Help with varying pressure on radiant floor system

infloorradiantheat

Jamie,
Yes that is what we have. All three have their own expansion tanks, air bubblers, temp and pressure gauges etc. They are separated by heat exchangers. A pressure change in one of them does not affect the others.

infloorradiantheat

Pressure has continued to drop and no signs of wetness anywhere.

Thanks Zman for the following recommendation. We are having them come out on Mon. to fill the system with gas and see if we can find a leak.

Zman said:

These guys specialize in both listening and helium testing and would be able to help you out. https://www.americanleakdetection.com/

Still don't know why the expansion tank isn't functioning like our prerusting out one did.

We did have a hydronics pro out on another matter and quered him on methodology of working with the thermal imaging camera. Still no luck with that method.

infloorradiantheat

Does anyone think tanking more air pressure off the expansion tank would keep the pressure from getting up to 24 psi?

To recap we set it to 20 psi prior to installation. We then let off about 2 psi to see if helps. When we took 2 psi of we isolated the tank but did not drain the pressure off since we didn't want to allow any air into the system.

infloorradiantheat

We started measuring the loop temps at the manifold, both coming and going. They vary between 110 ft to 190 ft and then two 300 ft. The two longest ones don't seem to get up to the return temp of the shorter ones. Could the flow be restricted by air bubbles trapped in the lines?

It could just be that the longer length leads to more disappation. I'm not sure if we should try to flush this zone again or just leave it for the gas guy to check for leaks on Mon. He has to empty the lines to put the gas in so we may be facing an air in the lines problem again.

Jamie Hall

You should expect the return temperature of the longer lines to be lower. First place, there's a lot more area to heat. Second, the flow rate may be lower since the head loss is much more.

Whether this is desirable or not is another question. Do they heat the areas they are in adequately? If so... you're fine. If not, you may need more flow.

infloorradiantheat

Thanks Jamie. I think I was just wishfully thinking that it was air in the pipes working it's way out that is causing our pressure drop. We closed the bubbler and still had some drop after the system ran so that indicates a leak. I'm nervous having someone all over the house in this time of Covid. Considering cancelling the leak detection appt and waiting for a month but that makes me nervous also as water is probably going somewhere.

infloorradiantheat

We didn't have luck with the leak detection. They weren't convinced we have a leak on the upper floors. They though it was maybe a leak in the basement zone even though we have that zoned off.

So they want to do a week test when the system is off in the spring where there is a pressure gauge on each zone and we see if any of them falls. I can't disagree with the logic of the approach but I'm disappointed that we can't move on from filling the system every two weeks.

They said if we start to see water leaking then we will know that we really have a leak.

In the mean time we are going to buy the pressure gauges and look for opportunities to put them on during the day or if we get a series of days that aren't projected to be freezing. We have a secondary heat system with our minisplits so we could actually do this in the winter on a warmer series of days.

hot_rod

If anything, you may have a pinhole leak or two. Those are not easy to pinpoint.
I'd be tempted to add a small, low flow accurate water meter on the fill to actually see how much water is leaking. Watch it for a few weeks- month.

infloorradiantheat

Yep, I felt their hesitation to tackle it. I also asked about repair as if we do locate one who the heck will tackle digging it out of the gypcrete and repairing it?

I don't think it will be easy to add a flow meter to the water fill. Also we don't leave it turned on, just use it to refill in between pressure drops which seem to be at a two week interval for about 6 psi so 3 psi per week.

Jamie Hall

Your pressure drop is small enough that I completely agree with @hot_rod : the actual leak is small. Although I have been heard to say that I've never known a leak to get smaller with time, I suspect that in this case you can probably get away with leaving it as a nuisance until summer. I hasten to add -- no guarantee on that.

Further, a small leak like that could be in the basement zone, even though it's turned off. Valves do not always close completely drip tight, and you are looking for a drip size leak.

I think your best bet to even isolate where it might be at all is a variant on the pressure gauge on each loop -- the variant being that you close all the loops (why I'm suggesting summer!) and compare the pressure changes on each loop over a period of time. Include gauging the boiler and pumps and manifolds in your isolating as well, and cut the expansion tank out of the system. Hopefully one -- and only one -- of the isolated zones will show a dropping trend, while the rest may just go up and down with temperature more or less together.

Not a fun problem.

infloorradiantheat

Jamie,
Thanks for the validation. You are saying basically the exact same thing that the guy from the leak detection company said. He also commented that we couldn't rule out the basement because it was valved off.

