FINAL REPORT COMPARATIVE ROOF TESTING AT ONONDAGA COUNTY CORRECTIONAL FACILITY Final October 2011 Submitted to: Onondaga County Dept. of Correction 6660 East Seneca Turnpike Jamesville, NY 13078-0143 Submitted by: Ashley-McGraw Architects, PC 500 South Salina St, Suite 1100 Syracuse, NY 13202 CDH Energy Corp PO Box 641 2695 Bingley Rd Cazenovia, NY 13035 315-655-1063
44
Embed
ashley roof final report-Oct 2011 - CDH Energycdhenergy.com/presentations/ashley roof final report-Oct 2011.pdf · Onondaga County extended the design contract with Ashley McGraw
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
FINAL REPORT
COMPARATIVE ROOF TESTING AT ONONDAGA COUNTY CORRECTIONAL FACILITY
Final
October 2011
Submitted to:
Onondaga County Dept. of Correction
6660 East Seneca Turnpike Jamesville, NY 13078-0143
Submitted by:
Ashley-McGraw Architects, PC 500 South Salina St, Suite 1100
Syracuse, NY 13202
CDH Energy Corp PO Box 641
2695 Bingley Rd Cazenovia, NY 13035
315-655-1063
EXECUTIVE SUMMARY
Onondaga County sought to evaluate the energy and water retention performance of green or vegetative roofing systems relative to other conventional and energy-efficient roofing systems. A major roof replacement project on multiple buildings at the Jamesville Correctional Facility offered the opportunity for a side-by-side test to evaluate different roofing systems. Monitoring equipment and instrumentation were installed to measure the performance of the different systems. The test considered four different roofing systems:
1. A conventional roof with 4 inches of foam insulation and a black Ethylene Propylene Diene Monomer (EPDM) membrane
2. A roof with 4 inches of foam insulation with a white Thermoplastic Polyolefin (TPO) roof membrane.
3. A vegetative roof with 4 inches of foam insulation 4. A highly-insulated roof with 8 inches of foam insulation and a TPO roof membrane.
Onondaga County extended the design contract with Ashley McGraw Architects to complete this testing. CDH Energy was hired to develop and implement a monitoring approach to quantify compare the performance of the four roofing systems. Temperature sensors and other instrumentation were installed in the roof assembly during construction in the Summer and Fall of 2009. Continuous data collection at 15-minute intervals has continued since October 2009 to obtain performance data to assess performance of the different roofing systems for all seasons of the year. The measured results showed that the TPO and vegetative roof systems had much lower roof temperatures than the conventional EPDM surface. The reduction in solar absorption reduced solar gains in the summer but also increased heat losses during the heating season. Compared to the EPDM membrane, the TPO roof had 30% higher heating losses and the vegetative roof had 23% higher losses. The TPO roof with extra insulation did have lower heating losses than the EPDM roof. Overall the TPO roof was cost neutral compared to the EPDM roof when both heating and cooling losses are considered. The vegetative roof had net cost savings of $7 per year per 1000 sq ft of roof area. The vegetative roof retained a significant amount of the rainfall across the year. On an annual basis only of about 20% of the measured rainfall was sent into the storm drain system.
TABLE OF CONTENTS Introduction ..................................................................................................................................... 1 Description of Roofs ....................................................................................................................... 2 Monitoring Approach...................................................................................................................... 5
Instrumentation ........................................................................................................................... 5 Major Events During Monitoring Period .................................................................................. 12
Measured Results .......................................................................................................................... 13 Roof Thermal Performance ....................................................................................................... 13 Water Retention and Drainage .................................................................................................. 27
Appendix A – Monitoring System Details Appendix B – Comparison of Rainfall Data from Various Local Weather Stations
LIST OF FIGURES
Figure 1. Aerial View of the Four Units at Jamesville Facility ..................................................... 2 Figure 2. Description of the Green Roof on Unit 3 (Roof Garden System from Carlisle) ............. 3 Figure 3. Photos of Roofs at Onondaga County Correctional Facility (before installation) ......... 3 Figure 4. Photos of Roofs at Onondaga County Correctional Facility (after installation) ............ 4 Figure 5. Thermocouple installations TRI and TRO ..................................................................... 6 Figure 6. Thermocouple Installation TAI ....................................................................................... 6 Figure 7. Detailed Drawing of Thermocouple Locations at each Station (the colored lines
indicate how the thermocouple wires are routed through assembly and back to the datalogger) .............................................................................................................................. 7
Figure 8. Unit 1 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position) ....... 8 Figure 9. Unit 2 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position) ....... 8 Figure 10. Unit 3 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position) ..... 9 Figure 11. Unit 4 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position) ..... 9 Figure 12. Vegetative Roof Mockup with Rain Gauge to Measure Water Drainage ................... 11 Figure 13. Green Roof Mockup on Unit 2 ................................................................................... 12 Figure 14. November 9, 2009, Temperature Profiles, Insolation Profile, and Resulting Heat Loss
