2.2 Field Monitoring

• Air Handler Study

To determine the impact of air handler location on heating and cooling energy use, researchers measured the amount of air leakage in air handler cabinets, and between the air handler cabinet and the return and supply plenums.  To assess this leakage, testing was performed on 69 air conditioning systems. Thirty systems were tested in the 2001 and 39 in 2002.  The 69 systems were tested in 63 Florida houses (in six cases, two air handlers were tested in a single house) located in seven counties across the state - four in Leon County in or near Tallahassee, 17 in Polk County, three in Lake County, 13 in Orange County, one in Osceola County, two in Sumter County, and 29 in Brevard County.  All except those in Leon County are located in central Florida.  Construction on all houses was completed after January 1, 2001, and most homes were tested within four months of occupancy.

In each case, air leakage (Q₂₅) at the air handler and two adjacent connections was measured.  Q₂₅ is the amount of air leakage which occurs when the ductwork or air handler is placed under 25 Pa of pressure with respect to its surrounding environment.  Q₂₅ also can be considered a measurement of ductwork perforation.

To obtain actual air leakage while the system operated, it was necessary to measure the operating pressure differential between the inside and outside of the air handler and adjacent connections.  In other words, it was necessary to know the perforation or hole size and the pressure differential operating across that hole.  By determining both Q₂₅ and operating pressure differentials, actual air leakage into or out of the system was calculated.

Field Testing Leakage Parameters:  Testing was performed on 69 air conditioning systems to determine the extent of air leakage from air handlers and adjacent connections. Testing and inspection was performed to obtain:

• Q₂₅ in the air handler, Q₂₅ at the connection to the return plenum, and Q₂₅ at the connection to the supply plenum.
• Operating pressure at four locations - the return plenum connection, in the air handler before the coil, in the air handler after the coil, and at the supply plenum connection.
• Return and supply air flows were measured with a flow hood.  Air handler flow rates were measured with an air handler flow plate device (per ASHRAE Standard 152P methodology).
• Overall duct system and house airtightness in 20 of the 69 homes.
• Cooling and heating system capacity based on air handler and outdoor unit model numbers.
• The location and type of filter.
• Dimensions and surface area of the air handler cabinet.
• The fractions of the air handler under negative pressure and under positive pressure.
• The types of sealants used at air handler connections.
• Estimated portion of the air handler leak area that was sealed "as found."

Air Handler Leakage:  Leakage in the air handler cabinet averaged 20.4 Q₂₅ in 69 air conditioning systems. Leakage at the return and supply plenum connections averaged 3.9 and 1.6 Q₂₅, respectively. Using the operating pressures in the air handler and at the plenum connections, these Q₂₅ results convert to actual air leakage of 58.8 CFM on the return side (negative pressure side) and 9.3 CFM on the supply side (positive pressure side). The combined return and supply air leakage in the air handler and adjacent connections represents 5.3% of the system air flow (4.6% on the return side and 0.7% on the supply side). This is a concern, when considering that a 4.6% return leak from a hot attic (peak conditions; 120^oF and 30% RH) can produce a 16% reduction in cooling output and 20% increase in cooling energy use (Cummings and Tooley, 1989), and this was only from the air handler and adjacent connections.   (Please see Figure 56.)

"Total" Duct Leakage:  Some important observations were made from the extended test data in 20 houses.  Total leakage on the return side of the system (including the air handler and return connection) was 53 Q_25r,total. With weighted operating pressure on the return side of about -100 Pa (including the air handler), operating return leakage was calculated to be 122 CFM, or 9.7% of the rated system air flow.

Total leakage on the supply side of the system (Q_25s,total) was very large, at 134.  The ASHRAE 152P method suggests using half of the supply plenum pressure as an estimate of the overall supply ductwork operating pressure, if the actual duct pressures are not known.  For the 20 systems with extended testing, supply plenum pressure was 73.3 Pa.  Based on a pressure of 37 Pa, actual leakage should be 167 CFM or about 13.3% of the rated air flow.  To test the ASHRAE divide by two method, supply duct operating pressure measurements were taken from 14 representative systems.  These averaged 35.9 Pa, compared to 65.7 Pa for the supply plenums for those same 14 systems.  For these systems, the duct pressure was 55% of the supply plenum pressure - making the ASHRAE method a reasonable method for estimating central Florida home's supply ductwork operating pressures. 

However, the ASHRAE method wasn't reasonable for estimating central Florida home's return ductwork operating pressures.  For these 20 systems, 38% of the Q_25r,total was in the air handler and 62% of the Q_25r,total was in the return ductwork.  Given an air handler pressure of -133 Pa, a return plenum pressure of -81.5 Pa, and return duct pressure of approximately -70 Pa, the weighted return side pressure was approximately -95 Pa.  By contrast, the ASHRAE method predicted  -41 Pa.  Clearly, in systems with a single, short return duct plenum like those commonly found in Florida, the actual operating pressure should be greater than the return plenum, maybe by as much as 1.2 times the plenum pressure.

Return side leakage is available on 58 of the 69 systems.  Return leak air flow (Q_r,total) combined for the air handler, return connection, and the return ductwork was found to be 152.4 CFM, or 11.8% of total rated system air flow for this group.  For this larger sample, Q_r,total is considerably greater than for the 20 houses with extended testing. These alarming results show that even in these newly constructed homes about 12% of return air and 13% of supply air duct systems are leaking.

