This post is a revised version of the last post about Evaluating Combustion Air. The changes are a result of the many discussions I had with the experts in the previous post. You know who you are. Thank you. I continue to value your suggestions. Feel free to comment and thanks again.
Evaluating Combustion Air
This procedure helps evaluate the need for supplemental combustion air during worst-case testing in a building with a naturally drafting chimney. Don’t make openings in the building without evaluating the need for combustion air. These new openings can lead to unintended consequences like pressurizing or depressurizing the CAZ or admitting cold drafts.
Evaluating the CAZ Volume
In the average building with more than 0.40 natural air changes per hour, the combustion appliance zone or CAZ should contain more than 50 cubic feet of volume for each 1000 BTUH of combustion-appliance input. However, a smaller volume may provide adequate combustion air and a larger volume may not depending on the airtightness of the CAZ.
Evaluating O2 and CO Readings
During worst-case testing, use a combustion analyzer to measure both CO and oxygen (O2). The O2 is an indicator of excess combustion air, and CO may be an indicator of insufficient combustion air.
- Sample undiluted flue gases as they leave the appliance’s heat exchanger during worst-case conditions.
- If the reading from the combustion analyzer is more than 5% with an atmospheric burner or more than 3% with a power burner or well adjusted and maintained barometric draft control, combustion air is probably adequate assuming CO is minimal
- If the O2 reading from the combustion analyzer is less than the above O2 values, this indicates that combustion air is inadequate or that the appliance is over-fired. We would expect significant CO to accompany such low O2 readings.
- If O2 is too low, measure fuel input to verify that the firing rate is at or below the manufacturer’s BTUH specifications for input. An excessive firing rate could also cause low O2 and high CO.
- If O2 is too low at the correct firing rate, open a door or a window connected to the CAZ. If opening the CAZ door, a nearby window, an exterior door, or any combination of these increases the O2 reading and decreases CO, then install supplemental indoor or outdoor combustion air as specified below.
The best solution to a persistent combustion-air problem may be to isolate the CAZ from the building and provide outdoor combustion air.
Installing Supplemental Combustion Air
The table shown here gives sizing guidelines for combustion air openings. If testing indicates the need for supplemental combustion air, install openings to one of these spaces.
- Another indoor space
- A ventilated intermediate zone, such as a ventilated attic or ventilated crawl space.
- From outdoors into an isolated CAZ
- From outdoors to the appliance by replacing atmospheric combustion appliances with sealed-combustion (direct-vent) appliances.
This is a revised procedure based on comments to the last blog post at this link.
You might suggest doing a worst-case test after any supplemental combustion air openings are added to the CAZ. Better safe than sorry.
In the field guide, I added:
After installing supplemental combustion air, repeat the worst-case testing to verify that the combustion air problem is solved and that the combustion is safe.
Lloyd Hamilton wrote the following on LinkedIn:
I read your new post.
While I agree with you for the most part, I posit that it would be even better to measure draft over fire. When adjusting combustion, draft over fire is the first thing adjusted. The combustion is then adjusted for this specific draft. The barometric draft device should maintain this draft over fire under all conditions. When there is insufficient combustion air the draft over fire will go down and combustion will be negatively affected. It might lead to more smoke if oil and more NOX if gas without changing O2 or CO significantly, but if draft over fire changes combustion will always be negatively affected.
The thing that is still an issue for me in this process is that it does not address something we spend a fair amount of time trying to fix in our Wx program. We are not a fan of adding “combustion air” to fix a make-up air problem. Exhaust fans and depressurization are really the problem and I really don’t think putting a passive opening into the building is the way to address it. The process does not distinguish between a legitimate combustion air problem and a depressurization problem.
Also, the sizing criteria is insufficient to address make-up air deficiencies.
I think this is the best sentence in the process: “The best solution to a persistent combustion-air problem may be to isolate the CAZ from the building and provide outdoor combustion air.”
Only, I might write it as: “The best solution to a persistent combustion-air or make-up air problem may be to isolate the CAZ from the building and provide outdoor combustion air or to install direct vent appliances.”
Thanks for the suggestion on that sentence. I will make that change. I appreciate your willingness to think through this whole process with me and our colleagues.
I’ve spent a lot of time thinking about make-up air and ventilation air over the past few days. We’re primarily concerned about the functioning of the combustion appliance. If not for the atmospheric combustion appliance, we wouldn’t be as concerned about depressurization caused by the lack of make-up air for fans. The point is that worst-case conditions may deny the atmospheric appliance the air it needs for combustion and dilution air. If the fans and the combustion appliances are competing for the same air, there may not be enough air for them both. The ventilation air to dilute a possible spill of combustion gases: how do we evaluate that? At worst-case, when the appliance is connected to adjacent spaces and probably to the whole building, is there really a difference between combustion air and make-up air? How often does anyone make a hole in a single-family home just to provide make-up air?
Isolation is kind-of a band aid, but works well in cases where the CAZ may already be pretty separated from the home. I wish we could convince more folks to use sealed-combustion appliances, but that’s a hard sell, even for new homes.
In comments on the latest BPI 1200 draft I raised a point similar to Tom Andrew’s, (re: page 40, Annex D., table D1A if backdrafting occurs with the CAZ door closed, Action Required says, “Recommend measures to improve air transfer between the CAZ and the core of the house”).
