way, use some alcohol in addition. The transmission piece to the sample tube must also be cleaned with particle free compressed air (spray can). Attention: Do not damage the cover foil of the counting tube!

249 Page 58 F Rev. 4 Fig. 16: Upper section of filter housing (from below) The beta source in the lower section of the filter adapter should be cleaned carefully! with a dry cloth or with particle free compressed air (spray can). Also blow out the opening of the mash. To do this, detach the intake hose at the bottom first. After cleaning, replace the intake hose and secure with the clamp. Again, if glass fiber disposals can t be removed completely, use a cloth drenched in alcohol. Attention: Do not damage the cover foil of the source! Fig. 17: Lower section of filter housing (top view) After reassembling of the filter adapter check the leak tightness of the sample pass. Connect the device to power and switch it on. Start a single measurement. Block the intake of the transmission piece and watch whether the flow rate lowers to a value close to 0 l/h.

250 F Rev. 04 Page 59 Check the transport reel and the pressure rollers for soiling, particularly at the edges, and clean if necessary, e.g. with a soft brush. Fig. 18: Transport reel The vacuum pump should be maintained regularly after the first year of operation. The seal, filter and carbon blades should be replaced in line with the maintenance instructions for the pump. Depending on the ambient conditions, the carbon blades should be checked around every 1000 hours of operation (corresponds to continuous operation of around 6 weeks). The intervals can be adjusted depending on the wear identified. For further details, refer to section Check the internal hoses and the waste air filter for soiling at least every six months and clean or replace them as required Sample inlet maintenance Depending on the type, the sample inlet should be maintained, cleaned and lubricated regularly in line with EN12341 (Appendix B) or US National Ambient Air Quality Standard (NAAQS) 40 CFR, Part 50 (Appendix L). For example EN stipulate cleaning and greasing of the impact plate of the sample inlet after 20 samples (= 20 days). Under consideration of the local dust concentration a maintenance interval of 4 weeks is recommended.

251 Page 60 F Rev Instructions for replacing a filter reel Open the supply reel: Remove the front section of the supply reel by turning it, while holding the rear section in place. Open the filter housing by pressing the key in the maintenance menu.

252 F Rev. 04 Page 61 Opening the take-up reel: Remove the front section of the take-up reel by turning it, while holding the rear section in place. Turn the pressure roller to the left and engage it at the location provided. The remaining section of the tape can then be removed. Check the transport reel and clean it if necessarily. Retain the empty cardboard reel and re-use it for the take-up reel.

253 Page 62 F Rev. 4 Position the new filter reel in the direction shown. Slide the tape through the open filter housing.

254 F Rev. 04 Page 63 Run the tape along the transport reel. Mind to avoid rubbing the filter tape against the transport reel, because the fine glass fibres could soil its rough surface. Secure the tape on the cardboard sleeve with adhesive tape.

255 Page 64 F Rev. 4 Pull the filter tape somewhat more in the direction of the take-up reel without rubbing the tape against the rough surface of the transport reel. Turn the take-up reel 2 to 3 times to the left to further secure the tape.

256 F Rev. 04 Page 65 Turn the pressure roller back to the right and engage it at the location provided, so that filter tape is pressed onto the transport reel. Screw the front section of the take-up reel back on and secure it by turning it while holding the rear section in place.

257 Page 66 F Rev. 4 Screw the front section of the supply reel back on and secure it by turning it while holding the rear section in place. Close the filter housing by pressing the key in the maintenance menu.