Profile .................................................................................................................................... 14 Figure 15. Comparing Heat Transfer Rates for A (solid) and B (dotted) Locations on Each Unit
............................................................................................................................................... 15 Figure 16. Roof Temperature (TRO) and Heat Loss Profiles for Summer Conditions .............. 17 Figure 17. Roof Temperature (TRO) and Heat Loss Profiles for Fall Conditions ...................... 18 Figure 18. Roof Temperature (TRO) and Heat Loss Profiles for Winter Conditions ................. 19 Figure 19. Roof Temperature (TRO) and Heat Loss Profiles for Spring Conditions ................. 20 Figure 20. Daily Heat Loss Compared to Daily Outside Temperature ........................................ 22 Figure 21. Plot of Monthly Heat Transfer with Four Roofing Systems ...................................... 24 Figure 22. Temperature Profiles, Heat Loss Profiles, and Rainfall/Drainage for December 2010
............................................................................................................................................... 25 Figure 23. Monthly Drainage Rate versus Rainfall (WUG data, Airport) .................................... 28 Figure 24. Percentage of Monthly Drainage/Rainfall (WUG) Compared to Insolation .............. 29 Figure 25. Impact of Rainfall on Measured Soil Moisture Content ............................................ 30 Figure 26. Comparing the Impact of Roof Moisture Content on Roof Temperatures ................. 31
LIST OF TABLES Table 1. Construction Details for the Roofs on Each Unit ............................................................. 2 Table 2 Instrumentation Installed for EACH Measurement Station .............................................. 6 Table 3. Instrumentation for Additional Measurements .............................................................. 10 Table 4. Summary of Major Events During Monitoring Period .................................................. 12 Table 5. Summary Days Included in the Plots Below ................................................................. 16 Table 6. Monthly Heat Loss Rate ................................................................................................ 23 Table 7. Annual Heating Load and Costs for Each Unit .............................................................. 26 Table 8. Monthly Rainfall and Drainage Data Along with Weather Conditions ......................... 27
Introduction
CDH Energy Corp. 1 October 2011
Introduction Onondaga County sought to evaluate the energy and water retention performance of green or vegetative roofing systems relative to other conventional and energy-efficient roofing options. A major roof replacement project on multiple buildings at the Jamesville Correctional Facility offered the opportunity for a side-by-side test to evaluate different roofing systems. The results of this testing are intended to provide technical feedback and guidance to inform the county’s decision making process for future roofing renovations for the all buildings across the county. For this test, four different roofing systems were installed:
1. A conventional roof with 4 inches of polyisocyanurate foam insulation with a black EPDM single ply-rubber roof membrane.
2. A conventional roof with 4 inches of poly-iso insulation with a white TPO roof membrane.
3. A vegetated roof with 4 inches of poly-iso insulation underneath it. 4. A highly-insulated roof with 8 inches of poly-iso insulation with a white TPO roof
membrane. A side-by-side test of these four roofing systems provided the means for thermal performance to be quantitatively assessed for:
• White TPO vs. conventional EPDM (1 vs. 2) • 8 inches vs. 4 inches of insulation (2 vs. 4) • vegetated vs. non-vegetative roof systems (3 vs. 1 or 2)
CDH Energy was contracted by Ashley-McGraw Architects to develop and implement a monitoring approach to quantify and compare the performance of the four roofing systems. Temperature sensors and other instrumentation were installed during construction in the Summer and Fall of 2009. The data collection system was fully vetted and commissioned by late 2009. Continuous data collection at 15-minute intervals has continued since then to obtain performance data to assess performance of the different roofing systems for all seasons of the year. During the monitoring period, CDH Energy has also posted the data to a website where County staff could review plots and tables summarizing the collected data. The database and website was updated nightly with the newest data throughout the monitoring period. The website is available at: www.cdhenergy.com/dataaccess.php (Click on “Comparative Roof Testing at Onondaga County Detention Facility”).
Description of Roofs
CDH Energy Corp. 2 October 2011
Description of Roofs The four buildings or Units at Jamesville that were included in this test program are shown in Figure 1 below. Each Unit had a different roofing system installed, as described in Table 1. All of the roofing systems included a ½-inch layer of Georgia Pacific DensDeck™ fiberglass-reinforced gypsum board between the insulation and the roof membrane. In each case the insulation was secured to the roof using adhesive foam.
Figure 1. Aerial View of the Four Units at Jamesville Facility
Table 1. Construction Details for the Roofs on Each Unit
Notes: 1- Polyisocyanurate foam board applied in 2-inch layers 2- Black EPDM (Ethylene Propylene Diene Monomer) single-ply rubber roof membrane
3- White TPO (Thermoplastic Polyolefin) roof membrane
Unit 1
Unit 2
Unit 3
Unit 4
Description of Roofs
CDH Energy Corp. 3 October 2011
The vegetated roofing system was the Roof Garden System by Carlisle. The assembly includes a drainage board on top of the membrane followed by a moisture retention mat and 2-3 inches of small aggregate. The 12 inch by 15 inch sedum tiles are place on top of the aggregate. The drainage board includes plastic cavities or cups to retain water.
Figure 2. Description of the Green Roof on Unit 3 (Roof Garden System from Carlisle)
Figure 3 shows photos of the roofs before installation and Figure 4 shows the new roofing systems.
