Mr. Chris Cocallas University Architect and Director Capital Planning and Construction Colorado School of Mines th St. Golden, Colorado 80401

Size: px

Start display at page:

Download "Mr. Chris Cocallas University Architect and Director Capital Planning and Construction Colorado School of Mines th St. Golden, Colorado 80401"

Nicholas Walters
5 years ago
Views:

Mr. Chris Cocallas University Architect and Director Capital Planning and Construction Colorado School of Mines 1801 19th St.

1 Mr. Chris Cocallas University Architect and Director Capital Planning and Construction Colorado School of Mines th St. Golden, Colorado Re: GRL and GRLA Building Noise Study Wave #1434 Dear Chris, We completed the noise assessment for rooftop equipment on the General Research Laboratory (GRL) and General Research Laboratory-Annex (GRLA) buildings at the Colorado School of Mines (CSM). I measured sound levels on the roof near each noise producing piece of equipment and on the ground in the vicinity of nearby residences. This report is an update of our original report dated April 8, We have re-measured sound levels at night after modifications have been made to fans on both the GRL and GRLA rooftops. The study shows that the most significant noise sources were initially the large General Exhaust fans and motors on the GRLA building and the induced flow Lab Exhaust fans on the GRL building. The GRL Lab Exhaust fans were re-sheaved on March 25, 2016, to slow the fan speed, and this significantly reduced noise from these fans. The Variable Frequency Drives serving the GRLA exhaust fans were adjusted and the annoying chirp or whine was reduced and shifted to a higher frequency. The annoying chirp is no longer audible at most locations on the ground. The GRL Lab Exhaust fan noise is reduced below the background sound and other equipment. The GRLA General Exhaust fans are still audible to the north, but the smaller GRLA Lab Exhaust fans are louder than the larger fans at locations northeast of the building. This report is intended to summarize the measurements and mitigation that have been done so far, and to comment on additional noise mitigation measures that can be considered.

Page 2 Test Procedures To minimize the influence of background noise from traffic on Highway 93 and local streets, I measured sound levels in the area on Friday March 25, between 12:30 a.m. and 3:30 a.

2 Page 2 Test Procedures To minimize the influence of background noise from traffic on Highway 93 and local streets, I measured sound levels in the area on Friday March 25, between 12:30 a.m. and 3:30 a.m. I measured sound levels at or near five locations that you had previously measured at to allow comparison with those measurements. The same testing procedure was repeated on April 22, 2016, at the same time of day. The measurement locations are shown on Figure 1. Figure 1: Measurement locations

3 Page 3 The following test equipment was used. Sound Level Meter M1 (Locations 1, 4, and 8) Larson Davis Model 831 sound level meter S/N , Type 1 per ANSI S1.4 PCB preamp PRM831, S/N PCB ½ʺ microphone Model 377B02, S/N Sound Level Meter M2 (Location 7) Larson Davis Model 831 sound level meter S/N , Type 1 per ANSI S1.4 PCB preamp PRM831, S/N PCB ½ʺ microphone Model 377B02, S/N Sound Level Meter M3 (Location 2) Larson Davis Model 831 sound level meter S/N , Type 1 per ANSI S1.4 PCB preamp PRM831, S/N PCB ½ʺ microphone Model 377B02, S/N Larson Davis Electronic Calibrator CAL200, S/N 2905 The system calibration was checked in the field before and after each series of measurements. The calibration of all equipment has been certified by the manufacturer. Meter M1 was moved between Locations 1, 4, and 7 during each equipment scenario. The data shown in Table 1 are two to three minute averages during each scenario. Meters M2 and M3 ran continuously at their respective locations from approximately 12:45 a.m. to 3:30 a.m. on Friday March 25, 2016, and Friday, April 22, The data shown in Table 1 are five minute averages during each scenario. I measured sound levels under a variety of operation conditions to help determine the individual contributions of each piece of equipment. Mike Willey of CSM turned equipment on and off to create the different equipment scenarios. Please note that the GRLA General and Lab Exhaust systems, and the GRL Lab Exhaust systems each include two fans, but only one runs at a time. Only one fan from each system was running during our tests. I also measured sound levels close to the equipment on the rooftops of the GRL and GRLA buildings (with Meter M1). We used this data in our analysis to help judge the most significant noise sources, and to evaluate mitigation. These measurements were performed on March 10, 22, 25, 29, and April 6, 2016.

