skip to Main Content

Can a Single CO Reading Really Confirm Combustion Safety?

Understanding Combustion Is The First Step In Combustion Safety.

Guest post by David Richardson of Thermal Performance Services in the Lexington, Kentucky area. David is an EVER rater, HVAC contractor, and combustion safety instructor for the National Comfort Institute. This is his second guest post in the EV Blog, the first one discussing his education in combustion safety. ~ab3

Guest post by David Richardson of Thermal Performance Services in the Lexington, Kentucky area. David is an EVER rater, HVAC contractor, and combustion safety instructor for the National Comfort Institute. This is his second guest post in the EV Blog, the first one discussing his education in combustion safety. ~ab3


In the BPI Standards, measuring carbon monoxide (CO) in undiluted flue gas of combustion equipment is a critical part of combustion safety. This is one of several measurements that allow us to gauge the safety of an appliance and ensure that it operates properly.

According to BPI Standards, if a natural draft appliance passes worst case spillage, the building analyst is to take a CO reading after 10 minutes of operation or after the equipment reaches steady state operation. But, does a single CO reading really give us enough information to verify that a piece of equipment is really operating safely?

The information available in watching carbon monoxide readings can be eye-opening. One thing that always holds true about proper combustion is that the longer a piece of equipment burns, the cleaner the burn should get. This is critical to remember when using CO readings as a diagnostic tool; the burn should always get better, not worse. If the burn gets worse as the combustion process continues, there are underlying problems that need to be uncovered.

CO testing is similar to watching your favorite sporting event – If you’re able to watch only a couple of minutes in the middle of the game, you’re likely to miss not only the most exciting parts of the game, but also the game-changing moments as well. CO testing should be no different. If a building analyst relies on a single-point CO test, they’re likely to miss some of the most important parts of the combustion process.

If you’re wondering where I’m going with this, let me remind you that the burn should get cleaner as the combustion process continues. When a piece of equipment is operating safely and properly, the CO readings should be stable and stay in a range of acceptable levels. This tells us that the flue gases are leaving the equipment properly and that combustion air is making its way into the burner compartment. Smoke testing the draft hood can’t tell us if this is occurring as it only shows if room air is moving up the flue.

The CO numbers should never go up or continue to climb once the initial components have warmed up. If the CO numbers continue to climb, the equipment is operating in a deteriorating condition. This pattern of numbers tells us that the combustion process is breaking down inside the equipment itself. When rising CO numbers are present, combustion air is not making its way into the burner compartment and flue gases are not leaving the heat exchanger.

Rising CO numbers are one of the most dangerous and overlooked conditions in CO testing and cannot be discovered with a single-point CO test. The reason this condition is so dangerous is that you have no idea of where the CO numbers will stop. There have been many instances where rising CO numbers have gone from numbers that would look acceptable in a single-point CO test to very dangerous in mere seconds due to the deteriorating condition inside the equipment.

It doesn’t take as long as you would think to find this problem if it exists. Watching the pattern of the CO numbers will reveal if this problem is present. The next time you perform a combustion safety test, watch the whole combustion process instead of just a portion of it. You might be surprised at what you find.

This Post Has 4 Comments

  1. Important information, David.
    Important information, David. Thanks for posting. Do you happen to have any guidelines for multiple tests. E.g., number of tests, time between samplings, etc.,? Was just wondering. Thanks again!

  2. You have obviously been
    You have obviously been influenced by Jim Davis and you are absolutely correct. Constant monitoring of CO and O2 can tell an analyst a great deal about the operation of the gas appliance. However, BPI is not listening. They used to when Jim was their first instructor.

  3. There is ongoing confusion
    There is ongoing confusion about undiluted and CO Air Free readings using combustion analyzers. Since analyzers provide both readings from a undiluted sample, which reading is the one that BPI references? John Jones of BPI says, “Ovens and all combustion appliances with exhaust vents (pipes) should be tested using the air-free method (setting).” If the auditor is using a basic CO measuring device like a Monoxor that doesn’t have an COAF setting, this would mean that they would be getting the wrong information. Is that correct?

  4. My apologies in taking so
    My apologies in taking so long to respond to you all, thank you for all the comments. 
    John there is a set of guidelines for how the testing is to be performed on multiple test locations such as would be the case with a natural draft furnace. 
    The easiest way to do this is just as you would read a book going from left to right. Starting in the first heat exchanger cell outlet you would take a reading during the run cycle for close to 30 to 40 seconds or once stabilized and then move to the second cell outlet obtaining a sample there for 30 to 40 seconds again. 
    The process repeats again till you have one set of numbers for each heat exchanger cell outlet then the process is repeated again till there are three readings taken per cell. This will help in determining the stability of combustion. 
    Jim you are absolutely correct about Jim Davis being an impact in my life on this. I am fortunate enough to be the first person that Jim has passed the torch to for teaching his life’s work. It’s good to know you know the history between Jim and BPI, their standards have changed much since he first gave input on them many years ago. My hope is that they will move back towards so many of the important concepts that seem to have been forgotten. 
    Paul you are right about the air-free CO readings needed for ovens in regards to ANSI standard Z-21. The BPI standards from what I gather are designed around 100 PPM of CO as read on the analyzer which in my humble opinion is too high. 50 PPM as read should be the most any gas oven produces. 
    This rule helps give auditors a number to shoot for if they are using a CO only test instrument or if the Oxygen sensor in their combustion analyzer goes bad. 
    One thing about CO air free is that it is a calculated number being figured on the analyzer and it has a chance of being off depending on certain variables. 
    In my opinion sticking with 50 PPM as read gives all auditors a useful reference when making these checks in the field.  
    In every class I teach the numbers to be used for diagnostics are as read CO numbers. 
    Air free numbers are important for code compliance and have their place in that regards. 
    It is much easier from what I have found to use as read numbers when performing diagnostics in the field. 
    Thanks for all the comments and great discussion!

Comments are closed.

Back To Top