Duct Leakage to "Out":  In 20 homes, duct leakage to "out" was measured.  (Please see Table 15, below.)  On average, 56% of the leakage of the return ductwork and supply ductwork was to "out."   "Out" is defined as outside the conditioned space, including buffer spaces like an attic or garage.  The fraction of leakage that was to "out" varied by air handler location.  For return ductwork, the proportion of total leakage to "out" is 81.4% for attic systems, 67.6% for garage, and 28.0% for indoors.  For supply ductwork, the proportion of total leakage to "out" was in the range of 52% to 56% for all three locations.

The attic return ductwork was the most predictive variable to "out" leakage findings.   All of the return ductwork for attic units was located in the attic.  Much of the return ductwork for other units was located in the house.  As a consequence, the energy penalty associated with locating the air handler in the attic was greater than indicated in the computer modeling results in Table 16, since the modeling only considered the leakage of the air handler cabinet and the adjacent connections, and not the return ductwork leakage.

Table 16 (above) shows that the operating supply leakage to "out" was large for all three air handler locations, averaging 89 CFM.  The average operating return leakage to "out" was slightly larger, at 90 CFM.  However, there was a large variation between air handler locations; 96 CFM for attic systems, 131 CFM for garage systems, but only 38 CFM for indoor systems.  From an energy perspective, the attic systems experienced the greatest "real" energy penalties, because all of the return ductwork and air handlers were located in the attic.  (Please see Table 15, p. 80.)  By contrast, a majority of the return leakage for the garage systems likely came from the garage (which is considerably cooler than the attic).  For indoor systems, the return leakage to "out" most likely originated from the attic.  However, since the return leakage was so much smaller, the energy impact was likely considerably less than both the attic and the garage systems.

Correlation of Supply Duct Leaks with Number of Registers:  When analyzing the supply leakage in the extended test data, a surprising correlation was observed. This correlation indicated a systematic and consistent duct fabrication problem across a wide range of air conditioning contractors. Figure 57 illustrates this correlation, showing that each supply duct has a remarkably predictable total duct leakage.  The coefficient of determination is 0.86, indicating that 86% of the variability in total supply duct leakage was explainable by the number of supply registers.  Figure 58 shows a similar relationship between supply leakage to "out" and the number of supply registers.  In this case the coefficient of determination was 0.69, indicating that 69% of the variability in total supply duct leakage was explainable by the number of supply registers.  Note that one of the two houses with 13 registers showed considerably less leakage than expected.  In this case, supply ducts were located in the interstitial space between floors. When the house was taken to -25 Pa, it is probable (though not measured) that the interstitial spaces were substantially depressurized as well, so leaks in those supply ducts would show less air flow (i.e., less pressure differential = less leakage air flow) and therefore be under-represented.

The following two figures (Figures 57 and 58) suggest that a duct leakage problem occurs in nearly all new homes. Researchers identified three issues that create most of the leakage: (1) the connection of the supply register or return grill, (2) the boot (supply box) to sheet rock connection, and (3) the flex duct to collar connection.  The supply register or return grill leakage typically shows as supply leakage in the "total" test.  It usually occurs when the register or grill does not fit snugly to the ceiling or wallboard.  Issues two and three show up as leakage to both "out" and "total." (Please see Figures 59 and 60.)

Figure 57.  Supply CFM25 "total" leakage versus the number of supply registers.	Figure 58.  Supply CFM25 "out" leakage versus the number of supply registers.

Figure 59.  Flexible duct to metal collar connection.	Figure 60.  Register to ceiling leakage.

Computer Modeling for Florida Energy Code Air Handler Multipliers:  FSEC researchers performed simulations and developed air handler multipliers for the Florida Energy Code using this study's simulation results.  Researcher used the FSEC 3.0 model, a general building simulation program developed in 1992.  This program provided simultaneous detailed simulations of a whole building system, including energy, moisture, multi-zone air flows, and air distribution systems.

In 2001, modeling had been performed to develop initial air handler multipliers. These multipliers were based on estimated Q₂₅ and duct operating pressures.  At the time of the 2001 modeling, there was essentially no data on air handler and connection leakage.  Modeling for this project was performed again, but this time using the results of the 69 field tested homes.

The modeling inputs used in 2001 and those from the current study are shown below.  (Please see Table 17.)   Note that the same Q₂₅ and operating depressurization (dP) values was used for all air handler locations, since there was essentially no difference between the Q₂₅ values for attic, garage, and indoor air handler locations when gas furnace units were removed from the analysis.

While the Q₂₅ leakage for the air handler and connections was about 65% less than earlier estimates, operating pressures were much higher. The air handler multipliers based on the current computer modeling results are presented in Tables 18, 19, and 20.  Modeling of air handler energy use also was performed for the air handlers located outdoors, despite the fact that no field data was collected for outdoor units.  The modeling input parameters were the same as the other air handler locations as shown in Table 17.  Note also that the air handler multipliers for the attic, indoors, and outdoors are normalized to the garage, since this location was considered the baseline.

AH Location	Winter			Summer
	old	2001	new	old	2001	new
attic	1.04	1.11	1.08	1.04	1.10	1.08
garage	1.00	1.00	1.00	1.00	1.00	1.00
indoors	0.93	0.92	0.94	0.93	0.90	0.92
outdoors	1.03	1.09	1.05	1.03	1.02	1.01

The final report for this study can be viewed online at http://www.fsec.ucf.edu/bldg/pubs/cr1357/index.htm.