Final Program. PVP Pressure Vessels & Piping Conference. New Horizons in Global Pressure Vessel and Piping Technology

Transcription

1 Final Program PVP Pressure Vessels & Piping Conference New Horizons in Global Pressure Vessel and Piping Technology July 15-19, 2012 Sheraton Centre Toronto Hotel Toronto, Ontario, Canada

2 Proceedings of the ASME 2012 Pressure Vessels & Piping Conference PVP2012 July 15-19, 2012, Toronto, Ontario, CANADA EVALUATION OF STUFFING BOX FORCES ASSOCIATED WITH THE THERMAL EXPANSION OF e-ptfe AND GRAPHITE PACKING SETS Carlos D. Girão Keith Guenther Teadit Industria e Comercio Ltda Av Martin Luther King Rio de Janeiro, RJ Brazil PVP ABSTRACT This paper evaluates the stuffing box forces associated with mechanical packing sets comprised of a combination of e-ptfe and Graphite materials. Utilizing a simulated stuffing box test device and protocol, radial and axial forces associated with thermal stresses are measured for different packing configurations. The testing varies the quantity and location of both e-ptfe and Graphite packing material within a five ring packing set. Test results are reported with comparative data and correlations among the configurations. NOMECLATURE - Shaft torque r - Shaft Radius - Lateral Deformation Factor - Friction Coefficient N - Number of Packing Rings A - Contact area between the Packing and the Shaft - Axial Stress i - Axial Stress on Ring number i 1 - Gland Stress % 1 - Percentage of Gland Stress Increment due to Temperature Increase % - Percentage of Torque Increment due to Temperature Increase INTRODUCTION Industry has recognized the inherent benefits of braided packings produced from e-ptfe filaments. Low coefficient of friction and outstanding chemical stability make it well-suited for sealing aggressive fluids. Nevertheless, PTFE packings tendency to expand due to temperature variations, can greatly increase stuffing box internal stresses. This leads to high shaft torque and eventually scorching the packing. The correlations among thermal expansion, stuffing box stress and the shaft friction force for different e-ptfe braided packings were carefully studied in a previous PVP paper [1]. The study showed that packing thermal expansion can be controlled with engineered filaments by increasing the amount of fillers in the packing and therefore reducing the amount of PTFE. However, this is not a variable that can be modified by the end-users. Since different combinations among e-ptfe and other type of braided packings can be easily done by any user, it was decided to perform a similar study with packing combination sets. Tests to evaluate the performance of these configurations were conducted by changing the number and set arrangement of e-ptfe and Flexible Graphite braided packing rings. THERMAL EXPANSION TEST RIG A test rig that simulates a stuffing box for 6,4mm (1/4in) cross section packings with a 25,4mm (1 in) diameter shaft was used to perform the studies. In order to meet the requirements of surface finish for valves under API 600 [2], API 602 [3] and MSS SP 120 [4] it was assured that the shaft had a surface finish (R a ) of 0,8 m (32 in) and the stuffing box a surface finish (R a ) of 3,2 m (125 in). This was the same rig cage developed for the previous study about the influence on the stuffing box of the forces generated by packing thermal expansion. The same test protocol was also utilized in order to allow comparisons between the two studies. The detailed description of the test 1 Copyright 2012 by ASME

3 rig and the instrumentation are all presented in the previous PVP paper [1] mentioned above. The test rig image and schematics are shown in Figures 1 and 2. FIGURE 1 TEST RIG 2. Random loads are applied on a metal bushing to assure the load cell and the gauged bolts are correctly calibrated. 3. Five packing rings are installed with an initial stress of 50MPa (7252 psi) to seat the packing and the load on the last ring registered. 4. After one hour, the readings are registered and the gland bolts are relaxed to be again taken up until finger-tight. 5. The packing is allowed to relax for ten minutes and the loads on the gauged bolts and on the load cell are again recorded. 6. The temperature is increased to 100ºC (212ºF). 7. After a dwell time of twenty minutes the loads are registered. 8. The shaft torque is then measured using a torque meter. Four readings are made with a half-turn of the shaft each. The first reading records the static torque while the three subsequent readings record the dynamic torque. 9. After the two complete shaft turns, the readings on the gauged bolts and on the load cell are registered. 10. The temperature is decreased to room temperature and steps 7-9 are repeated. 11. Increase the temperature 100ºC (212ºF) and repeat steps 7-9 to finish the test. PACKING SETS ARRANGEMENT The tests were performed with five different packing combination sets and their dynamic properties registered. Packing Combination Set A was built from 100% PTFE braided packing rings. Packing Combination Sets B, C and D were built from arrangements of PTFE along with Flexible Graphite braided packings rings. Packing Combination Set E was built from 100% Flexible Graphite braided packing rings. Table 1 indicates the number of rings in each combination set tested. TABLE 1 COMBINATION SETS NUMBER OF RINGS TEST PROCEDURE FIGURE 2 TEST RIG SCHEMATICS The test procedure developed to analyze the different packing behavior when subjected to temperature cycles is described below. The packings are initially seated, then relaxed to be again taken up until finger-tight as it would be done in a pump installation procedure. The evaluation of the thermal expansion phenomena under the high stresses used in valve applications is not within the scope of this study 1. The Rig is cleaned to assure that the shaft and the stuffing box are free from any unwanted material/dust. COMBINATION Flexible PTFE SET Graphite A 5 0 TEST RESULTS B 4 1 C 3 2 D 2 3 E 0 5 GLAND STRESS x TEMPERATURE The tests allowed the analysis of the combination sets behavior when subjected to temperature cycles. The charts 2 Copyright 2012 by ASME

