Pressure Testing (furnace installation leak)

[CosmicDriftwood]

Chasing a ghost perhaps but in the process of pressure testing a simple propane furnace installation I installed a water Column pressure gauge. Over a period of 11 hours the pressure gauge reading will go from 10" WC to 0" WC. In the course of trying to understand this I installed a manometer in conjunction with the pressure gauge. I should add that this furnace is new and has never be used beyond an initial checkout.

Extensive testing shows that with an initial test pressure of 11" WC the rate of pressure loss is approximate 0.5" WC/hour. Exhaustive soap bubble testing of the short run of pipe and six fittings between the manual shut off valve and the furnace electric gas valve shows no leaks. So that suggest the the gas valve is leaking either through and/or from the valve body. I will say that there has never been a hint of propane odor.

The phenomenon that I need help with understanding it that after 15 hours or so with the test in progress and having been pressured to 10" - 12" WC the manometer will show a negative pressure of -3" WC -. This is a U-Tube manometer and is open on one side to atmosphere. So If indeed there is a leak through the electric gas valve - since all piping connections are tight, then the -3" WC is the difference between atmospheric pressure and the pressure inside of the piping and electric gas valve. I should add that the outlet side of the electric gas valve is always at atmosphere. It really looks like that there is a 3" vacuum relative to atmosphere right? But how can this be since the system will go to zero pressure, because of this "ghost" leak and then start to show negative (vacuum)? Nature abhors a vacuum so if there is a leak then the manometer should go to zero and stay there right?

The ANSI standard for electric gas valves allows for 235 cc/hour leakage passing through the valve and 200 cc/hour out of the valve for a total of 435 cc/hours. Relatively speaking I can't get a handle on what that means in terms of pressure loss. With a base pressure of 11" WC at an ambient temperature of 60 F; I'm trying to figure out of this valve is within spec.

Any help and/or advice will be appreciated. Thanks...

 

[JBA]

The first questions that come to mind is whether or not the ambient air temperature is constant at the beginning and the end of the test and if any part of the piping system, including attached piping and equipment exposed to radiant or conductive heating or cooling during any period of the test from the sun or another source such as a nearby furnace?

 

[CWatters]

+1 on the temperature issue.

Is the electric valve directional? Some valves will appear to work but leak if installed the wrong way around.

 

[CosmicDriftwood]

The first questions that come to mind is whether or not the ambient air temperature is constant at the beginning and the end of the test and if any part of the piping system, including attached piping and equipment exposed to radiant or conductive heating or cooling during any period of the test from the sun or another source such as a nearby furnace?

Thanks JBA for you your reply. Your comment was right on track as it turned out. Wish that I had gotten back to checking for replies to my post sooner than I did because I concluded that I have indeed been chasing a ghost and that ghost had a lot to do with the "attached piping and equipment exposed to radiant or conductive heating or cooling during any period of the test from the sun or another source" as you indicated in your reply. The ambient temperature in the immediate area was fluctuating to some degree but the real variable was the heat sink effect from the through the wall 3/4" black pipe coming from the outside of the building. That effect in conjunction with the barometric pressure fluctuation due to a couple of storm fronts and was impacting my manometer by as much as 3" WC. So sometime on Friday I concluded that those two effects were creating my "ghost".

My test port for the manometer connection and the WC gauge was in line after a 1/2" valve that when closed would isolate both of them from the rest of the piping. So this morning the outside ambient temperature was 28 F and the ambient temp inside the building was 48 F, and the 1/2" valve had been closed off for 48 hours. The manometer read a negative 9" WC (that's a -9"). So you nailed it and for me this ghost said "boo". There's no leak.