Issue : October-December 2003

Duct Work Leakage & Leak Testing

“How much does ducting leak? Where does ducting leak? How much should it leak? What are the applicable standards? How to use them? How to set up a leakage target – and, stay within it? What is a Leak Testing Rig and how does it help test and measure the leaks? These are some of the questions addressed in this two-part article.”
Part 2 of 2

By R. V. Simha
Air Conditioning Consultant
Airtron, Bangalore

R. V. Simha is a graduate engineer in both mechanical and electrical engineering, with over 40 years of experience in HVAC. He has been a practising consultant for the last 26 years. He is an active member of ISHRAE and ASHRAE South India chapter.

The leakage budget

We now have in place the necessary tools to plan “leak testing”. The first step is to determine the permissible system leakage. It is the responsibility of the design engineer to fix this all-important parameter. The consideration is essentially the nature and requirement of the project. Once this parameter is fixed, it will be necessary to apportion it amongst the various elements of the air distribution system. As we have seen, the leakage depends on the static pressure in the duct and that is not the same for all sections of the ducting. Leakage also depends on the "leakage class", i.e., the class or quality of duct construction, which will not – and need not – necessarily be the same for all sections of the ducting. In addition, there are various elements of the air distribution system, which are also sources of leakage. It should be remembered in this context that the maximum system leakage referred to for ductwork is the system leakage, less leakage through other elements of the system – which can be called the duct leakage. The design engineer will accordingly find that he has to formu-late a “leakage budget”, which, will look somewhat like this:

  1. Duct work
  2. Air Handling Unit
  3. Damper d. Silencers
  4. Heaters
  5. Filters
  6. Plenums
  7. Cooling coils
  8. Equipment connections
  9. Collars-to-diffusers connections

The duct leakage will be the system leakage less the sum of leakages of items b to j.

AHU leakages

Photo 1

An important component of leakage is obviously the air-handling unit. It becomes a major player in the budget particularly where the system static pressure is high, say, 75 to 150 mm. It is therefore, necessary to specify and calculate the AHU leakage.

An applicable AHU leakage standard is the European Standard EN 1886. Extracts from this standard are furnished in Table 6 and 7 below:

Table 6 : Leakage limit for casing at 400 Pa negative pressure
Leakage class Max. leakage
rate L/s per m2
efficiency Em %
Em < 40
1.32 F5-7 40 ≤ Em < 90
B 0.44 F8-9 90 ≤ Em

The appearance of "filter class" and "filter efficiency" may be explained in terms of the need to relate leakage rates to quality of air i.e., the higher the air quality, the tighter should be the ducts.

Table 7 : Leakage limit for casing at 700 Pa positive pressure
Leakage class Max. leakage
rate L/s per m2
B 0.63

The following equation can be used to determine maximum allowable leakage in case the units are tested at pressures different from 700 Pa.

fm = maximum allowable leakage rate at the actual test pressure (L/s per m2)
f700 = maximum allowable leakage rate at 700 Pa (L/s per m2)

It is recommended that the specified leakage class for AHUs be lower than that specified for the ductwork, because of the larger number / length of joints per m2 of casing area than in ductwork and the greater difficulties encountered in sealing air handling units, due to penetrations, access doors etc.

A standard AHU from a "high quality" manufacturer can be expected to be within the limits of DW 142 Class A and Class B. Class C may be targeted where project requirements so stipulate – but then, the specifier must accept that there is a price penalty to be paid for specialized unit construction and testing. Figure 3 shows AHU leakage calculations.

Fig. 03


Calculations for carrying out leak testing

Having seen how AHU leakages can be calculated, the way is now clear for arriving at the duct leakage. To be sure, there are other items of leakage too, but focus has been on the AHU, since it happens to account for a significant fraction of the system leakage.

The method of calculation of duct leakage is best understood by studying a worked example (see box on following page). The procedure described therein is also shown graphically in Figure 4.