I really hate to give up on getting this solved until spring but it may be necessary. I did put in a question to Fernox about their F4 leak sealant to see if it would work on a leak in pex tubing. I haven't heard back yet to see if that is a possibility for getting this taken care of.

infloorradiantheat

We just ruled out the basement. We have a hose pressure gauge so we closed all the zone valves for the day on active house portions (East, West and Upper). The basement valves have been closed since our last fill to 20 psi. Then this evening we checked them all.
Basement was at 20 psi.
Upper was 9.9 psi.
East was 10 psi
West was 11 psi.

The Upper, East and West zones have all been active through the pressure drop while the basement has been valved off since we last did a fill to 20 psi. Hence the basement isn't part of this pressure drop issue.

Another odd thing we can't figure out is the pressure gauge on the expansion tank side of the heat exchanger shows 15 psi while the gauge in the return side of the heat exchanger shows 10 psi. When the system heats up the pressure gauge on the expansion tank side goes to 24 or 25 psi while the return side goes to 20 psi. So maybe both the tri indicator pressure gauges were reading wrong in the same way and the hose screw on one is the accurate one.

Jamie Hall

One step at a time! You'll get there!

I wouldn't be a bit surprised if the tridicator gauges were off.

hot_rod

A gauge on the suction side of a circulator will show a pressure drop when the circ is running. Ideally you do "static" pressure testing, with nothing running.
Once circulator start running you have different "dynamic' pressure conditions.
Ideally you want to use one gauge and move it around as needed, that eliminates the gauge error between various gauges.

Trying to get accurate gauge reading on a system that is turning on and off, heating and cooling, etc is going to drive you crazy.

And of course the higher the gauge on the system, the lower the pressure. .433 psi for each 1 foot of lift, or vice versa

My suggestion for adding a water meter was to get a handle on how much water is being added. IF for instance it is a gallon a week, let it run through the winter. If it takes on gallons per day, or more, that will cause scaling issues, address it sooner rather than later..

Here is how I add a meter for temporary trouble shooting.

Meter, fill valve, a few garden hose adapters and short hoses. Connect into the system most any place you have a hose connection, like the boiler drains. Turn the current fill off, let this setup add and meter how much of a leak you have.

Now you eliminate all the guesswork that may be keeping you up at night.

infloorradiantheat

Hot rod,
Thanks for the idea and explanation of how to do it outside of the current fill valve. We are going to think on the best way to do that for our next fill. We put it back to 20 psi last night so we should have about a week and a half to consider how to do this.

infloorradiantheat

I thought I'd follow up for anyone that reads this. Turns out our tri-indicator gauge was wrong. We tried two of the same and they were both reading wrong. We though we were filling to 20 psi but it was really 15 psi. We figured this out when we put hose connection pressure gauges on all the drains and measured the pressure at 4 different points. Once we filled the system to 20 psi, we have not had a pressure drop for the last month. So it seems that the expansion tank was functioning properly at the 20 psi it had been set to.

But why the drop we did see? The only thing I can think of is that the balloon in the expansion tank may have been stuck on the side wall (I saw this somewhere as something that could happen) and it wasn't being fully distended until we got to 20 psi, which is what the tank was pumped up for. Or maybe air in the system?

We did follow hot rod's suggestion to use a water meter and pressure fill gauge to fill the system via a hose connection. To move from 10 psi to 15 psi was less than 1/10th of a gallon. So we were losing less than 1/10 gallon over each two week period. How much less? I don't know since the gauge only had that reading. We did verify the reading first to use the system to water some bushes outside. Then stopped it when the marker turned over.

Anyways, happy to not have any pressure drop for the last month. Will continue to monitor it just in case there is a tiny leak that is only draining excess water in the expansion tank in case we over filled it. So those who guessed, no leak - wait it out may have been right. And Jaime who guessed bad tri-indicator gauge(s) was right.

We did have a scare that we thought might be our mystery leak. After pressurizing to a real 20 psi we started to get a drip noise in the wall where the upstairs pipes go. But we couldn't figure out how a drip noise could be generated due to it being a straight run from the basement to the upstairs. Any water coming out should have just run down the pipe noiselessly and eventually shown in the basement.

After some internet research we figured it was expansion/contraction of the copper pipe. Subsequently (after being tortured by the noise for 3 weeks) we noticed the pipe supports in the basement were broken plastic. We strapped the pipes up and the noise stopped. It was -8 here last night and there was a small amount of noise so we likely need to insert an additional piece of straping between the pipe and the strap to give them a little more support.

Thanks to everyone who helped! We would never have known how to trouble shoot this with out your help.

Jamie Hall

The balloon -- good term for it -- goes the other way, and if was filled to 20 psi air and you had only 15 psi in the system the poor thing was plastered against the bottom and sides of the tank. Give it a break!