View from Unit 3 looking towards Unit 1
View from Unit 2 looking towards Unit 3
Figure 3. Photos of Roofs at Onondaga County Correctional Facility (before installation)
Description of Roofs
CDH Energy Corp. 4 October 2011
View from Unit 3 (vegetative) looking towards Unit 2 (TPO)
View from Unit 4 looking towards Unit 3
Figure 4. Photos of Roofs at Onondaga County Correctional Facility (after installation)
Monitoring Approach
CDH Energy Corp. 5 October 2011
Monitoring Approach Several approaches to quantifying the energy impact of the different roof systems were considered including measuring the heating energy use of the HVAC system before and after retrofit. Ultimately the approach of measuring the temperature differences within the roof assembly was ultimately selected as most compatible with the project schedule, building configuration, and limited access inside the facility. Two independent monitoring stations were installed on each roof, for a total of eight stations. Each station used a Campbell Scientific data logger. The eight loggers were located on top of the roof, space out over several hundred yards. A mix of hardwired and wireless networking was used to connect the loggers. Communications outside of the building was provided by a phone modem link. Each monitoring station was based around a Campbell Scientific CR800 or CR1000 data logger (Station 2A uses a CR1000 to accommodate the extra data points). The data loggers were programmed to sample all sensors once per second. Calculated averages and totals were recorded for each 15-minute interval. After all records were created at each station, the data logger located at 3A collected each record from all the other data loggers. That master data logger was called and data was downloaded each night by phone modem. The data was loaded into a database at CDH Energy for automatic verification, processing and display on the web. Appendix A provides more details on the monitoring system. The rational for installing these points is given below.
Instrumentation The overall experimental approach was to measure and compare the temperatures across the assembly for the different roofing systems in a side-by-side test. The heat transfer through the roof surface is proportional to the temperature difference through each layer. Since all the roof systems are exposed to the same ambient conditions, as well as similar indoor temperatures, the performance of the different systems can be directly compared at each time. At each monitoring station, three temperatures are collected. The top point is the roof temperature above the insulation and below the DensDeck™ (TRO). The middle point is under the insulation but above the deck (TRI). The third point is indoor air temperature measured just below the ceiling in the space below (TAI). These points are compared to the outdoor air temperature (TAO) which is measured at one location. The measurements listed in Table 2 are taken at two separate locations (A & B) on each Unit (1, 2, 3, & 4) for a total of 8 locations. Figure 5 and Figure 6 show where the thermocouples were installed at each station. Figure 7 schematically shows the locations of each sensor through roof assembly. The indoor temperature sensor was difficult to fish through the roof and into the space below. However, we were able to get at least one sensor into the space for each unit.
Monitoring Approach
CDH Energy Corp. 6 October 2011
Table 2 Instrumentation Installed for EACH Measurement Station
Point Description Instrument Eng Units
Locations
TRO Roof Temperature (on top of insulation, under roof brd)
Type-T Thermocouple
ºF At each station
TRI Roof Temperature (under roof insulation, above deck)
Type-T Thermocouple
ºF At each station
TAI Indoor Temperature (just below the roof)
Type-T Thermocouple
ºF At each station
Thermocouple installed on top of Roof Deck (TRI ) - Before Insulation is Installed
Thermocouple installed above the insulation and below the dense deck for location 1B (TRO)
Figure 5. Thermocouple installations TRI and TRO
Thermocouple installed just below the ceiling for Location 4B
Thermocouple installed just below the ceiling for Location 4A
Figure 6. Thermocouple Installation TAI
Insulation Board
Monitoring Approach
CDH Energy Corp. 7 October 2011
Figure 7. Detailed Drawing of Thermocouple Locations at each Station (the colored lines indicate how the thermocouple wires are routed through assembly and back to the datalogger)
Monitoring Approach
CDH Energy Corp. 8 October 2011
Figure 8. Unit 1 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position)
Figure 9. Unit 2 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position)
Monitoring Approach
CDH Energy Corp. 9 October 2011
Figure 10. Unit 3 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position)
Figure 11. Unit 4 Station Positions (‘O’ is the datalogger, ‘X’ is the thermocouple position)
Monitoring Approach
CDH Energy Corp. 10 October 2011
One of the stations (2A) also included additional measurements for the overall site. Table 3 lists these additional measurements. A weather station was installed to measure ambient temperature (TAO), horizontal solar flux (ISH) and rainfall (RAIN). Table 3. Instrumentation for Additional Measurements
Point Description Instrument Eng Units
Location
TAO Outdoor Temperature Type-T Thermocouple
ºF Station 2A
ISH Solar Flux or Insolation (horizontal)
Licor LI200x W/m2 Station 2A
TGR Green Roof Temperature (in middle of soil layer)
Type-T Thermocouple
ºF Station 3A
MGR Green Roof Moisture Content (in middle of soil layer)
Campbell Scientific CS616
0-1 Station 3A
RAIN Rainfall Texas Electronics 525
Inches Station 2A
WF Water Flow from Green Roof Mockup
Texas Electronics 525
Gal/h Station 2A
A 4 ft by 4 ft mockup of the vegetative roof system was also created and located on the roof of Unit 2. The purpose of this mockup was to provide the means to directly measure the water retention of the vegetative roof assembly. Figure 12 shows the mockup with the instrumentation added to measure its water retention performance. The instrumentation included a rain gauge (RAIN) and as well as a modified rain gauge to measure the water flow down the drain of the roof assembly (WF). The rain gauge included an electric heater so it could provide a reading of snowfall as well as rain. Figure 13 schematically shows the analysis approach. The rainfall data was compared to the total water flow from the drain of the 4 ft by 4 ft mockup. Comparing these values provided a direct measurement of the moisture holding ability of the green roof surface.
