TORONTO CITY HALL WINDOW RETROFIT: CONVERSION OF SINGLE GLAZED WINDOWS TO A DOUBLE GLAZED SYSTEM ON A HERITAGE BUILDING

Transcription

1 TORONTO CITY HALL WINDOW RETROFIT: CONVERSION OF SINGLE GLAZED WINDOWS TO A DOUBLE GLAZED SYSTEM ON A HERITAGE BUILDING R. Wood ABSTRACT From 2009 to 2012, over 1200 replacement windows were designed, fabricated and installed at Toronto City Hall. The original, single-glazed windows were replaced with energy efficient, sealed units. The scope included preliminary design of a lock strip gasket system, a stainless steel frame, and an aluminum frame. Interior and exterior glazed options were considered for each. Wind tunnel testing was completed to determine the wind loading criteria. Given City Hall s designation as a heritage building, all changes had to be approved by Heritage Preservation Services. No parts of the historic building fabric could be removed, the new system had to be reversible, and the overall appearance had to remain consistent. The alternatives were evaluated on Health and Safety, Heritage, Capital Cost, Lifecycle Cost, and Technical Performance. A custom, stainless-clad, aluminum frame system was selected. It harnesses the structural and performance benefits of extruded aluminum systems, while respecting the original architect s use of stainless steel. The exterior contains a pressure-equalized, drained cavity to provide optimal performance. A comprehensive quality control program was developed. Full scale replicas of the existing system were used for testing, training of glaziers, and exploration of improvements to the design and installation. Material and workmanship were regularly reviewed through production, fabrication, and installation. Finally, a rigorous in-situ testing program was implemented. Due to the complexity of this type of project, close collaboration among all members of the project team was crucial to a successful outcome. Integrating the architectural, engineering, fabrication, and installation teams from the onset of the project enabled numerous improvements in process and design. These improvements resulted in significant capital cost savings for the client, a higher performing window system, and on-time completion. Temperature regulation complaints have decreased significantly and reports of condensation or frost on the interiors have been eliminated. INTRODUCTION Toronto City Hall was officially opened in In 2009, C3 Polymeric was awarded the contract to undertake the design and replacement of 780 original windows on the east tower. They were to be replaced with energy efficient, double pane units. Not only were the existing windows stained, but occupants complained of discomfort due to inconsistent temperatures in their working environment. This had negative impacts on both energy efficiency and employee productivity and morale. The original City Hall design by Finnish architect Viljo Revell was the winning entry of an International Design Competition for a new Toronto City Hall. City Hall is an established landmark and tourist destination. It was listed as a heritage property in 1976, and officially designated in In undertaking the retrofit, the architectural significance and heritage designation of the building required consideration. 511

2 Toronto Hydro Energy Services (TH) completed a feasibility study to address renewable and energy efficiency opportunities at in early A design charrette in May 2007 brought together a number of experts to generate ideas on modernizing City Hall. Measures to improve the building automation system and integrate the security and lighting system were identified and are being completed. Window replacement was also one of the options identified. The window replacement project was planned to remove and replace the existing single pane windows with a double pane design. There are a total of 960 large windows and 264 corner windows on the two towers. The east tower is the larger of the towers, with 630 large and 150 corner windows. DESCRIPTION OF ORIGINAL SYSTEM The original glazing system was comprised of a lock strip (zipper) gasket on a non-thermally broken stainless steel frame (Figure 01). The vision units are in strips of punch openings. They are 6mm, annealed units, approximately six feet wide by eight feet high. The locking strip is on the interior of the assembly. A trough or gutter extends around all four sides of the vision unit on the interior. The depth of the trough varies between the head, sill, and jambs. Single glazing typically has an insulation value of approximately R The thermal properties and glare control of the original system were very poor. Temperature regulation was difficult to achieve, especially on the upper floors of the east tower, which are exposed to significant afternoon and evening sun. An interior film and blinds had been added to help control these issues, but had provided minimal benefit. During summer months, interior temperatures in excess of 28 Celsius were common. Further, the film had degraded from its original condition, exacerbating the problem with glare. Building operations reported 40 to 50 weekly complaints related to the interior environment. FIGURE 1: EXISTING SECTION, HEAD, AND SILL DETAILS The original system was also prone to both water penetration and condensation. Relative humidity in the building was maintained at a low level to avoid condensation. Low humidity levels contributed to occupants discomfort. Previous attempts had been made to humidify the Council chambers but the condensation on the sloped chamber windows became too much of a problem and the humidification was turned off. 512