4 Page 4 The equipment scenarios for the early morning March 25 and April 22 measurements are identified as follows. All off) Scenario 3a) Scenario 3b) Scenario 3c) Scenario 4d) Scenario 4f) All on) all rooftop equipment on GRL and GRLA turned off. This condition is used to determine the background noise levels GRLA small Lab exhaust fan & MAU on, GRL rooftop equipment off GRLA Gen Exh fan & MAU on, GRL rooftop equipment off GRLA Gen Exh fan, small Lab Exhaust fan & MAU on, GRL rooftop equipment off GRL Gen Lab Exh Fans on, MAU on, Lab Exhaust fan off, GRLA rooftop equipment off GRL Gen Lab Exh Fans on, MAU on, Lab Exhaust fan on, GRLA rooftop equipment off all rooftop equipment on GRL and GRLA was on. Test Results March 25, 2016 Table 1: Sound pressure levels (in dba) under various equipment scenarios, 3/25/16 March 25, 2016 Location No. Scenario All off, background a b c d f All on 39 47/47* 45 47/44* 47 * Mar 25/Mar 29 I measured during the early morning on March 25 except where noted otherwise. The GRL Lab Exhaust fans were re-sheaved after the March 25 measurements. I re-measured on March 29 during the day at Locations 2 and 7. That data is also shown in Table 1. The sound levels in Table 1 are L90 sound levels in A-weighted decibels (dba). The L90 is the sound level that is exceeded 90% of the time. Since the sound from fans and mechanical equipment is relatively steady, the L90 can be used to show the sound level of the equipment while minimizing the effect of intermittent cars passing on Highway 93 or local streets, aircraft flyovers, and other intermittent noises. This gives us the best indication of sound levels from the mechanical equipment. The L90 was also used for the background sound levels (all equipment off), so sound levels shown represent the quieter periods during lulls in traffic.

5 Page 5 Initial Results Table 1 data shows that the GRLA small Lab Exhaust fan raised the sound level at Location 8, but was insignificant at other locations. The larger GRLA General Exhaust fan sound was significant at Locations 2 and 8. The whine associated with the motor was audible at Locations 1, 4, and 7, but the sound level was not significantly raised by the fan. Locations 4 and 7 are screened from direct line-of-sight of the General Exhaust fans. The GRL General Exhaust fans and MAU were measureable at all locations except Location 1, but only 2 to 4 dba above the background sound level. The GRL Lab Exhaust fan was much more significant and clearly audible at Locations 4, 7, and 8 before it was re-sheaved. It is worth noting that sound that appeared to be from the Coors plant(s) to the east was distinctly audible at Location 4 and is part of the background noise. Results after GRL Lab Exhaust Fans were re sheaved The GRL Lab Exhaust fans were re-sheaved after the March 25 measurements to slow the fans down. This significantly reduced the sound produced by the fans. This was shown by the measurements on the roof and also by a measurement made during the day on March 29 at Location 8. Results after adjustment of the GRLA General Exhaust Fan VFDs The GRLA General Exhaust fans exhibited a loud whine or chirp that was audible at all of the measurement locations except Location 7. Since the March 29 measurements, the VFDs causing the audible motor whine were adjusted. I visited CSM again during the day on April 6 to observe the improvement. The whine was not audible on the ground at any of the measurement locations. I re-measured on the roof close to the equipment. The single-number dba level is not a good way to judge the severity of the whine, so I measured sound levels across the frequency spectrum in two different ways. You can think of them as two levels of resolution. First, I measured very close to the GRLA General Exhaust fan motor that was producing the strong whine using Narrow Band analysis. This measurement breaks the frequency spectrum from 0 to 5,000 Hz into constant bandwidth bands that are about 1.6 Hz wide. This gives a very accurate indication of the frequencies with the strongest sound energy and of the frequency of any tones. The result of the before measurement is shown in Figure 2.

