Different people do it different ways depending on how fancy a meter they have and how much training they have absorbed.
The most accurate way is to measure the oxygen content in the flue pipe before and after the blower comes on. If the oxygen increases after the blower comes on then the exchanger must be leaking. The requires drilling a hole in the flue which I don't agree with......
I like to disable the blower and insert a CO detector probe in the air stream (close to the heat exchanger) until the high limit trips. This will ensure a hot heat exchanger (opens cracks) and also test the limit at the same time. Those limits should be tripped every so often anyway to make sure they don't stick. If there is a leak the CO detector's ppm level will show it (must have 0-2000 ppm digital read out detector). Be careful not to get false reading from normal rust & nicotine burning.
A step down from that is closing off all the supply registers and turn the blower on high speed and observe the flame behavior. This is not fool proof because the air pressure may not necessarily be coming back in the flame chamber. But most of the time it will show a leak by the flames rolling out or acting like a tornado is trying to blow them out.
I wouldn't recommend old fashioned remedies like spraying salt in the blower compartment......adding a corrosive substance to metal is not good IMO.
I've never had any luck with smoke bombs either but some people swear by them.
Yea we used to use the old smoke bombs, but they were used for units that are oil fired and have a huge chamber. Aside form what Steve said you could also remove the chamber and due a hands on check for cracks.
Contractor Talk - Professional Construction and Remodeling Forum
A forum community dedicated to professional construction and remodeling contractors. Come join the discussion about the industry, trades, safety, projects, finishing, tools, machinery, styles, scales, reviews, accessories, classifieds, and more!