I said that, in my opinion, opening the CAZ to the interior is seldom the best solution; better to simulate closed combustion if at all possible. I would re-prioritize (i.e., flip upside-down) the list of supplemental spaces in your procedure, so that direct vented is first, isolated CAZ second, etc., with “another indoor space” at the bottom.
Where I’m from (Michigan) 90% of CAZs are basements and >90% of those are unintentionally conditioned. Fuel code says you can open the basement door and get your 50 cu.ft./kBtu., but is that really a good idea?
To address the two questions in your response: Evaluating the lack of air for the appliances is accomplished in a number of ways and I don’t have an issue with adding combustion analysis to the mix. And, yes, when the appliances are connected to the whole house, the source of air is the same whether it is for combustion or make-up.
I do think that guiding someone to add outdoor combustion air as the only solution may not be the best thing to advise. The process states:
“5) If O2 is too low at the correct firing rate, open a door or a window connected to the CAZ. If opening the CAZ door, a nearby window, an exterior door, or any combination of these increases the O2 reading and decreases CO, then install supplemental indoor or outdoor combustion air as specified below.”
Pressure relieving the CAZ to the house (indoor combustion air)or adding a hole (outdoor combustion air), in my opinion are not typically appropriate responses to a make-up air problem. They may be absolutely appropriate if it is truly a combustion air issue but my experience dictates that adequate combustion air is usually available and it is make-up for fans that is an issue (worst case).
It has been more the solution that I disagree with than the evaluation.
OK, we need a solution to the make-up air problem, I agree. I’m searching for a procedure that makes sense to a reader of that procedure. That’s why I’ve suggested that we don’t talk about so many types of air. When I say combustion air, I mean a combination of combustion air, dilution air, and ventilation air. They’re all the same air aren’t they? If we separate make-up air from combustion air, do we need two separate evaluation procedures and two separate solutions?
I need to re-explain why I started this conversation. There is a parallel conversation between Paul Francisco and Thom Knoll in the review of BPI Standard 1200 that is relevant too and cited here.
Thom Knoll believes that combustion air from the heated space to the CAZ is undesirable because of that heated air going up the chimney. Thom says in an email to Paul, copied to me: “I think the list of steps in Krigger’s procedure is still upside down so, if the CAZ is too tight: first install direct vent; if not direct-vent, isolate the CAZ and add air; and, finally, if there is no other option, increase communication from CAZ to inside.”
I have to agree with Thom on this. I was coming from the usual situation of no budget for the sealed-combustion appliances, but that is an unnecessary assumption. Yes sealed-combustion (direct-vent) appliances are the best solution.
My main guidance-writing problem is the assumption that if the CAZ has less than 50 cubic feet of volume per 1000 BTUH, that the combustion appliances necessarily have a combustion-air problem. NFPA 54 and the IRC have the same convoluted procedure to tell you where to put the holes and how big to make them. This procedure doesn’t work because the combustion-air problem isn’t verified and probably doesn’t exist. However, folks are cutting holes to indoor and outdoor spaces based on this guidance daily.
Tom Andrews and I agree that 1) combustion air problems are rare and that 2) insufficient make-up air and depressurization are more common problems. The challenge between Tom Andrews and me is finding a procedure we can agree on.
If make-up air is the main problem, then what is the procedure for verifying that problem and solving it? We may have created the problem by air-sealing the building shell. Do we now cut an expensive hole to provide make-up air?
From Paul Francisco to Thom Knoll and the blog via email:
Here is my view. I am not sure if this is different from Tom Andrews’ view.
If there is a problem simply because of the level of exhaust (which is what I believe Tom is referring to), then increasing the connection of the CAZ to the rest of the house does not help unless the exhaust is in the CAZ. Also, simply adding a hole to outside is not a good idea. In that case, isolating the CAZ from the rest of the building and providing dedicated combustion air from outdoors does make sense, and is something that I have recommended in the past.
Now, let’s look at a different case. You have a water heater in a basement. The dryer is in the basement. With the basement (CAZ) door closed the dryer backdrafts the water heater. When the CAZ door is opened there is no problem. In this case I think it makes sense to add connection between the basement and the house. For one, it is a lot cheaper. Truly isolating the basement from the house can be extremely time-consuming and expensive and it will save virtually no energy. You will also need to insulate the basement ceiling. If you add a sufficient connection to the house you have not had to punch any leaks in the house. Basements are typically considered part of the conditioned space, even if they do not have registers supplying conditioned air directly to the space. If having two parts of the conditioned space communicating with each other avoids a spillage problem, I think that is better than trying to separate off one part of the conditioned space and punching a hole. Keep in mind that a hole from outside to a basement will be at the highest stack pressure point, so it will allow the most air in when the appliance is off.
So, in my view the answer depends on the situation. But let me frame it one more way, in the form of a question. If you had an appliance that was performing fine, without isolating it and adding a hole to outside, would you isolate it and add a hole to outside? If the answer is no, then why wouldn’t you create the same condition when there is a spillage issue, if it is reasonable and works?
From Tom Andrews to Thom Knoll, Paul Francisco, and myself.
Just for clarification, I don’t have a problem with attaching a CAZ to the house if it will fix a depressurization problem – like pressure relieving a utility room to the house if the water heater does not work when the dryer is on. If the problem is an issue with exhaust throughout the home, then, obviously attaching the CAZ to the house doesn’t fix anything. The other thing I have been speaking about is the addition of a hole to the outside to provide make-up air. That is not realistic. Often, the best options are to isolate the CAZ and provide combustion/ventilation to the space or replace the appliances.
I think we are similar in our thinking.