The series of many sharp spikes, or tones, create an unusual chirp or whine at about 2,200 Hz and higher

6 Page 6 Figure 2: Narrow band measurement of GRLA General Exhaust fan motor before VFD adjustments Figure 2 shows the decibel level (y-axis) versus frequency (x-axis). The series of many sharp spikes, or tones, create an unusual chirp or whine at about 2,200 Hz and higher frequencies. Figure 3 shows a measurement at the same location after the VFD adjustment. Figure 3: Narrow band measurement of GRLA General Exhaust fan motor after VFD adjustments

7 Page 7 The difference between Figures 2 and 3 is dramatic. You can see the reduction in amplitude and number of spikes, and also that heavy blue areas (indicating more sound energy) shift to higher frequencies. The most distinct spikes now start at about 4,000 Hz. Higher frequencies are attenuated at a higher rate as sound travels through the atmosphere. The plots in Figures 2 and 3 are interesting, but it is somewhat difficult to see how this affects the sound levels that we hear or measure. Figure 4 shows the sound levels measured at the same times and locations, but presented in onethird octave bands. One-third octave bands show the sound levels with less frequency resolution, but are standardized so the actual sound levels are more meaningful. Figure 4: 1/3-octave band measurements of GRLA General Exhaust fan motor before and after VFD adjustments The one-third octave band data show that the peak sound energy was reduced in amplitude and moved from the 3150 Hz to 5,000 Hz band. I could still hear the much reduced whine when standing near the fan, but I could no longer hear it on the ground. Having said this, my visit on April 6 was during the day when ambient noise was higher than at night. After my work on the roof, the wind picked up and made measurements on the ground meaningless. The sound levels should be evaluated again at night when ambient sound is low.

8 Page 8 Reduction of the GRL Lab Exhaust Fan Noise I took similar measurements next to the GRL Lab fan discharge stack. Figure 5 shows the onethird octave data before and after the fan was re-sheaved to slow its speed. Re-sheaving the fan reduced the sound level across the frequency spectrum. The overall sound level was reduced by about 18 dba. The resulting reductions at Location 7 was about 3 dba since other sources contribute as well. Further reduction should not be necessary for the GRL Lab fans.

9 Page 9 Test Results April 22, 2016 The nighttime round of testing was repeated on April 22, The results are shown in Table 2. Table 2: Sound pressure levels (in dba) under various equipment scenarios, 4/22/16 4/22/16 testing Location No. Scenario ) All off, 41* 41* 43* 38* 41* background 3a 42* 44* 45* 38* 46* 3b 41* 45* 42* 41* 43* 3c 41* 45 42* 39* 46* 4e 41* 40 41* f 40* ) All on 41* 46 42* * Background sound from the wind at times, traffic, and other sources was louder than equipment at GRL and GRL-A. There was a strong hum audible much of the time from the direction of Coors or Golden. The intent of this round of measurements was to get a full set of data after improvements have been made to equipment on GRL and GRLA. Unfortunately the background sound was louder on April 22 and makes direct comparison with the March 25 data (Table 1) difficult. In Table 2, the * indicates when the sound of wind, traffic, or equipment in the direction of Golden or Coors was more significant than sound from the GRL and GRLA equipment. The sound levels from GRL and GRLA equipment could not be determined (even though it may have been audible in some cases) in the presence of the background sound. We only know that the fan noise is no louder than the measured sound level, and in many cases is less than the measured level. The GRL Lab Exhaust fans are no longer the loudest source on the GRL roof. Sound from the condensing units and/or make-up air unit are louder than the fans. The GRLA General Exhaust fans are more noticeable at Location 2. The GRLA Lab Exhaust fans are more noticeable at Location 8.

