GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Green Communities

Combustion Safety and the Worst-Case Depressurization Test

Make sure that if all the exhaust fans in a home are running, you don't end up with dangerous exhaust inside the living space

Image 1 of 2
This manometer, or pressure gauge, is reading - 6.1 Pascals between the outside reference and the combustion zone with an oil-fired natural draft boiler during a worst-case CAZ test.
Image Credit: Peter Yost
This manometer, or pressure gauge, is reading - 6.1 Pascals between the outside reference and the combustion zone with an oil-fired natural draft boiler during a worst-case CAZ test.
Image Credit: Peter Yost
This Wizard "smoke" stick is an easy and harmless way to reveal air flow. In this case it is showing air being drawn into the back of the clothes dryer as the clothes dryer exhausts about 150 cubic feet per minute (cfm) during operation.
Image Credit: Peter Yost

The problem

Vented combustion appliances—gas and fuel oil water heaters, furnaces, boilers, even fireplaces—need to exhaust all of their combustion by-products, all the time, outside the home. An especially dangerous combustion exhaust component is carbon monoxide because it is odorless, colorless, tasteless and quite toxic. If you have no combustion appliances or they are sealed-combustion, no worries. But if any of these appliances is atmospherically-vented or has induced draft or even if they are power-vented, we need to test to see if conditions in the home can create backdrafting (backdrafting is combustion exhaust rolling out of the intended exhaust pathway and into the home).

The test

There is actually a series of combustion safety tests but we are going to focus on the simplest and most often the first test, the Worst Case Depressurization Test, also known as the worst-case CAZ—combustion appliance zone—test. So what is the worst case? Essentially, you turn on every device in the home that can create negative pressure in the room or space in which the combustion appliance is located and compare that pressure to the outside. Here is a list of these devices:

• Bath exhaust fans – typically pull somewhere between 25 and 100 cubic feet per minute (cfm) out of the house)

• Kitchen hood (if it really vents to the outside) – usually around 100 – 150 cfm but downdrafting or commercial types can exhaust more than 1500 cfm!

• Clothes dryer – usually around 150 cfm

• Laundry room exhaust fan – about the same as a bath fan

• Attic exhaust fan – see below

• HVAC air handler – see below

There are three things to note about this list:

1. Attic exhaust fans: Aren’t these just to cool off the attic and the attic is outside of the conditioned space? They are, but if the ceiling plane of the top floor is not airtight, attic exhaust fans can end up depressurizing a home’s interior.

2. Air handlers: They just move air around the INSIDE of the home, right? Well, that is what they are SUPPOSED to do, but if there are ducts that run through unconditioned space that leak—or if there is unbalanced delivery/leakage between the supply and return side of the system that expresses itself in the space with a combustion appliance—the air handler can de-pressurize the CAZ. So, we need to test worst case with and without the air handler on, if the home has a forced-air HVAC system.

3. Whole-house attic ventilation fans: These are mighty large exhaust fans; shouldn’t they be on the list? While these fans can certainly depressurize a home big time, they are not included in the CAZ test. It is assumed that since these fans are for cooling by air movement, enough windows are open during operation of the fan that depressurization and subsequent backdrafting are not an issue.

The equipment

The test requires just two pieces of equipment—a manometer (pressure gauge, pictured above) and a smoke source (several different types, but the “toy” smoke generator pictured is the most fun).

The procedure

After surveying the home for combustion appliances and exhaust fans, the procedure goes like this:

1. Seal the house: close all exterior windows and doors.

2. Turn off all combustion appliances.

3. Close all interior doors.

4. Measure the baseline pressure. This is the the pressure relationship between the combustion space and the outside with no exhaust fans on and should be no more than a pascal or two. NOTE: Wind can really make it difficult to get good readings. Sometimes turning fans off and on is necessary to get decent readings.

5. Turn on all the fans and measure the worst-case depressurization.

The results

Here is a representative chart (from the CMHC Chimney Safety User’s Manual Reference #4) for safe limits for the CAZ test.

Combustion Appliance Chimney or Flue Height (ft) Unlined Chimneys on Exterior Walls Metal Lined, Insulated or Interior Chimneys
Gas Furnace 13 or less – 5 Pa – 5 Pa
Gas Boiler 14 – 20 – 5 Pa – 6 Pa
Gas Water Heater More than 20 – 5 Pa – 7 Pa
Oil Furnace 13 or less – 4 Pa – 4 Pa
Oil Boiler 14 – 20 – 4 Pa – 5 Pa
Oil Water Heater More than 20 – 4 Pa – 6 Pa
Fireplace N/A – 3 Pa – 4 Pa
Wood Stove N/A -10 Pa -10 Pa
Induced Draft Appliance N/A -15 Pa -15 Pa

If you get test results that exceed these limits, there are only really four options:

1. Replace the appliance with a power-vented or sealed combustion unit.

2. Switch out to a non-combustion appliance.

3. Isolate the combustion space (build a room with dedicated outside air to the room with the combustion appliance)

4. Ensure operation of exhaust fans that avoids worst-case depressurization, backed up by a CO monitor in the combustion zone.