Monitoring Approach
CDH Energy Corp. 11 October 2011
The mockup of the vegetative roof assembly with rain gauge underneath
The drain gauge (WF) underneath the roof mockup
Figure 12. Vegetative Roof Mockup with Rain Gauge to Measure Water Drainage
WF
Monitoring Approach
CDH Energy Corp. 12 October 2011
WF
RAIN Green Roof Mockup Rain
Gauge
Roof of Unit 2
Figure 13. Green Roof Mockup on Unit 2
In addition, monitoring station 3A has an extra thermocouple that is embedded in the “soil” layer of the vegetative roof (TGR). This location also has a water content reflectometer embedded within the vegetative assembly to measure the moisture content of the “soil” (MGR). These sensors were installed to provide further performance information about the vegetative roof.
Major Events During Monitoring Period Table 4. Summary of Major Events During Monitoring Period
May 2009 • Installed embedded thermocouples for 1A • Installed embedded thermocouples for 1B
June 2009 • Installed embedded thermocouples for 2A • Installed embedded thermocouples for 2B • Installed embedded thermocouples for 3A • Installed embedded thermocouples for 3B • Installed embedded thermocouples for 4A • Installed embedded thermocouples for 4B
September 2009 • Vegetative roof was installed
October 2009 • Installed data loggers for each monitoring station • Mockup sensors installed • Indoor temperature sensors installed • Data collection begins
November 2009 • The phone line for data collection operational • The conduit for the pyranometer had been tilting; it was been straightened and reinforced.
March 2010 • The collector on the rain gauge was knocked off the top of the device over the winter (precise date
unknown) The top was replaced.
Measured Results
CDH Energy Corp. 13 October 2011
Measured Results The data collection system at the site was fully commissioned starting on October 22, 2009. This section analyzes the data collected from November 1, 2009 through March 31, 2011.
Roof Thermal Performance The calculation to determine the heat loss for each roof is made by determining the temperature difference across the layers of insulation and dividing by the rated R-value for the installed insulation:
TAO
TRO
TRI
TAI
Rinsulation q Insulation
Board
Concrete Deck
Veg roof (optional)
Deck Board membrane
Where: Rinsulation - R-value for Insulation layer (ºF-h-ft2/Btu) q - Heat flux through the roof assembly (Btu/h-ft2)
q = (TRI-TRO) Rinsulatiion Where q is defined as positive for heat loss from the space through the roof to ambient. On each roof, a different amount of insulation board (with an R-value of 5.7 ft2-h-°F/Btu per inch) was installed on each Unit. The resulting R-values are given in Table 1. Because the insulation board has very low thermal mass, the calculated heat loss values for each 15-minute interval determined by this method are expected to provide a representative estimate of the dynamic (or time-varying) heat transfer through the roof. Figure 14 shows the temperatures, insolation, and resulting roof heat flux for the four different roofs during a sunny Fall day (November 9, 2009). The top of the insulation for the EPDM membrane is at a much warmer temperature on this day since more heat is absorbed from the sun. The TPO roofing surface significantly reduces this impact as expected. The vegetative roof provides thermal mass that smoothes out the fluctuations across the daily cycle.
Figure 14. November 9, 2009, Temperature Profiles, Insolation Profile, and Resulting Heat Loss Profile
Ambient Temperature, Top of Insulation, Bottom of Insulation, Indoor Temperature
Unit 1: 4 in, EPDM Unit 2: 4 in, TPO
Unit 3: 4 in, Veg Unit 4: 8 in, TPO
Measured Results
CDH Energy Corp. 15 October 2011
Figure 15 shows that the heat transfer performance was very similar for the two locations or stations on each Unit. The solid lines correspond to station A while the dotted lines correspond to station B. Stations A and B generally showed very similar responses for each different roofing system. Therefore the plots beyond this point in the report use the average heat transfer rates for the A and B locations.
Unit 1: 4 in, EPDMUnit 2: 4 in, TPOUnit 3: 4 in, VegUnit 4: 8 in, TPO
Figure 15. Comparing Heat Transfer Rates for A (solid) and B (dotted) Locations on Each Unit
The series of plots shown in Figure 16 through Figure 19 below compare the performance of the four roofing systems in the various seasons. In each case, specific days were selected to represent or highlight a condition common to that season.
Measured Results
CDH Energy Corp. 16 October 2011
Table 5. Summary Days Included in the Plots Below
Season Date Condition
Summer, Figure 16 May 29, 2010 Summer, Sunny, 71F May 16, 2010 Summer, Sunny, 59F May 12, 2010 Summer, Cloudy, 44F
Fall, Figure 17 November 2, 2010 Fall, Sunny, 38F November 9, 2010 Fall Cloudy, 40F
Winter, Figure 18 January 30, 2010 Winter, Sunny 2F, after notable melt January 24, 2011 Winter, Sunny 3F, min temp -13F January 19, 2011 Winter, Cloudy, 25F
Spring, Figure 19 April 17, 2010 Spring, Morning, 41F April 14, 2010 Spring, Sunny, 49F
Summer Days Figure 16 compares roof temperatures (i.e., at top of insulation) and heat loss rates for three different summer days. On the sunny days the roof temperature for the EPDM roof is more than 50ºF hotter than the TPO surface. The vegetative roof was even 20ºF cooler than the TPO surface presumably due to evaporation at the surface. The thermal mass of the vegetative assembly above the temperature sensor (TRO) also mitigates heat loss and results in much less variation in both temperature and heat loss across the day. The heat gain (or negative heat loss) with the EPDM surface on the sunny days is considerably greater than for the other roof systems. The EPDM roof has a summer time heat gain that is greater by 2-4 Btu/h per square foot (or about 0.2-0.3 tons per 1000 sq ft) than the other roofing systems. Adding the 4 inches of insulation with the TPO membrane reduces the peak cooling load by about 0.5 Btu/h per square foot (or about 0.04 tons per 1000 sq ft). Fall Days Figure 17 shows the profiles for two fall days. Even on cloudy days the heat gain with the EPDM roof increases surface temperatures by 20-30°F. Winter Days Figure 18 shows profiles for several winter days. The temperature just under vegetative roof stays in 30-40F range regardless of ambient conditions, presumably because of its thermal mass and its ability to retain snow cover. The cloudy winter day with snow cover showed the temperature just under the vegetative roof remaining near 40F while the other roofs have surface temperatures very near the freezing mark – as would be expected for a snow-covered roof. This implies the vegetative layer is some thermal resistance by raising the freezing layer above the roof surface. Spring Days Finally, Figure 19 shows profiles for some spring days. The impact of a rain event on this plot is apparent for one of the days.
