Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

Testing leak-by on gas valves

aircooled81 Member Posts: 205
Preparing to record running preasures on natural gas boiler. Noticed the pressure after the primary main gas valve was allowing pressure to increase though it was off. So I checked after the 2nd and 1st stage gas valves and they were leaking past too. ?
So it goes, 1" gas into a honeywell V48A2169-3, then after that the gas train tees into 2 other GV's, 1 is a honeywell V48A2151-3 (3/4") and the other is another V48A2169-3 (1"). After they tee back together on their way to the manifold, there is an additional 1" ball valve with test cock.
I put my guage on the test cock and with the entire gas train off, I witnessed the gas pressure rise 0.44"wc in 3.5 minutes, and then checked on it 15 minutes later and it had risen to 1.65"wc and still climbing.
Now the room temp was a steady 74*F, and the boiler had not run all day, actually all week. So I dont think this can be attributed to heat.
Is there an exceptable leak rate for a gv, do honeywell gv's bleed by for any particular reason?
The inlet natural gas pressure to the regulator is about 8.5"wc, and it steps the pressure down to 4.5"wc when flowing.

I think I need to yank these GV's, check the seats if possible (?) or just replace them. Any thoughts?


  • ratio
    ratio Member Posts: 3,627
    According to Honeywell, there's a allowed amount of bleed-through, on at least some of the valves I've read about. They give a checkout procedure in the paperworks. Here's the manual for the V48A valve, check the last page. I'd imagine the other valves are similar, but check.

    I don't thing you are supposed to open the valve & check the seat unless there's a checkout/test procedure for it. I'd imagine doing that would interfere with any kind of listing unless the mfgr permitted it in writing.

  • Tim McElwain
    Tim McElwain Member Posts: 4,625
    Because the V48 is a single seated diaphragm type gas valve there is an allowable loss through the valve. It is well below the Lower Explosive Limit (L.E.L.) for both natural and propane gases. This is why on commercial CSD-1 applications we have two gas valves.

    All residential heating valves since 1979 are redundant dual seated valves and they are usually a combination of direct acting solenoid and then a diaphragm valve. This insures a more positive shut done on the direct acting solenoid which is the first valve in the residential gas valves. Some companies (White Rodgers) use two solenoids which have a more positive shut down than diaphragm gas valves to insure less leakage

    There is however an allowable leakage under ANSI Standards for gas valves but it is well below what is detectable as to odorant added to the gas.
  • Jamie Hall
    Jamie Hall Member Posts: 23,322
    To which I might add that even a very small bleed through on one valve will raise the pressure downstream, if there is another valve beyond which has less bleed through... doesn't take much difference.
    Br. Jamie, osb
    Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
  • aircooled81
    aircooled81 Member Posts: 205
    @Ratio, thank you for the litterature. That is phenomenal, I will go back and count the bubbles to see if the leak rate is excessive. How cool is that!
    @Tim McElwain , @Jamie Hall , thank you both for your replies. I will be adding this stuff to my boiler annual check lists, it only takes a few moments to set up some tests like this.
  • ratio
    ratio Member Posts: 3,627
    No problem. I actually took the plugs out of the test ports on my boilers & installed Pete's Ports, just to make that kind of thing easier.