4 illustrated below (Figures 3 through 7) were plotted indicating the gland stress variations from room temperature to 100ºC in two thermal cycles. Each column pair indicates the stress in: room temperature; 100ºC (212ºF); room temperature after torque measurement; and at 100ºC (212ºF) again after torque measurement. FIGURE 5 GLAND STRESS x TEMPERATURE PACKING COMBINATION SET C FIGURE 3 GLAND STRESS x TEMPERATURE PACKING COMBINATION SET A FIGURE 6 GLAND STRESS x TEMPERATURE PACKING COMBINATION SET D FIGURE 4 GLAND STRESS x TEMPERATURE PACKING COMBINATION SET B 3 Copyright 2012 by ASME

5 FIGURE 9 COMBINATION SET B ARRANGENT (4 e-ptfe rings and 1 Flexible Graphite ring) FIGURE 7 GLAND STRESS x TEMPERATURE PACKING COMBINATION SET E In the charts above (Figures 3 through 7) it can be noticed that temperature changes have a direct influence on the gland stress values. The stress for packing Combination Set A, for instance, is 50% greater at 100ºC (212ºF) than at room temperature. Comparisons among the packing combination sets show that the use of graphite rings reduced the effect of thermal expansion and contraction as the combination sets where exposed to temperature cycles. The difference among the packing combinations sets are not characterized only by the quantity of braided packing rings but also by the arrangement of the e-ptfe and the flexible graphite rings along the set. Figures 8 through 12 show how the packing were arranged in the combination sets tested. FIGURE 10 COMBINATION SET C ARRANGENT (3 e-ptfe rings and 2 Flexible Graphite rings) Legend: FIGURE 11 COMBINATION SET B ARRANGENT (2 e-ptfe rings and 3 Flexible Graphite ring) FIGURE 8 COMBINATION SET A ARRANGENT (5 e-ptfe rings) FIGURE 12 COMBINATION SET C ARRANGENT 4 Copyright 2012 by ASME

6 (5 Flexible Graphite rings) SHAFT TORQUE x TEMPERATURE In a previous PVP paper [5] the following equation was proposed to calculate the shaft torque. N r A (1) i 1 According to the formula above the variation of the shaft torque is directly affected by the gland stress, 1. The percentage increase in gland stress, 1, from room temperature to 100ºC (212ºF) is shown in Figure 13. This value was calculated by subtracting the gland stress value at 100ºC (212ºF) from the measurement at room temperature and dividing by the initial stress at room temperature. i i FIGURE 14 TORQUE INCREASE DUE TO TEMPERATURE INCREASE The above graphs illustrate the large influence of e- PTFE on a packing set expansion and consequently on gland stress and shaft torque. Comparing charts from Figure 13 and Figure 14 it can be observed that both, gland stress and shaft torque, exhibit the same behavior. PUMP SIMULATOR TEST RIG The packing sets were also tested in a Pump Test Rig that simulates the working conditions of a pump using water as test media. The main packing performance parameters like leakage, power consumption and number of gland packing adjustments were monitored in a 100-hour test. A scheme of the Test Rig is illustrated in Figure 15 and the actual Rig is shown in Figure 16. The test parameters are shown in Table 2. FIGURE 13 GLAND STRESS INCREASE DUE TO TEMPERATURE INCREASE The chart shows that Combination Set A has a % 1 nearly four times greater than Combination Set E. It also shows that the set E, the combination set with the least e- PTFE rings content, generated lower thermal expansion forces than the others. It is also possible to notice the direct correlation between the amount of e-ptfe rings that each packing set have and the gland stress increase with temperature. The higher the number of e-ptfe rings the higher the percentage of gland stress increase, % 1 The effects of these variations on shaft torque are shown in Figure Centrifugal Pump 5 Water Reservoir 2 Control/ Monitor Tests 6 Control Valves 3 Shaft Leak Rate Detection 7 Pressure Indicator 4 Gland Leak Detection 8 Programmable Controller FIGURE 15 PUMP SIMULATOR TEST RIG SCHEMATICS 5 Copyright 2012 by ASME