Figure 19. Roof Temperature (TRO) and Heat Loss Profiles for Spring Conditions
Unit 1: 4 in, EPDM, Unit 2: 4 in, TPO, Unit 3: 4 in, Veg, Unit 4: 8 in, TPO
Rain
Measured Results
CDH Energy Corp. 21 October 2011
Figure 20 plots the integrated heat loss for each day versus the daily ambient temperature. Each symbol corresponds to a day and each roof system is shown as a different color. Each daily value is average of the A and B locations on each roof. All the roof systems show the expected trend of more heat loss at lower outdoor temperatures. The EPDM roof on Unit 1 with 4 inches of insulation has the most heat gain the summer and nearly the highest amount of heat loss in the winter. The TPO roof on Unit 2 with 4 in of insulation has less heat gain the summer – as expected. However the TPO roof also shows more heat loss in the winter, presumably when the roof is clear of snow. The TPO roof with extra insulation also shows the expected reduction in heat transfer. All the roof surfaces show the impact of snow on the roof in the winter – though this effect is most pronounced for the vegetative roof. The heat loss flattens out once the ambient temperature drops below 35°F or so on some days because the phase change associated with snow on the roof tends to hold the roof surface at a constant temperature near the freezing point. There is more scatter at these temperatures since the heat transfer performance is much different whether the roof surface is snow covered or exposed. The highly-insulated TPO roof shows less variation and a more linear pattern because the phase change impact of freezing and thawing is not apparent until much lower ambient temperatures are reached.
Measured Results
CDH Energy Corp. 22 October 2011
0 20 40 60 80 100Outdoor Temperature (F)
-40
-20
0
20
40
60
Hea
t Los
s (B
tu/ft
^2-d
ay)
Unit 1: 4 in, EPDMUnit 2: 4 in, TPOUnit 3: 4 in, VegUnit 4: 8 in, TPO
Figure 20. Daily Heat Loss Compared to Daily Outside Temperature
Heat Loss
Heat Gain
Impact of Snow Cover
Measured Results
CDH Energy Corp. 23 October 2011
Table 6 shows the result of integrating the heat loss over each month for the four roofing systems. Figure 21 shows a plot of this data for 2010. From a heating season perspective, the EPDM membrane provides a benefit of lower overall heat loss in the winter (October to April) because of the solar gain. The heat loss with the TPO and vegetative roof are actually 30% and 23% greater when using the same insulation level. Adding 4 inches of insulation with a TPO surface reduces the heat loss by 11% compared to the conventional roof. In the cooling season1 the TPO and vegetative roofs reduce the heat gain to the roof compared to the EPDM roof. The added roof insulation has only modest impact on the heat gain compared once a TPO membrane has been used (the impact of insulation would have much greater had insulation been added to the EPDM surface).
2010 Clg Season (1,188.5) 642.2 1,141.3 698.0 (May to Sep) 1,830.7 2,329.9 1,886.5
Heat Loss (Btu/ft^2)
1 While the correctional facility has does not have cooling, we completed an analysis to assess the impact of the roof systems assuming the facility did have cooling.
Measured Results
CDH Energy Corp. 24 October 2011
(600)
(400)
(200)
-
200
400
600
800
1,000
1,200
1,400
Jan-
10Fe
b-10
Mar
-10
Apr-1
0M
ay-1
0
Jun-
10
Jul-1
0Au
g-10
Sep-
10
Oct
-10
Nov-
10De
c-10
Hea
t Los
s (B
tu/ft
^2)
Unit 1: 4 in EPDMUnit 2: 4 in TPOUnit 3: 4 in VegUnit 4: 8 in TPO
Figure 21. Plot of Monthly Heat Transfer with Four Roofing Systems
The impact of snow cover and the associated freezing and thawing at the roof surface is shown by Figure 22. On December 1, it starting raining and the temperature dropped until the rain turned to snow. In this case the temperatures on top of the insulation reached freezing (32oF) and stayed there for several days. On December 4 through 9, there was precipitation but no drainage from the roof mockup (indicating snow). The surface of all the EPDM and TPO roofs stayed near 32oF, indicating freezing at the roof surface. The temperature probe under the vegetative roof never reached freezing and started to get warmer as the snow started to build up and the snow provided an insulating layer. The temperature inside the vegetative layer (green dotted) did reach freezing but increased as the snow layer built up) However, while this change in temperature is noticeable, the change in heat loss rate was modest and may even show a slight increase in heat loss from the roof.
