Ten years ago, residential duct leakage testing was nonexistent. Today, it is routine for many energy conservation programs. Everyone has a favorite duct test, and many managers have adapted tests to meet the needs of their particular program. Each test gives slightly different information, though every bit of information is a piece of the same puzzle: What is going on in this house?

Home Energy Rating System (HERS) programs and regulators are struggling with the question of whether to require duct testing and, if so, what testing method to use. Since the primary purpose of HERS is to consistently compare one house to another, it is important to have standard test procedures within each state, if not across the whole country.

How important is duct testing? Collected field studies have shown that, on average, duct leakage accounts for 15%-30% of heating and cooling costs, depending on the location of the duct system. But houses vary considerably, and there is no way to judge whether a house will have leaky or tight ducts just by looking at it. To encourage testing, most HERS rating programs assume that ducts are very leaky when testing is not actually required.

Many duct diagnostic and repair programs only need to focus on how big the holes are, but HERS programs need to find duct system efficiency to determine the energy rating. ASHRAE is developing a standard method for determining duct system efficiency, which is already being considered by California for inclusion in the state's HERS regulations. In addition to more established tests with a duct blower and blower door, ASHRAE is considering a new test as a tool for rating existing homes in the standard.

Figure 1. Leakage into and out of the duct system while the air handler is operating changes the pressure in the house compared to outside. The house-pressure test uses this change in pressure to mathematically calculate the leakage that caused it. Note that the house pressure is taken with respect to the attic to minimize the effects of wind. (Source: Energy Performance of Buildings Group, LBNL)

To come up with an energy use estimate, one needs to know:

• The supply leakage flow to unconditioned spaces while the air handler is in operation.
• The return leakage flow from unconditioned spaces while the air handler is in operation (and, especially for cooling, whether the leakage is from the attic or crawlspace).
• The effect of holes in the ducts on overall house air exchange when the system is off.

The procedure assumes the use of a digital manometer with two channels, so that the user can easily switch between different measurements. Some measurements are repeated many times and averaged to reduce the uncertainty of measuring very small pressures.

Note that

• All pressure measurements should be five-second averages, measured to the nearest tenth of a Pascal.
• The attic should be well vented, since it is being used as a surrogate for outside pressure.
• The pressure differences across the ceiling (ÆPon, ÆPoff, and ÆPRB) are measured using the attic as the reference pressure.
• The pressure differences for the supply and return ducts (ÆPret, ÆPsupon, ÆPretRB, and ÆPsupRB) are measured using the house as the reference pressure.

Procedure

1. Close all the exterior windows and doors in the building and open all interior doors. If the basement is conditioned space, open the doors between the basement and other conditioned spaces and close the basement windows. If the basement is unconditioned, then close the doors between the conditioned space and the basement.
2. Install one plastic tube between the building and the attic by passing one end of the tube through the attic access hatch and connecting the other end to the reference port of a manometer. Close the hatch and tape the gap around the tube.
3. Install a second plastic tube between the return duct and the building by passing one end through the return grille and connecting the other end to the input port of the manometer. The end of the tube within the return duct should be midway (within 3 ft of the midpoint) between the grille and the air handler return plenum. The return pressure must be measured on the furnace side of the filter. If the filter is not located at the grille, the filter should be removed and any leaks created by removing the filter from its access should be sealed.
4. Connect a third plastic tube to a pressure pan. (A pressure pan is a device used to temporarily seal individual registers and grilles.)
5. Turn on the air handler fan and wait at least 30 seconds for the fan to reach steady operation. Record the pressure difference between the house and the attic ten times (ÆPon). Record the pressure differential between the return duct and the house once (ÆPret). If the system does not have a fan switch, turn on the fan by raising or lowering the thermostat setting.
6. Turn off the air handler fan and wait for it to come to a complete stop. (Some air handler fans continue to operate for some time after they have been switched off.) Record the pressure difference between the house and the attic ten times (ÆPoff).
7. Turn on the air handler fan and wait at least 30 seconds for the fan to reach steady operation. Record the pressure difference between the house and the attic ten times (ÆPon). Record the pressure difference between the return duct and house once (ÆPret). This is a repeat of step 5. Cover one supply register with the pressure pan and wait ten seconds for the system to return to equilibrium. Record the pressure difference between the pressure pan and the house once (ÆPsupon). Remove the pressure pan after this measurement is complete.
8. Turn off the air handler fan and wait for it to come to a complete stop. Record the pressure difference between the building and the attic ten times (ÆPoff). This is a repeat of step 6.
9. Turn on the air handler fan and wait at least 30 seconds for the fan to reach steady operation. Slowly block the return grille with newspaper until the pressure differential between the return duct and the house is approximately -100 Pascals (Pa) (-0.4 inches H₂O), or until the register is fully blocked, whichever comes first. For systems with more than one return grille, block each grille by an approximately equal amount in order to keep the pressures in each branch of the return as uniform as possible.
10. Record the pressure difference between the house and the attic ten times (ÆPRB). Record the pressure difference between the return duct and the house once (ÆPretRB). Cover one supply register temporarily with the pressure pan and wait ten seconds for the system to reach equilibrium. Record the pressure difference between the pressure pan duct and the building once (ÆPsupRB).
11. Turn off the air handler fan and wait for it to come to a complete stop. Record the pressure difference between the building and the attic ten times (ÆPoff).
12. Average all measurements of ÆPoff from steps 6, 8, and 11. Average all measurements of ÆPon from steps 5 and 7.
13. Use the equations in "House-Pressure Test Equations" to determine supply and return leakage flows.