258 F Rev. 04 Page Error messages / Troubleshooting Errors that occur during operation are shown by the device in the display and at the 50-pin data output, and are also saved in the database. The following errors are monitored: Error type Cause/Identification Measures Volume error error code 1 Filter crack Measuring range error Vacuum break error code 2 Group fault error code 3 Volumetric flow outside the range 950 to 1050 litres per hour, the error must be present continuously for at least 30 seconds. If there is no filter paper between the source and the GM tube (e.g. at the end of the reel or if the filter paper is torn), the radiation intensity is no longer attenuated by the filter paper. As a result, the count rate rises above 138,000 pulses per minute. The measured dust concentration is above or below the programmed measuring range. The vacuum switch is indicating a vacuum of more than -0.4 bar relative in the intake pipe after the filter. One or more of the specified errors have occurred. 1. Check pump function. 2. Check seals on pipes. 3. Check whether pipes are clogged. 4. Check whether filter adapter closes correctly. 5. Check function of flow control: Block intake pipe and observe whether the display changes. 6. Check function of control valve: Remove cover from control valve, block intake pipe and observe whether the lever in the control valve changes its position. Insert new filter paper in the device. Start measurements. Note: The end of the filter tape reel is marked by a hole in the filter approx. 1m before the end of the filter tape. Open the filter adapter and check whether this marking is currently in the filter adapter. The reel must then be completely replaced (see section 10.3). If necessary, program higher measuring range limits. (Parameters / FS conc µg/m3, see section ). This normally only occurs with very heavily loaded filter spots. The measured absolute mass on the last filter spot should be checked, and is then normally over 600 µg. The cycle time and/or the number of loads may need to be reduced. (Table 7) Error messages / Troubleshooting

259 Page 68 F Rev Calculating the filter tape and cover foil consumption t Reel [h]= t Cycle [h] x SC x 45m/0.036m t Reel Time in which the reel is consumed t Cycle Cycle time parameter SC Sample counts parameter 45m Length of tape 0.036m Collection spot spacing Example: t Cycle = 1h SC = 3 t Reel = 1h x 3 * 45m/0.036m = 3750h, i.e. approx. 5 months A 30m will need to be replaced earlier.

260 F Rev. 04 Page Technical data Concentration measuring ranges Detection limit Zero point temperature dependency Sensitivity temperature dependency Change of zero point over time Between µg/m³ and 0-10 mg/m³, selectable <1 µg/m³ Availability >95 % Assembly and commissioning Running in time <0.8 µg/m³ at ΔT of 15 C (27 F) in line with suitability test report July 2006 <3.3% of 40 µg/m³ at ΔT of 15 C (27 F) in line with suitability test report July 2006 Automatic zero point correction Approx. 2-3 hours, depending on preparation Approx. 1 hour Filter tape replacement interval Cycle time * Multiple load number * Length / 0.036m Maintenance interval (except filter tape, sample inlet, volume pump) Volume pump maintenance interval e.g. 0.5h*1*45m/0.036m=625h=26 days or 3h*2*45m/0.036m=7500h=312 days Approx. 1/2 year After the first year of operation 6 weeks Sample inlet maintenance interval 4 weeks (in line with EN / EN 14907) Power supply 230 V / 50 Hz +10% / -15% or 115 V / 60 Hz +10% / -15% See rating plate Power consumption Approx. 0.4 kva Internal fuse 230V 4 A (inert) 115V 6.3 A (inert) Permitted ambient temperature for device 0 C to +40 C (32 F to 104 F), non-condensing (Type-approved: +5 C to +40 C (41 F to 104 F)) Permitted sample gas / measuring head temperature Measured value output / interfaces Measurement database Message / error database Operation Measurement cycle Multiple sample spot loading Source -20 C to +50 C (-4 F to 122 F) 2 x 4-20 ma, 1 x RS232, Gesytec protocol (Bavaria/Hesse) 9 months 9 months Touchscreen User selectable, between 15 minutes and 24 hours Max. 24 cycles per filter spot C14 surface emitting source, enclosed, half life = 5,730 years, total radioactivity <450 kbq (<12.5 µci)