10 Page 10 Noise Level Criteria and Assessment Colorado Revised Statute sets a limit of 50 dba at night at a location 25 beyond a residential property line. The Table 1 and Table 2 data show that the GRL and GRLA rooftop equipment did not exceed the state limit even before re-sheaving the GRL Lab Exhaust fans. Nonetheless, sound from the equipment on both buildings was clearly audible above background sound in the area on March 25, and the whine of the GRLA General Exhaust fan motor was distinctly audible and drew specific complaints from some neighbors. Since then, the whine has been mitigated and is no longer noticeable. The April 22, 2016 measurements continue to show that the GRL and GRLA rooftop equipment does not exceed the state limit. Potential for Additional Mitigation Two of the most significant noise sources have been mitigated. The remaining GRL and GRLA noise sources contributing to noise levels in the area include the GRL Make-up Air unit and condensing units, and the GRLA General and Lab Exhaust fans (fan noise as opposed to motor noise). The cooling tower on the GRLA may be contribute to a lesser extent. Exhaust Fan Stack silencers The GRLA General Exhaust and smaller Lab Exhaust fans produce noise, but the noise does not appear to be as annoying as the motor whine was. It is possible to reduce the General Exhaust and Lab Exhaust fan noise with stack silencers. Some manufacturers of these induced flow fans make Circular Dissipative silencers that are essentially a collar around the duct above the fan. Unfortunately this type of silencer is not available as an off the shelf item from Twin City Fans, the manufacturer of the GRLA General Exhaust and Lab Exhaust fans. We have received preliminary designs for inline stack silencers from Vibro-Acoustics. The silencers would be inserted between the fan discharge and the induced flow stack. They would raise the General Exhaust stack height by approximately 72ʺ, and the Lab Exhaust stacks by approximately 36ʺ. If you wish to pursue the stack silencers, a structural engineer will need to determine if additional structural support, guy wires, etc., are necessary to support the increased stack height. Fan noise could be reduced by 5 to 10 dba. Straight or elbow silencers may also be necessary the bypass damper intakes on each set of fans. Keep in mind that as the neighborhood sound level is reduced closer to the background sound level, further reduction will be less noticeable.

11 Page 11 Tall Screen Wall We evaluated the possibility of a screen wall around the GRLA rooftop equipment. A screen has to completely block the line-of-sight to reduce noise. Screens are more effective at mid and high frequencies than at low, and are most effective when placed close to the noise source. I understand that the GRLA building structure cannot accommodate the additional weight and wind load of a screen. A screen would have to be located at the edge of the roof along the northwest and northeast sides. It would have to be supported from the ground on a new structure. It would have to be approximately 50 to 55 feet tall (from ground level) to be effective. In theory a screen could reduce GRLA equipment noise in the neighborhood from 10 to 15 dba, but this much reduction would not be realized because noise from other sources would become more significant. For example, even if the fan noise is reduced by 10 to 15 dba, the noise level at Location 2 would only be reduced by about 3 dba (from 44 to 41 dba per Table 1) due to the background sound level. This is a just noticeable change. For your reference, a 2 to 3 dba change is just noticeable by most people. A 5 to 6 dba change is clearly noticeable, and a 10 dba change is perceived as half as loud. I have only evaluated the screen wall conceptually. At this point I have not tried to evaluate its feasibility, or if and how it would work architecturally and structurally. If feasible, it is sure to be quite expensive for the relatively small improvement to be gained. Conclusions Sound levels from GRL and GRLA rooftop equipment comply with State of Colorado limits at residential property lines. The previous source of noise complaints was most likely the GRL Lab Exhaust fans and the GRLA General Exhaust fans and motors. Sound from the GRL Lab Exhaust fans has been greatly reduced by re-sheaving and slowing the fans. The whine of the GRLA General Exhaust fan motors has been reduced or eliminated. Sound from the GRLA General Exhaust and Lab Exhaust fans is still contributing to noise levels that are likely above the nighttime background sound at some locations in the neighborhood. There are times, even at night, when the background sound is too loud measure the fan noise. The GRL condensing units and MAU may also be contributing to noise in the area. Additional mitigation measures could include silencers on the exhaust fans or a tall screen wall around two sides of the GRLA building.

12 Page 12 I am available to meet or discuss this further by phone if you wish. Please let me know if you would like to proceed with design or investigation of additional noise mitigation. Sincerely, Jeff Kwolkoski, P.E., INCE Bd. Cert. President cc: Mike Bowker, CSM

FC Cincinnati Stadium Environmental Noise Model

Preliminary Report of Noise Impacts at Cincinnati Music Hall Resulting From The FC Cincinnati Stadium Environmental Noise Model Prepared for: CINCINNATI ARTS ASSOCIATION Cincinnati, Ohio CINCINNATI SYMPHONY