Fig. 04

Sizing the leak testing rig

Let us suppose that it is required to test the 5 nos. branch ducts of the worked example. The approach is shown below:

Units IP SI
No. of branch duct
Mean perimeter (70% of max.)
Duct length
Surface area
Leakage cfm / branch duct
5 Nos.
5 ft x 2 ft.
9.8 ft.
50 ft.
490 sft
490 x 0.044 =
21 cfm
5 Nos.
1.5 m x 0.6 m
3 m
15.25 m
46 sqm
46 x 0.214 =
9.8 L/s

It is seen that with a test rig capable of handling air flow rates down to 21 cfm, an entire branch can be tested in a single run (in practice, it would be designed to handle a some what lower flow rate, say 15 cfm – to keep a safety margin). If however, it is not possible to put together all the pieces of an entire branch in one go, the flow rate will be less than 21 cfm. In that case, the rig must be able to measure the smaller flow rate called for – perhaps, around 10 cfm. This fixes the lowest flow rate that the rig must be able to handle (the ducting being tight – as it will be, if it is constructed to higher pressure classes, can also call for low flow rates). The difference between the maximum and minimum flow rates, at the lower end of the range of the rig, might some times warrant the use of two different orifice plate assemblies – one for low flow rates and another for larger flow rates – like say 5 cfm and 20 cfm. If on the other hand, all the 5 branches are to be tested at the same time, the rig should be able to handle 21x5=105 cfm; the rig would then have a rating of 130 cfm approx. Thus the maximum and minimum flow rates for the rig would be 130 cfm and 5-10 cfm respectively.


Worked Examples
The design procedure is best understood by a worked example
(See Figures 3, 4 & 6 for clarifications)
Conditioned area (say 160 x 50ft.)
1 air change per hour
Cleanliness level
Total air
8000 sft
10 ft
80000 cu.ft.
1330 cfm
Class 10,000
40 ach
53330 cfm
743 sqm
3 m
2266 m3
604 L/s
Class 10,000
40 ach
24160 L/s
No. of Air Handling Units (AHU) 2 2
Surface Area of each AHU :
a. Positive Pressure side
b. Negative Pressure side

237 sft
463 sft

23 sqm
43 sqm
Leakage per AHU : See Figure
a. Positive Pressure side
b. Negative Pressure side
Total per AHU
Total for 2 nos. AHUs

51 cfm
41 cfm
92 cfm
183 cfm
185 cfm

23 L/s
18.6 L/s
41.6 L/s
83.2 L/s
85 L/s
Summary :
a. Total permissible leakage

1 ach
1330 cfm

1 ach
604 L/s
b. AHU Leakages 187 cfm 85 L/s
c. Others :
HEPA Filter, damper,
duct-to apparatus
connections, apparatus
-to-duct connections,
apparatus casing … etc

90 cfm

41 L/s
d. Sub Total (item b + item c) 277 cfm 126 L/s
Net permissible duct leakage
(item a – item d)
Duct surface area -
SA duct @ 1250 fpm
RA duct @ 1000 fpm

1052 cfm

10600 sft
13250 sft
23850 sft.

478 L/s

985 sqm
1231 sqm
2216 sqm

Permissible leakage rate
Supply Air
Duct No. / size
Duct test pressure
Leakage Rate
converted to (at 250 Pa)

0.044 cfm\ft2
53330 cfm
5 Nos./5ft x 2ft.
2.36 in. wg

0.024 cfm/sft.

0.214 L/s.m2
24240 L/s
5 Nos./1.5m x 0.6m
600 Pa
= 0.121 L/s m2
Duct pressure class selected Class C- DW
Class 3- SMACNA

This calculation serves to indicate the Duct Pressure class required. In this case, it will be seen that Class 3 SMACNA leakage rate is 0.15 L/s per m2 against 0.121 L/s per m2 required. This means that Class 3 just about fails to make the grade. Options available are:
a. Improve seal levels
b. Go for flat oval ducts (which are not yet made in India)

As far as DW is concerned, the leakage rate for Class C is only 0.109 L/s per m2; hence, it will suffice. In any case, DW has a higher duct pressure class (Class D) also.


The leak testing rig

The rig consists essentially of a fan (centrifugal), fan motor, flow measuring arrangement, flow adjusting arrangement, a manometer or similar pressure indicating device (for displaying the test pressure). See Figure 5.