7 The pump test leak rate results for packing combination sets A, B and D are shown in Figures 18, 19 and LEAK RATE (ml/min) SHAFT SIDE STUFFING BOX SIDE FIGURE 16 PUMP SIMULATOR TEST RIG TABLE 2 TEST PARAMETERS TIME ( hrs ) FIGURE 18 PACKING COMBINATION SET A LEAKAGE TEST RESULT DESCRIPTION PARAMETER 1000 Shaft Speed (max.) Shaft Size Number of Rings Test Media Media Pressure Test Temperature Test Duration 8m/s (26fps) 47mm (1,8in) 5 rings Water 6 bar (87psi) Room Temperature 100hours The leak rates along the shaft side and the stuffing box side were controlled at all times along the tests. The water leakage was collected into two separate small reservoirs. One of these reservoirs collected the leakage from the stuffing box side which is the leakage between the stuffing box and the gland follower as illustrated in Figure 17. The other reservoir collected the leakage from the shaft side. The origin of the collection of this leakage is also illustrated on Figure 17. The reservoirs were instrumented to measure the volume of the fluid collected along the time allowing the registration of the leak rate throughout the test LEAK RATE (ml/min) SHAFT SIDE STUFFING BOX SIDE TIME ( hrs ) FIGURE 19 PACKING COMBINATION SET B LEAKAGE TEST RESULT LEAK RATE (ml/min) SHAFT SIDE STUFFING BOX SIDE TIME ( hrs ) FIGURE 20 PACKING COMBINATION SET D LEAKAGE TEST RESULT FIGURE 17 SHAFT SIDE AND STUFFFING BOX SIDE LEAKAGE Packing Combination Set A showed higher leakage with very intense responses to gland adjustment. The addition of one flexible graphite ring in Combination Set B considerably improved the results as shown in Figure 19. The addition of more Flexible Graphite rings improved the set overall reaction to gland adjustments and leak rates decreased 6 Copyright 2012 by ASME

8 as shown in Figure 20. Te test summary chart can be seen in the table below (Table 3). TABLE 3 SUMMARY CHART Packing Set Average Leakage (ml/min) A 68 B 17,7 D 10 [4] MSS SP 120. Flexible Graphite Packing System for Rising Stem Steel Valves Design Requirements [5] VEIGA, J., CIPOLATTI, C. GIRÃO, C. The Influence of Different Braiding Packing Materials and Number of Rings on Stem Torque and Sealability. PVP [6] The Fluid Sealing Association / European Sealing Association. Pump & Valve Installation Procedures The temperature and power of the pump was also monitored along the tests. Figure 21 shows that there is a direct correlation between these two variables. Temperature ( C) TEMPERATURE AND POWER x TIME 40 0,5 20 Temperature 0 Power 0, TIME (hrs) FIGURE 21 PACKING COMBINATION SET D TEMPERATURA AND POWER PROFILE 2,5 2,0 1,5 1,0 Power (kw) CONCLUSIONS The tests show how the role played by thermal expansion influences packing performance on pumps. Packing expansion due to temperature variations can greatly increase stuffing box internal stresses and lead to high shaft torque and eventually scorching the packing. The previous knowledge of this behavior can help end users adapt installation procedures [6] in order to improve performance results. REFERENCES [1] VEIGA, J., CIPOLATTI, C. GIRÃO, C. et al. The Influence on the Stuffing Box of the Forces Generated by Packing Thermal Expansion. PVP [2] ANSI/API STD 600. Bolted Bonnet Steel Gate Valves for Petroleum and Natural Gas Industries. 11 th Edition (ISO 10434:2001 Modified, Bolted bonnet steel gate valves for the petroleum, petrochemical and allied industries) [3] ANSI/API STD 602. Steel Gate, Globe and Check Valves for sizes DN 100 and Smaller for the Petroleum and Natural Gas Industries. 8 th Edition (ISO 15761:2002, Steel gate, globe and check valves for sizes DN 100 and smaller, for the petroleum and natural gas industries) 7 Copyright 2012 by ASME