3 DESIGN PROCESS The design phase of the project involved the establishment of wind loading criteria, preliminary engineering design of six alternatives, and evaluation of the alternatives to determine which should proceed to the testing and mockup phase. WIND LOADING CRITERIA It was recognized that the traditional methods of calculating the design loads might not be appropriate for. The area surrounding City Hall had changed significantly since its construction in the 1960s. Further, the impact of the curved structure on window pressures is difficult to calculate. As such, wind tunnel testing was undertaken to refine the wind loading criteria. The wind tunnel testing was completed by the Alan G. Davenport Wind Engineering Group at the Boundary Layer Wind Laboratory at the of Western Ontario (Figure 02). A 1:400 scale rigid model of City Hall and its surroundings was constructed. The building model, built from ABS Plastic, was equipped with 354 pressure sensors, 188 on the east tower and 166 on the west. The proximity model of the surrounding city was built in block outline from Styrofoam for a radius of approximately 500 m. FIGURE 2: MODEL OF CITY HALL (WHITE) AT BOUNDARY LAYER WIND TUNNEL LABORATORY The building model and the proximity model were rotated in 10 degree increments to simulate different wind directions, with the upstream terrain being changed as appropriate. The experimental pressure data did not show any unusual sensitivity to the precise angle of the wind. The Canadian building code recommends an allowance of 0.2 kpa for every 100m of height, about 0.2 kpa total for the east tower, to allow for stack effects. It also recommends an allowance of 0.1 kpa for mechanical system effects. An importance factor of 1.10 was used to factor the load. TABLE 1: MAXIMUM DIFFERENTIAL PRESSURES AND SUCTIONS 513

4 The maximum pressures from the wind tunnel testing were 1.9 kpa and 3.2 kpa for the large and corner units respectively (Table 01). Factoring these results as described above resulted in a design pressure of 2.42 kpa (50.5 psf) for the large windows and 3.85 kpa (80.4 psf) for the stair windows. INSULATED GLAZING UNIT DESIGN Pilkington Energy Advantage Grey Eclipse (surface 2) was selected for the glass. A warm edge spacer and argon gas were also specified. This particular makeup of glass was selected by the project architect, and pre-approved by Heritage Preservation Services (HPS) prior to the start of engineering design. The goals at this point were to improve the thermal performance and decrease glare, while keeping changes to the overall aesthetic to a level acceptable to HPS. An analysis was completed to determine the required glazing thickness required to counteract the expected maximum design pressures from strength of materials perspective and to work within an acceptable deflection criteria for the windows. The analysis resulted in an IGU makeup consisting of two, 6mm tempered panes and an air space of 12.7mm. DESIGN ALTERNATIVES The scope of the project included the preliminary design and evaluation of six different alternatives. The alternatives were classified into three types: a locking strip gasket system; a stainless steel frame; and an aluminum frame. Both an interior glazed and exterior glazed option was considered for each type of system. EVALUATION CRITERIA Engineered drawings were produced for each of the alternatives. The project team evaluated each option based on the following criteria: 1. Safety 2. Heritage 3. Capital Cost of Retrofit 4. Lifecycle Costs 5. Technical (Air, Water, Thermal Performance) HERITAGE CONSIDERATIONS Given s designation as a heritage building, all changes to the building had to be approved by Heritage Preservation Services. Should they not be satisfied, they had the ability to veto the project. Consultation with representatives from Heritage Preservation Services (HPS) was conducted, and the characteristics of the heritage attributes related to the façade, glazing, and frame identified. These included the following: Strong horizontality created by faceted pre-cast concrete panels and large glazing elements bookended by vertical precast elements. Strong secondary vertical stainless steel elements between glazed units. Minimal visual impact of glazing frame due to relatively thin matt black gasket system which floats the glazing between minor stainless steel frame elements. Film on glazing and interior blind system which impacts the façade are not original to building. 514