Fig. 05

Typically, the fan duty required would be about 200- 300 cfm against a static pressure of about 6" to 8" wg. The flow rate thru the test duct work is varied by operating the bleed (control) valve.

An orifice plate assembly together with a manometer can be provided for flow measurements. Air flow results in a pressure drop across the orifice plate. This pressure drop is measured by a manometer.

The Orifice Plate has to be made to precise tolerances – to the applicable ASME Standard. The orifice is sized to yield a pressure drop that is large enough – even at the lowest stipulated flow rate – to be read easily under site conditions. (If the pressure drop is too large, the static pressure for which the fan has to be selected becomes correspondingly large – this is why the fan static pressure required turns out to be some what high – at 6” to 8” wg).

Fig. 06
Click to view the clear picture

The orifice plate needs to be calibrated in laboratories equipped with instruments of appropriate accuracy levels; also, their calibration should be traceable. It is good practice to calibrate (the orifice plate) with more than one instrument. The calibration results in a pressuredrop vs flow rate curve which can be used to provide a table from which flow rates can be read off pressure drop measurements.

When an orifice plate assembly is used, a pressure drop as high as about 10 mm can be obtained for flows as small as 10 cfm. Such pressure drops can be read easily on manometers in the field – sometimes even U-tube manometers will do – free from hassles of level adjustments. Sophisticated instruments, it may be noted, are required only for calibration, and instruments commonly used by contractors will suffice for field measurements with the rig. The U-tube manometers can, if required, be put together in a matter of few minutes or an hour or two, at the most, in the field.

Photo 02

The fan-motor unit, the bleed control valve and the manometer can be assembled together as a portable unit, while it is convenient to mount the Orifice Plate Assembly close to the ductwork under test.

Test rigs as described above have been fabricated in our country and used successfully during the past 5 years. They have been applied for duct work leakage in the field as well as for AHU leakage measurements. See Photo 1(in AHU manufacturing plant). They can be built to meet specific requirements. In Photo 2 a "Low Velocity Air Leakage Tester" as manufactured by Air Flow Developments Ltd., England is shown.

A typical hook-up of the rig at site can be seen in Figure 7, while Figure 8 furnishes information about orifice plate assembly, orifice sizes, and location of vena contracta taps.


The leak testing procedure

Once the leakage testing rig is in place, the following test procedure may be adopted:

  1. Complete Part 1 of the Test Sheet (Appendix - 1).
  2. Connect test apparatus to section of ductwork to be tested.
  3. Adjust test apparatus until the static pressure differential is obtained.
  4. Check that the measured leakage is within the permitted rate. (No addition shall be made to the permissible leakage rate for access panels or dampers where these are included in the ductwork).
  5. Maintain the test for fifteen minutes and check that the leakage rate has not increased.
  6. Reduce pressure in section to zero by switching off the fan; then immediately re-apply test pressure to establish that the air leakage rate is not greater than the previous reading.
  7. Record details on Part 2 of the Test Sheet and complete, including witnessing.

Fig. 07


Location and rectification of leaks

The likely leakage sites are shown in Figure 2 in Part 1. Leaks may be located:

  1. By listening for them
  2. By feeling for them especially with a wet hand
  3. By applying soapy water over the seams or joints
  4. By using a smoke pellet (with the agreement of the client)

For rectifying leaks:

  1. Check that all joints are fastened tightly, change gaskets and sealants, if indicated.
  2. Reseal wherever leak locations have been identified.

Fig. 08


This article attempts to heighten the awareness of ductwork air leakage; besides it covers duct construction standards, permissible leakages for various ductwork classes and interprets the two well-known standards – SMACNA and DW, which are widely in use all over the world.

A procedure for establishing targets for total system leaks and ductwork leaks (which form a part of system leaks) and a method for planning and staying within system leakage flow rates so established, has been indicated. A worked example illustrates the procedures.

An understanding of the subject of ductwork leakage goes a long way towards aiming and securing superior quality of ductwork and achieving better performance. Considering, the poor standards of workmanship and the generally lax approach to quality of workmanship in our country, it is advisable to over-emphasize the importance of leak testing ductwork rather than looking for opportunities to waive testing requirements on some ground or other.