Heat Loss
Heat Gain
Measured Results
CDH Energy Corp. 25 October 2011
30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15Nov Dec
0
20
40
60
Tem
pera
ture
(F)
30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15Nov Dec
0.00.5
1.0
1.5
2.0
2.53.0
Hea
t Los
s (B
tu/ft
^2-h
)
30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15December
0.00
0.05
0.10
0.15
0.20
0.25
Wat
er (i
nche
s)
Figure 22. Temperature Profiles, Heat Loss Profiles, and Rainfall/Drainage for December 2010
Rainfall Drainage
Unit 1: 4 in, EPDM, Unit 2: 4 in, TPO, Unit 3: 4 in, Veg, Unit 4: 8 in, TPO, Ambient Temperature
Rain Snow (no drainage)
Measured Results
CDH Energy Corp. 26 October 2011
Table 7 compares the overall heating and cooling costs for the different roofs. The EPDM roof is used as the reference or baseline; fuel use and cost savings are compared relative to that roof system. The TPO and vegetative roofs actually result in slightly higher fuel costs. Heating costs increase by $18 and $14 per year per 1000 sq ft assuming and 80% efficient heating system and a gas cost of $1.00 per therm. The TPO roof with extra insulation does save about $6 per year per 1000 sq ft. Comparing the results for Unit 2 and Unit 4 implies that the extra 4 inches of insulation saves about $24 per year per 1000 sq ft. Table 7. Annual Heating Load and Costs for Each Unit
Unit 1 4 in EPDM
Unit 2 4 in TPO
Unit 3 4 in Veg
Unit 4 8 in TPO
4.9 6.4 6.0 4.4 61.1 79.5 75.4 54.6
61$ 80$ 75$ 55$ (18)$ (14)$ 6$
Unit 1 4 in EPDM
Unit 2 4 in TPO
Unit 3 4 in Veg
Unit 4 8 in TPO
152.6 194.2 157.2 137.3 174.7 141.5
16$ 21$ 17$
Unit 1 4 in EPDM
Unit 2 4 in TPO
Unit 3 4 in Veg
Unit 4 8 in TPO
(2)$ 7$ 23$
COOLING
COMBINED
Savings per 1000 sq ft
Reduced Cooling (ton-hrs/yr per 1000 sq ft)Reduced Cooling Power (kWh/yr per 1000 sq ft)
NET Savings per 1000 sq ft
Annual Gas Use (therms per 1000 sq ft)Annual Cost per 1000 sq ft
Savings per 1000 sq ft
Annual Heat Load (MMBtu per 1000 sq ft)HEATING
Cooling costs are reduced by about $16 and $21 per year per 1000 sq ft for the TPO and vegetative roofs respectively, assuming an overall cooling efficiency of 0.9 kW per ton and an electric cost of $0.12 per kWh. Adding the extra 4 inches of insulation with the TPO roof had very little impact in the cooling season. The overall savings from the TPO roof are slightly negative, with the cooling and heating savings essentially canceling out. The vegetative roof does result in net savings of $7 per year per 1000 sq ft. The TPO roof with extra insulation provides the most savings ($23 per year per 1000 sq ft) since the TPO membrane helps in the summer and the extra insulation reduces heat loss in the winter.
Measured Results
CDH Energy Corp. 27 October 2011
Water Retention and Drainage The rain gauge under the roof mockup was calibrated to measure the volume of water draining from the 4 ft by 4 ft section. Dividing the volume of water by the area of the mockup provides the amount (and therefore the fraction) of total rainfall draining from the section. Drain (in) = water volume (in3) / area (in2) Table 8 shows the monthly rainfall and drainage data along with weather conditions for the site. The rainfall data in Table 8 is the Weather Underground (WUG) data from Syracuse Airport. We used this data because the rain gauge on the roof at the site malfunctioned for the part of the period (see Appendix B). For the 12 month period ending March 2011, 48.2 inches of rain were recorded at the airport and only 9.7 inches of that rainfall drained from the mockup roof (20%). Therefore on an annual basis the vegetative roof system retained about 80% of the rainfall. Table 8. Monthly Rainfall and Drainage Data Along with Weather Conditions
Figure 23 is a plot comparing the monthly rainfall and drainage. The data confirm the overall drainage rate was about 20%, but there is considerable variation from month-to-month. One of the factors thought to drive this variation is the amount of solar energy hitting the roof surface. Solar radiation causes more evaporation and decreases the drainage rate.
0 2 4 6 8Rainfall (in)
0
2
4
6
8
Dra
inag
e (in
)
Figure 23. Monthly Drainage Rate versus Rainfall (WUG data, Airport)
Measured Results
CDH Energy Corp. 29 October 2011
Figure 24 compares the monthly drainage fraction to the monthly solar energy. There is a weak trend of more drainage in months with modest solar flux and less drainage with more solar flux.
0 50 100 150 200Insolation (kWh/m^2)
0
20
40
Dra
in/R
ain
(%)
Figure 24. Percentage of Monthly Drainage/Rainfall (WUG) Compared to Insolation
The moisture content of the roof was shown to have a modest impact on overall heat transfer. Figure 25 shows a two week period in July 2010 that occurred after a long dry period. The measured soil moisture content starts off fairly low at 5-10% before July 23 when a rain event occurs. This storm causes the moisture content of the vegetative assembly to reach 25%. A second rain event a couple days later pushes the moisture content above 25%.