5 Based on these characteristics, and additional consultation with Heritage Preservation Services, the primary heritage criteria were identified as follows: Aside from the existing glazing and gasket, all components of the historic fabric of the building had to remain. The new system had to be reversible. That is, it had to be possible to reinstall the original glazing system at some point in the future. This criteria was included primarily for the insitu mockup phase, so that HPS could reject the mockups if they felt they did not comply. Minimize the decrease in the daylight opening (visible area). Respect the architect s original design intent. Respect the original building materials and fabric. Given HPS s ability to veto any of the design alternatives if they were not satisfied, the heritage criteria were the dominant criteria throughout the design process. EVALUATION Health and Safety: The original windows were replaced by moving the window panes through the stairwells as they did not fit in the elevator. Due to the weight and thickness of sealed units, this approach would not be acceptable with the new window system. Regardless of the system selected, new units will have to be brought up from the exterior. Given this constraint, interior glazed units will require additional handling to put into position. As the units weigh over 300 pounds, this is a significant disadvantage for the interior glazed systems. There were no other overriding health and safety concerns with any of the three types of system. Heritage: The lock strip gasket option would best meet the heritage criteria. All systems would result in some loss of daylight opening. The aluminum framed option would result in less loss than the stainless options. The aluminum system introduced a new material. This could be mitigated somewhat by cladding the interior of the system in stainless steel. Capital Cost: The lock strip gasket option would be slightly less expensive than the aluminum option. The stainless option would be considerably more expensive than either of the others due to substantial difference in material costs, and increased difficulties of working with stainless steel due to its increased weight and ability to cut and drill. Lifecycle Costs: The exterior options would have a slightly lower replacement cost than the respective interior options, as the interior options require additional handling of the units to move them into place. As noted above, all units have to be brought into position via the exterior. Higher failure rates were predicted for the gasketed systems because of poorer drainage and because the units are less well supported than the other systems. 515

6 Technical (Air, Water, Thermal Performance): Lock Strip Gasket Good thermal performance due to the gasket acting as a thermal break. The windows are held in place by the gasket system without a positive method of retention. Rarely used in Ontario so there is a lack of experience/trust with these systems. Highest potential for leakage. Susceptible to UV degradation. Highest probability of damage to vision unit during installation. Lowest tolerance for variation of existing frames dimensions. Manufacturer unable to verify that it can meet the structural loading requirements. Stainless Steel Solid structurally, not susceptible to UV degradation. More corrosion resistant than aluminum. Finish is durable. Fabricated stainless steel tubing has rounded corners and is subject to more tolerance variation. Framed joints need to be welded and the interior welds need to be ground smooth. The ability to key gaskets and thermal breaks into keyway is not available. There will be resulting shifts in the seal at the glazing line due to the effects of pumping of the sealed unit. Exterior Option The mechanical connection of the pressure plate to the mullion neck is not a very positive connection. The adequacy of the fastener depends on the number of screw threads engaged. Removal and replacement will likely damage the internal threads. The effects of long-term loading cycles will also tend to damage the threads. Over-torque of the fastening at the pressure plate can easily damage the internal threads, rendering the hole useless. Rolling tolerances of the snap on caps and the pressure plate may cause dislodgement. Provisions need to be made to mechanically hold the caps in place, especially at the vertical jambs. Interior Option The interior alternative uses a snap on cover arrangement that may be easily damaged during removal and replacement. The exterior gasket is not positively held in place, so dislodgment is possible. A heel bead is required for air seal continuity. A stud held pressure plate is difficult to fix or replace in the event of failure. Future air seal problems would be difficult to fix without major disassembly of components. Air and water seals at the main fastening points cannot be inspected. Aluminum Proper pressure equalized system has been incorporated in the design. Aluminum extrusion allows for tight tolerances and for very positive retention of the gaskets. Aluminum fabrication and construction is the most predictable and has a positive track record. Potential for some galvanic corrosion with stainless. 516

