I run a staple-up, hydronic floor. It was powered down between April and September of last year. Before I started it back up, I noticed there was no pressure in the system. I opened it's make-up valves, let it repressurize and powered it up. In the subsequent days there was small amount of leaking through the over-pressure valve on the boiler. I opened and closed the valve a few times, re-pressurized and everything seemed good. I attributed my earlier loss of pressure to this same valve leaking some during those six months the system was idle.

Between October and January, the system ran perfectly, behaved as expected and held pressure.

In the middle of January, I had to power down the system again. A couple weeks after powering it down, the pressure on the system had bled off. There was no sign of leaking at the boiler over-pressure valve. The expansion tank is still at its expected pressure.

I want to power it back up in a few weeks but am a little concerned. I must have a leak somewhere. Before I start a step-by-step check of every fitting a coupling, I'm looking for guidance. Where should I start looking for a leak which occurs in a static, cold system but not in a circulating, hot system?

John Thurston
Juneau, Alaska

well, hold on now.

You re-pressurized a cold system before heating it back up. then you heated it up and it blew the overpressure valve; then you went cold and pressure dropped again.

That sounds to me like an inoperable expansion tank. Either not pressurized to the pressure your system is running at, or filled with water outside the bladder due to a leak, or perhaps cut off from the rest of the system.

This should result in a noticeable pressure drop whenever the system cool down significantly though. As in, if you shut down the heat source and just circulated the water, letting it cool as it went, you should be able to see a pressure drop in that case.

the only other thing I could say is *maybe* when the system contracts from being cold, it creates a leak you don't otherwise have, perhaps a tube torques a fitting on a manifold somewhere or something like that... but you should be able to detect such a leak when the system is cold. If you can't see the pressure drop, I would go ahead and run the system, and wait until spring.. then shut down, and keep a close eye on the whole system.

quote:

Originally posted by NRT.Rob:
well, hold on now.

You re-pressurized a cold system before heating it back up. then you heated it up and it blew the overpressure valve; then you went cold and pressure dropped again.

That sounds to me like an inoperable expansion tank. Either not pressurized to the pressure your system is running at, or filled with water outside the bladder due to a leak, or perhaps cut off from the rest of the system.

I agree with that, but I've confirmed that the expansion tank is up to pressure. Also, there is air present at the charging valve and not water. And the only way the tank could be cut off from the rest of the system is by a mechanical blockage in the short T-pipe feeding it (there isn't a valve there). When the system was running, the top of the expansion tank was warm and the bottom half was cool. I don't think it's the tank, but I'm happy to hear someone tell me why I'm mistaken.

The other problem with this hypothesis is that it took a couple of weeks for the pressure to drop after it was powered down. If the pressure drop was caused by simply the thermal expansion/contraction of the water, I would expect that to occur in the same time as it takes the water to heat or cool. Yes?

quote:

the only other thing I could say is *maybe* when the system contracts from being cold, it creates a leak you don't otherwise have, perhaps a tube torques a fitting on a manifold somewhere or something like that...

That's what I suspect There isn't anything detectable in the accessible portion of the piping and fittings. There are two locations in the crawlspace with fittings. I'll have to go check them.

The amount of water required to remake the system pressure is extremely small. This means the amount leaking out is also extremely small. I fear I may not be able to detect such a small quantity. Is there such a thing as a moisture sensitive tape or paper with which I could wrap my fittings and return after a couple of weeks to read signs of leaking? I'm picturing something like a cloth tape which would change color in the presence of water.

John Thurston
Juneau, Alaska

Hmm, well you're right on all counts with the expansion tank, though it's pretty tough to check its charge level while it's connected to the system. That's why having a ball valve to isolate it is a good idea.

One easy way to detect miniscule leaks is by a bit of residue left behind, which is pretty normal even when the leaks are otherwise nearly undetectable.

A leak as small as you are indicating may not even be an issue, so don't lose sleep here, but it would be nice to figure this out.

Perhaps you could isolate part of the system from the rest, say half and half, and watch the pressure, then watch it when you un-isolate. this would indicate on which side you might look first.

Perhaps this is evaporation as well.. not even enough to be an actual water leak.

I don't have any other immediate ideas though.

.....

sounds like ,

contraction of iron boiler sections or perhaps bolts on a coil used for domestic hot water production..


small leaks often can be found at bonnets and some zone valves 1/4 turn tightening of these various parts will control the small leak without over tweaking things when you next shoot heat to it.. to get real technical you could torque to manufacturers recommendations ..

i live a bit North of you and have encountered the symptom on more than one occasions..

*~//: )