261 Page 70 F Rev. 4 Detector Filter material Filter area Geiger Müller end window count tube Glass fibre filter (free of heavy metals), retention = 99.95% at 0.3µm particle diameter 0.79cm² (0.12 in²) Filter feed 36 mm (1.4 ) Filter tape lengths Filter housing temperature Sample gas volumetric flow Sample volumetric flow - correction temperature Sample volumetric flow - correction absolute pressure Pump Dimensions (H x W x D) Weight Colour Protection type IP m, 45 m (98 ft, 148 ft) 20 C to 50 C (68 F to 122 F), regulated as absolute value or with offset relative to ambient temperature or to dew point temperature of ambient air 1000 l/h, max. variation ±2%, regulated Standard: Temperature entry by operator. Option: Automatic correction by measurement of temperature using temperature sensor, recommended for PM2.5 measurements. Standard: Absolute pressure entry by operator. Option: Automatic correction by measurement of pressure using absolute pressure sensor, recommended for PM2.5 measurements. Integrated rotary vane vacuum pump, 1m³/h 320 x 482 x 530 mm (12.6 x 19 x20.9 ) 19 Insert/ table-top unit 31 kg (68 lbs) RAL 7032 (grey) Sampling system TSP/ overall dust (VDI 2463, page 8); PT100 sensor (option) Meteorological sensor for temperature and relative humidity (option) Air pressure sensor (option) Probe heater (option) PM10 (EN 12341/ NAAQS 40 CFR, Part 50); PM2.5 (EN14907/ NAAQS 40 CFR, Part 50); Sample tube (single or double tube with active ventilation) Accuracy: T = ±0.5 K Accuracy: T = ±0.1 K rh = ±0.8 % (calibrated) Accuracy: p = typ. 2.5 hpa, max. 5 hpa Heating tape with temperature sensor (Table 8) Technical data

262 F Rev. 04 Page Service documents Parameter list Standard parameters (from software version 3.07): Parameter print : Master Data: Beta Particulate Monitor F Serial no.device Serial no. CPU Silicon Serial Number C C-DF-A2 SW Parameter: Cycle time 1 h Sample count 3 Replacem. T [C] 20 Replacem. p[hpa] 1013 Replacem. rh [%] 50 Mode start 60 Min FS conc [ug/m3] 200 Mode 20mA out 1 conc Mode 20mA out 2 mass Language English FS reference[ug] 1000 Integration 1 Mode temperature humidity Filter adapt.[c] 3 Samp.tube set[c] 3 Mode zeroizer inactive Mode synchr h:00-1 Max. mass [ug] 400 Offset mass [ug] 0 Span mass 1 Span volume 1.01 Offset temp. [C] 0 Offset p [hpa] -5 Mode RS232 terminal Gesytec No. 123 Baud 9600bd Parity no 8 RS485 inactive Factory adjustment: Offset sc1 [ug] 0 Offset sc2-n[ug] 0 Span Service Offs Fi-Ad[0.1C] 0 Filter adapt Filter adapt Tube heater Tube heater Temperature Temperature Volumesensor 1 V 3283 Volumesensor 5 V Pressure 4 ma 4025 Pressure 20 ma Reserve 4 ma 4000 Reserve 20 ma b 20 ma Out c 20 ma Out b 20 ma Out c 20 ma Out Message debounce 10 Beta sensor GM tube Device type F Filter motor micro Steps sens./tube 1600 Pitch dust spot 1600 Intell. corr. Active 1 ICC value 0.39 Vol. GM source 780 Mode filterprint inactive Sensor air T meteor. Sensor air p 4/20mA Pressure b Pressure c 675 Sensor air rh meteor. The values shown here are examples and vary from device to another. A device-specific parameter list is supplied in paper form with the delivery.

263 Page 72 F Rev Vacuum pump specification type VTE 3

264 F Rev. 04 Page 73

265 Page 74 F Rev. 4

266 F Rev. 04 Page Sample tube heater (option) a) Heating tape Type HBST Length 1.0 m Nominal voltage 230 V AC or 110 V AC Power 50 W Min. bending radius 15 mm Thickness 5 mm Breadth 25 mm Protective braid copper, nickel-plated Humidity-proof textile glass fabric This flexible heating tape is especially suitable for the conservation of heat and for antifreeze. HBST has a heat-resistant PTFE insulation, an additional protective conductor and a robust coating made of textile glass fabric. heating conductor PTFE protective braid b) Suitable mounting accessories GBB Broad glass fiber tape for the taping of heating cables, maximum operating temperature: 500 C. Thickness Breadth Unit 1.2 mm 75 mm 100 m roll GBW Fleecy glass fiber tape, about 3 mm thick for the insulation of heated sections. Breadth Unit 25 mm 30 m roll GB Thin, adaptable glass fiber tape for the taping and fixing of heating conductors and temperature sensors, maximum operating temperature: 450 C Thickness 0.15 mm Breadth 25 mm or 15 mm Unit 50 m roll

267 Page 76 F Rev. 4 GSO Glass fiber cord Diameter 2 mm or 3 mm Unit running m c)