7 DESCRIPTION OF THE SELECTED SYSTEM The lock strip gasket systems were eliminated from consideration due to technical concerns, primarily with respect to their ability to meet the loading criteria. The stainless systems were eliminated due to their cost. The selected system was a hybrid system of a custom aluminum frame clad in stainless steel (Figure 03). This enabled the design team to harness the structural and performance benefits of an extruded aluminum system, while at the same time respecting Revell s use of stainless as a primary design element. The system utilized a total of 16 custom aluminum, neoprene, and PVC extrusions. While off-the-shelf systems were considered, no single system was found to meet all of the evaluation criteria. Specifically, systems that may have met the structural and thermal criteria would not meet the heritage criteria. The primary frame member was very low profile, to minimize reduction of the daylight opening, with heavy walls to meet the structural criteria. A conventional would have resulted in a loss of daylight opening approximately 200% greater that the custom system. The new system incorporated a pressure equalized, drained cavity on the exterior. This is a feature not typically found in punch FIGURE 3: SECTION OF NEW WINDOW SYSTEM window systems. It helps to extend the expected lifecycle and improve the performance of the window system. The system was designed such that the vision units could be pre-glazed. This reduced the amount of on-site installation time required, and permitted glazing in a controlled environment. The only field applied primary seal was between the new and existing frames. It was easily accessible and could be inspected for quality on site. Custom aluminum tubes were designed to retrofit the surround so that it could accommodate the new frame. The original zipper gasket was seated on a relatively narrow frame. The design team determined that the existing narrow frame was not an adequate or practical support point for the new system. The tubes filled in the trough around the full perimeter of the window and made it level with the zipper gasket frame. The tubes were designed in a manner such that they could be removed at a later date if required, thereby helping to satisfy the heritage criteria. Two types of closures were added on the interior. A black anodized closure with a comparable width to the original lock strip gasket was used to hide the glazing leg of the frame. The remainder of the frame and aluminum tubes were clad in stainless steel, a primary material used in City Hall by Revell. A narrow pressure plate and black anodized snap cap was used. The cap was similar in dimension and appearance to the lock strip gasket, helping to minimize the impact on the façade. As viewed from the podium level, it was difficult to distinguish the new framing system from the original beyond the first few floors. 517