Measured Results
CDH Energy Corp. 30 October 2011
18 19 20 21 22 23 24 25 26 27 28 29 30 31July
0.0
0.2
0.4P
reci
pita
tion
(in)
18 19 20 21 22 23 24 25 26 27 28 29 30 31July
05
10
15
20
2530
Soi
l Moi
stur
e C
onte
nt (%
)
Figure 25. Impact of Rainfall on Measured Soil Moisture Content
Figure 26 shows the impact of moisture content on roof heat transfer. The plots compares the data for July 20 (when the roof assembly was dry) and July 27 (when the roof was moist). These two days were selected for comparison since both the ambient temperature and solar radiation were similar. The moisture content is clearly higher on July 27 and the resulting temperature at the roof deck is 4-5ºF cooler.
Rainfall Drainage
Measured Results
CDH Energy Corp. 31 October 2011
22: 0: 2: 4: 6: 8: 10:12:14:16:18:20:22: 0:40
60
80
100
Am
bien
t Tem
pera
ture
(F)
22: 0: 2: 4: 6: 8: 10:12:14:16:18:20:22: 0:0.0
0.2
0.4
0.6
0.8
1.0
Inso
latio
n (W
/m^2
)
22: 0: 2: 4: 6: 8: 10:12:14:16:18:20:22: 0:0
5
10
15
20
25
Soi
l Moi
stur
e C
onte
nt (
%)
22: 0: 2: 4: 6: 8: 10:12:14:16:18:20:22: 0:60
65
70
75
80
Roo
f Tem
p (F
)
Figure 26. Comparing the Impact of Roof Moisture Content on Roof Temperatures
July 20, 2010 July 27, 2010
Conclusions
CDH Energy Corp. 32 October 2011
Conclusions The vegetative roof clearly retains water and minimizes the amount of rainwater that drains from the roof. Drainage into the storm water system is only about 20% of the rainfall on the roof. Some modest variation was noted due to the amount of solar radiation striking the roof surface: in the summer when the insolation is higher, the amount of water draining from the roof decreases slightly. The thermal performance of the four roof systems was different in summer and winter. The EPDM surface did result in roof temperatures that were as much as 50ºF higher than the other surfaces. This surface had higher heat gains in the summer but also more modest heat losses in the winter. The TPO membrane significantly reduced the surface temperatures in the summer but also resulted in greater heat losses in the heating season (since beneficial solar gains are reduced). The vegetative roof adds thermal mass to the roof assembly that dampens the temperature swings. Evaporation at the surface also provides cooling in the summer and swing seasons. The vegetative roof may have also retained more snow cover more often. Overall the TPO surface with 4 inches of insulation had 30% higher thermal losses over the heating season and increases heating costs by $18 per year per each 1000 sq ft of roof area. However, the reduced summer time heat gains equate to about $16 per year per 1000 sq ft in cooling energy savings. Overall, heating losses and cooling savings tended to cancel out. The vegetative roof added thermal mass though the loss of solar gains in the winter still resulted in 23% higher heating losses. The estimated increase in heating costs was $14 per year per 1000 sq ft. However, the reduced summer time heat gains equate to about $21 per year per 1000 sq ft in cooling energy savings. Overall the net heating and cooling savings are about $7 per year per 1000 sq ft. The TPO roof with an additional 4 inches of insulation had best thermal performance. The thermal losses from the roof in the heating season were reduced by 11%. Heating costs were reduced by $6 per year per each 1000 sq ft of roof area. The reduced summer time heat gains equate to about $17 per year per 1000 sq ft in cooling energy savings. Overall the combined savings are about $23 per year per 1000 sq ft. The impact of the additional 4 inches of insulation reduce the thermal losses by about 31% (comparing Unit 2 and Unit 4). These heating cost savings are $24 per year per 1000 sq ft. The insulation has only a modest impact on cooling costs when combined with the TPO membrane. The cooling savings from adding insulation with the EPDM membrane would be more significant.
Recommendations Onondaga County should consider using green roofs on their facilities when water retention is the primary objective. The vegetative roof was shown to have some thermal benefit, though
Conclusions
CDH Energy Corp. 33 October 2011
similar thermal performance probably can be achieved more cost effectively by using a TPO membrane and/or adding additional insulation. The TPO membrane is energy neutral in the Central New York climate. The reduction in cooling energy use and peak cooling load is offset by the increase in thermal losses during the heating season. If TPO roofs are considered, insulation should be added to reduce heating energy use in the winter.
Appendix A A-1 July 2011
Appendix A Monitoring System Details
Watlow Gordon AFEC0TA060U8200 Thermocouples
The enclosure, data-logger, MD485, and battery for Location 1B
Conduit after installation of insulation & board (Location 1B)
Conduit mounted to roof deck at each station
Appendix A A-2 July 2011
Finished conduit assembly with boot and thermocouple wires shown (Station 2B)
Finished monitoring station (2A) and weather station (solar radiation and ambient temperature).