The idea is that the air handler is supposed to be a recirculation system, taking the same amount of air from the house through the returns as it puts back into the house through the supplies. So if interior doors are open, the operation of the air handler should not affect the pressure in the house.

However, leaks in the duct system will cause a change in the pressure in the house when the fan is turned on. Return leakage increases house pressure by introducing outside air into the house that must escape through the building shell. Supply leakage decreases house pressure, since the air lost from the supply ducts must be replaced by air sucked in through the building shell (see Figure 1.)

If supply duct leaks happen to balance the return duct leaks exactly, the house pressure won't change when the air handler is turned on. This is why the procedure includes a second step that involves blocking the return grille with a newspaper. Blocking that grille increases the pressure differential across the return leaks and decreases the pressure differential across the supply leaks. Thus, even if the leakage flows were balanced during normal operation, they won't be balanced when the return is blocked.

The supply and return duct leakage flows to or from outside can be translated into duct efficiency and energy costs using equations in ASHRAE Draft Standard 152P. The energy impacts of duct leakage while the fan is off are included in a standard blower door test.

This test does not distinguish between return leakage from an attic and that from a crawlspace, so the rater must determine the source of the return leakage from the location of the return ductwork. In the draft ASHRAE standard, the rater simply inputs the return duct location, which is then used to calculate the temperature and humidity conditions of the air being sucked into the return leaks.

The only equipment needed for this test is a digital pressure manometer with a resolution of 0.1 Pascals (0.0004 inches H₂O), a pressure pan, and a blower door to test house leakage.

The house-pressure test is not as precise as duct blower tests. The results are less reproducible, so an auditor, inspector, and contractor may all get different answers. It is also hard to compare this test directly to a duct blower test, since the two are measuring different things. The house-pressure test is a "quick and dirty test" for those occasions when time is limited, and the information it measures (leakage flow) is more useful for calculating the energy lost through duct leaks than are measurements of leakage area.

John Andrews at Brookhaven National Laboratory (BNL) also tested the house-pressure method against the duct blower plus blower door method in eight houses on Long Island. For supply and return duct leakage, Andrews found that the average discrepancy between the two methods was 7% of system fan flow (see Figure 3). While in some houses the results were fairly close, in other houses the tests gave very different results. When he compared the seasonal distribution efficiencies obtained from the tests, there was an average discrepancy of 5%.

Andrews noted that the house-pressure method was quick and easy to do. However, once again, because neither test represents a primary standard, he concluded that these results cannot be used to determine the accuracy of the ASHRAE standard. In the more complex return systems found in the Long Island houses, Andrews discovered that the correct measurement of pressure in the return duct is particularly important. Measuring improperly caused three of his results to show air leaking into the supply ducts.

A similar problem with incorrect flow signs (that is, directions of leakage) was discovered in approximately 20% of the California houses. In almost all cases these were houses that had very small leakage on either the supply or return side. The problem occurs because the house-pressure test makes direct measurements of the net leakage flows under normal operation and blocked-return operation. The calculations required to split these direct measurements into the separate contributions from supply and return leakage flows can be thrown off either by poor measurements of duct pressures or by limitations on the precision of the technique at low leakage flows.

Sensitivity analyses suggest that the uncertainty in the calculated supply and return leakage flows is on the order of 30-40 cubic feet per minute (CFM). The proposed ASHRAE standard treats negative supply-leakage results as return leakage (setting the supply leakage to zero), and positive return leaks as supply leaks (setting the return leakage to zero). A more rigorous uncertainty analysis of this technique is currently underway at LBNL and BNL.

Mark Modera is a staff scientist at Lawrence Berkeley National Laboratory in Berkeley, CA, and the chair of ASHRAE Standards Project Committee 152P on duct leakage. Jeanne Byrne is managing editor of Home Energy.

	the pressure in the house with respect to (WRT) the attic, while the air handler fan is on.
	the pressure in the return duct WRT the house, while the air handler fan is on.
	the pressure in the house WRT the attic, while the air handler fan is off.
	the pressure in the supply duct WRT the house, while the air handler fan is on.
	the pressure in the house WRT the attic, while the air handler fan is on, and the return grille is blocked.
	the pressure in the return duct WRT the house, while the air handler fan is on, and the return grille is blocked.
	the pressure in the supply duct WRT the house, while the air handler fan is on, and the return is blocked.

Equations

Calculate the net flow leaving the building due to supply and return duct leakage:

The unbalanced duct leakage flow under blocked-return conditions is:

The supply and return duct leakage flows under normal operating conditions are calculated from:

Supply pressure ratio:	Return leakage flow:
Return pressure ratio:	Supply leakage flow:

Then correct the results for the shift in neutral level associated with turning on the distribution system fan. High-level leaks are those in attics; low-level leaks are those in crawlspaces and basements (when the basements are outside the conditioned space). Multiply the supply and return leakage flows by one of the following correction factors:

1. Ducts in both high and low locations: no correction (Fnl=1.0)

2. Duct at high level only:

3. Ducts at low level only:

Math Notes:

sign (x) is simply the sign of whatever is in the parentheses. That is, when x is negative, sign (x)=-1, and when x is positive, sign (x) =1.

|x| is the absolute value of x. That is, if x is negative, make it positive. If x is positive, it remains positive.
 
 

Home Energy can be reached at: contact@homeenergy.org
Home Energy magazine -- Please read our Copyright Notice