8 THERMAL MODELING AND LAB TESTING Thermal modeling was completed using THERM. The thermal model was also reviewed and validated by an independent third party. The model showed that the new system would have a U-Value of approximately 2.34 W/m 2 C. Full scale replicas of the existing frame were constructed. Mockups of the new system were fabricated and installed into the replica frames. The entire assembly was tested for thermal, structural, and air and water performance. These tests are normally conducted on only the new components of a system. Including the existing surround created a more rigorous, authentic test environment. The new system passed all tests. IN SITU PROTOTYPE AND TESTING In situ mockups were installed on the 8 th and 10 th floors at Toronto City Hall early in These mockups were reviewed and approved by Heritage Preservation Services. Both the shop and in situ mockups provided opportunities for the design team to review the installation, make changes to the design and methods to reduce the time required for installation, and improve the quality. Based on both the shop and site mockups, it was determined that the installation of the structural channels was a bottleneck in the installation process. The channels in the original design were fabricated out of stainless steel. The installation process required drilling through the stainless to secure the channels to the existing surround. In addition, the tolerances on the fabrication of the channels resulted in slight differences in their height, which made the overall installation process more difficult and time consuming. To resolve the problem, the channels were re-engineered out of extruded aluminum. This provided very tight tolerances. The channels were also significantly lighter, and easier to cut and drill on site. These features helped to significantly improve the speed and safety of the installation for the crews. Another key element identified in the mockups was the indexing of the frames. The frames had to be installed in the correct vertical alignment so that there would be proper compression on the pressure plate gaskets. Proper compression is critical for a high performing window system as it helps to minimize water penetration. During the mockups, the alignment had to be manually measured at multiple locations around the perimeter of the window. Further, the geometry of the components made accurate measurements difficult to obtain. Even if measured properly, it was hard to keep the new frame in the correct position while drilling and fastening it in place. FIGURE 4: SAMPLE CHEATER A custom indexing guide (Figure 04) was developed for use in the installation. A number of prototypes were produced to refine 518

9 the tolerances, which were measured in thousandths of an inch. The guides were installed around the perimeter of the new frame. During installation, the crew had to only worry about applying pressure on the exterior so that the indexing guides pressed firmly against the existing stainless steel frames. No site measuring was required with this approach. As an added benefit, the indexing guides were designed to hold the glass in place until the pressure plates were installed. Typically cheaters or short pieces of pressure plate are used for this purpose. The spacing of the fasteners was also modified, as a small percentage of the fasteners was significantly harder to install, disproportionately increasing the amount of time required. The suggestion was reviewed by the engineering team, replicated in a mockup, lab tested, and implemented on the full scale installation, and resulted in a productivity improvement of over 20%. In addition, it improved health and safety, as less physical exertion and repetitive motion were required during installation. QUALITY ASSURANCE The replica frames enabled the design and installation team to conduct mock installations in a controlled environment. These were used to explore improvements to the design and installation methodology. The new system is significantly different than commercially available glazing systems. The installation team had to be trained on the unique aspects of the system, and the key points to monitor to ensure a quality installation. The models allowed the installation crews to train and become familiar with the differences and nuances of the new system prior to being on site. A quality assurance program was also developed for the installation. This included a series of checklists that were completed at various stages throughout the fabrication and installation process. Wherever feasible, components were colour-coded to enable a quick visual inspection to ensure they were installed in the correct location or orientation. A rigorous in situ testing program was also implemented. Multiple pressurized air and water tests were conducted throughout the project to ensure consistent quality and workmanship. ENERGY SAVINGS AND OCCUPANT BENEFITS Annual reductions in energy use of 800,000 kwh electricity and 10,000 klbs of steam were calculated by Toronto Hydro Energy Services at the beginning of the project. That is the equivalent to elimination of almost 2,000 tonnes of greenhouse gases every year. Actual energy use is being monitored. Based on occupant feedback, the new windows have had a significant positive impact on their work environment. Building operations report a significant decrease in complaints related to temperature regulation. Complaints have decreased from approximately to approximately 3-4 per week. The new windows have eliminated complaints of condensation or frost buildup on the interiors. RESULTS & CONCLUSION The east tower was completed in 2011 (Figures 05, 06), and the west in Both towers were completed on time and on budget. The new windows have had a significant positive impact on the work environment. Due to the complexity of this type of project, close collaboration among all members of the project team was crucial to a successful outcome. Integrating the architectural, engineering, fabrication, and installation teams from the onset of the project enabled numerous improvements in process and design. These improvements resulted in significant capital cost savings for the client, and a higher performing window system. 519

10 14 CCBST 2014 Proceedings Book_v11 closing _Layout :24 PM Page 520 FIGURE 5: INSTALLATION OF NEW WINDOWS ON THE EAST TOWER FIGURE 6: FINISHED WINDOWS AS VIEWED FROM INTERIOR 520