Position of the CS616 sensor from 3A
Appendix A A-3 July 2011
Network map Location Datalogger type Pak Bus Address Points1A CR800 11 Roof Temperatures and indoor
temperature4A CR800 41 Roof Temperatures and indoor
temperature4B CR800 42 Roof Temperatures and indoor
temperature
Appendix A A-4 July 2011
Database Setup Point Name Description Unit of measureInstrumentLID_1A Logger ID Number = 11BATTV_1A Battery Voltage Volts From dataloggerRT_1A Reference Temperature F From dataloggerTRO_1A Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_1A Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_1A Indoor Temperature F Watlow Gordon AFEC0TA060U8200 ThermocouplesLID_1B Logger ID Number = 12BATTV_1B Battery Voltage Volts From dataloggerRT_1B Reference Temperature F From dataloggerTRO_1B Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_1B Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_1B Indoor Temperature F Watlow Gordon AFEC0TA060U8200 ThermocouplesLID_2A Logger ID Number = 21BATTV_2A Battery Voltage Volts From dataloggerRT_2A Reference Temperature F From dataloggerTRO_2A Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_2A Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_2A Indoor Temperature F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAO Outdoor Temperature F Watlow Gordon AFEC0TA060U8200 ThermocouplesISH Solar Insulation kW/m^2 Campbell Scientific LI200X-L10 PyranometerTSF Total Flux MJ/m^s Campbell Scientific LI200X-L10 PyranometerRAIN Rainfall Inches Texas Electronics TR-525USW Tipping BucketWF Water Flow (Model) Inches Hydrolynx 5050 Tipping bucketLID_2B Logger ID Number = 22BATTV_2B Battery Voltage Volts From dataloggerRT_2B Reference Temperature F From dataloggerTRO_2B Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_2B Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_2B Indoor Temperature(bad) F Watlow Gordon AFEC0TA060U8200 ThermocouplesLID_3A Logger ID Number = 31 % BATTV_3A Battery Voltage Volts From dataloggerRT_3A Reference Temperature F From dataloggerTRO_3A Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_3A Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_3A Indoor Temperature (bad) F Watlow Gordon AFEC0TA060U8200 ThermocouplesTGR Soil Temperature F Watlow Gordon AFEC0TA060U8200 ThermocouplesMGR Soil Moisture Content % Campbell Scientific CS616 Water Content ReflectometerMPA Period Average uSec Campbell Scientific CS616 Water Content ReflectometerLID_3B Logger ID Number = 32BATTV_3B Battery Voltage Volts From dataloggerRT_3B Reference Temperature F From dataloggerTRO_3B Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_3B Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_3B Indoor Temperature F Watlow Gordon AFEC0TA060U8200 ThermocouplesLID_4A Logger ID Number = 41BATTV_4A Battery Voltage Volts From dataloggerRT_4A Reference Temperature F From dataloggerTRO_4A Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_4A Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_4A Indoor Temperature F Watlow Gordon AFEC0TA060U8200 ThermocouplesLID_4B Logger ID Number = 42 BATTV_4B Battery Voltage Volts From dataloggerRT_4B Reference Temperature F From dataloggerTRO_4B Top of insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTRI_4B Under insulation F Watlow Gordon AFEC0TA060U8200 ThermocouplesTAI_4B Indoor Temperature F Watlow Gordon AFEC0TA060U8200 Thermocouples
Appendix A A-5 July 2011
Instrumentation Information Instrument Output Multiplier and
OffsetNotes
Watlow Gordon AFEC0TA060U8200 Thermocouples
Type T output*9./5. + 32 converts from C to F, The thermocouple wires with a red tag are TRI while a blue tag indicates TRO
Campbell Scientific LI200X-L10 Pyranometer
mV 0.2 kW m-2 mV-1
Texas Electronics TR-525USW Tipping Bucket
Pulse 1/25.4 Converts from mm to inches of rain
Hydrolynx 5050 Tipping bucket Pulse 4.455/482 Converts from pulses to inches of rain using the calibrated value of 73ml/tip: volume (in^3) / area (in^2) = inches
Campbell Scientific CS616 Water Content Reflectometer
square wave with frequency depenent on water content
1 Volumetric water content = -0.0663+(-0.0063*PA)+(0.0007*PA) where PA = period average of the output
Appendix B B-1 July 2011
Appendix B Comparing Rainfall Data From Various Sources A high-quality rain gauge with an electric heater (to convert snow into water) was installed at the site to measure the amount of rainfall. Sometime in the winter of 2009-2010 the top of the rain gauge was blown off. The problem was found and corrected in March 2010. In order to assess when the data collected by the rain gauge was correct, we compared it to the other sources of rainfall data from area, including:
1. Weather Underground (WUG) data from Hancock International Airport and, 2. The weather station at the Syracuse Center of Excellence in downtown Syracuse.
The two sources of weather data are compared to the rooftop rain gauge at the Jamesville in Table B-1 Some discrepancy might be expected since the location of these gauges are different by a few miles. The data from the period of December 2009 through March 2010 for roof gauge is not reliable as discussed above. The COE data also had a hole in the data when the tower was moved in mid 2010.. Table B-1. Comparison of Jamesville Roof Rain Gauge to Other Sources
Figure B-1 shows a monthly comparison between the CDH data and the WUG data for the months after the rain gauge was fixed. During most months the two data sources are in good agreement, with the exception of the winter periods when the CDH rain gauge has failed.
Appendix B B-2 July 2011
0 2 4 6 8Jamesville Roof Rain (in)
0
2
4
6
8W
UG
-Airp
ort R
ain
(in)
Figure B-1. Comparison between Rainfall Sources
Therefore, the data from Weather Underground is used as the primary source in the main report.