In the first picture above, the manometer is reading -6.1 Pa during the worst-case depressurization test on my own home. This is with the clothes dryer, two bath exhaust, and a kitchen exhaust hood running and an average base line pressure reading of – 2.9 Pa (bit of a windy day but not gusting too badly). That’s a CAZ test of -3.2 Pa, well within the safety limits for a natural draft oil boiler with a center chimney more than 20 feet tall. We do keep a CO monitor in the oil boiler combustion zone as a bit of a belt-and-suspenders approach.

For more information on all of the combustion safety tests, visit the Building Performance Institute website.


  1. Naomi Davies | | #1

    Worst Case Safety Limits
    A most helpful supplement to using the BPI Standards alone.
    Thank you

  2. Frank Richter | | #2

    Dealing with wind
    Wind can effect the baseline reading if the exterior hose is exposed to wind. Placing the end of the hose inside a large soda bottle can protect it from the wind while still yielding an accurate outside reading. If the reading fluctuates wildly, wind might be the culprit.

  3. Peter Yost | | #3

    Dealing with wind and pressure readings
    Hi Frank -

    You are right; wind is quite often the culprit when outside pressure readings are erratic. I have tried shielding the hose end with a cardboard box with mixed results, but the large soda bottle is a slick idea. Will have to try that--thanks.

  4. Ed Voytovich | | #4

    When one door closes, another door opens
    Simply closing all the interior doors does not necessarily create the worst case. One certainly shouldn't close the doors leading to rooms where there are exhaust fans, dryers, or what have you. There's really no way to predict the effect of the air handling system on building pressures without trying various combinations of door configuration to achieve the worst case. If there are no cold air returns in the bedroom, or if a particular supply duct is leaky to the outside, or if the dresser is in front of the register, then the pressures are affected.

    The large bottle to attempt to control the effects of wind won't work because the bottle is "connected" to the ambient pressure by the opening through which the manometer hose passes. We can't fool Mother Nature! I'd recommend the long-term averaging setting on the manometer.

    I once overheard a building performance contractor from a mountainous area in the Carolinas complain at the Energy Conservatory booth at a conference that their manometers were no good because they don't work when the wind blows. The contractor had it backwards: it's not that they work poorly, it is exactly because they work so well that digital manometers are all over the place in the wind.

    If somebody told you it was going to be easy, I hope you didn't believe them!

  5. Todd VanOsdol | | #5

    "Airtight Woodstove" Defined?
    There is a large difference between the depressurization limits for open fireplaces (-3pa) and "airtight" woodstoves (-10pa). However, there is also a large grey area between these two extremes of solid-fuel appliances that have varying levels of airtightness. We often see many woodstoves that are hardly airtight and also run into fireplaces that seal up fairly well. Is there a standard by which we determine whether to use -3pa or -10pa as an acceptable amount of depressurization for a particular solid-fuel appliance?

  6. Peter Yost | | #6

    Standard for depressurization limits for solid fuel appliance
    Good question Todd. But I am not aware of any standard and I think your field experience may be the best answer we have. If we can "slide" the depressurization from around -3 to -10 Pa by adding exhaust loads one at a time, checking for backdrafting as we go, then that might be the best we can do.

    But maybe there is a standard and others will know of it. Any takers?

  7. David Whte | | #7

    re: "Airtight Woodstove" Defined?
    to make things more complex, the report does introduce the term "open wood stove" in a previous table, but does not mention open wood stoves in the table of depressurization limits.

    i called CMHC, who do not do technical Q&A on the report. i tried looking up the report authors (Scanada-Sheltair Consortium), and they appear to still exist in some form 22 years after the report was written, but their website ( did not load.

    anyone have insight/info on this? it's a big jump between 3 Pa and 10 Pa.

  8. GBA Editor
    MIKE GUERTIN | | #8

    Don't Overlook Central Vacuum - Worst Case
    Add Central Vac systems to Peter's list of exhaust equipment to turn on. Some systems pull 200+ CFM and can depressurize the CAZ above the limits. Some systems are exhausted internally and some externally so you may have to fiddle with door positions to reach the greatest depressurization.

  9. Peter Yost | | #9

    Central Vac and depressurization
    Wow - good one Mike, and I can't believe I missed this one. Good thing your work is "exhaustive." :-)

  10. grant_v | | #10

    If the CAZ passes a worst-case depressurization test on its own without makeup air (no spillage, good draft, and low enough CAZ pressure), does the CAZ still need to have a volume meeting 50 cubic feet per 1000 BTU's in the eyes of a mechanical inspector? It would seem pointless at that point if I verified the appliances will be safe through other methods.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |