So stumped. Anyone can help me troulbeshoot lack of circulation in the radiant heat loop?

Lagerbaer

Okay, this is going to be a long one...
The TL;DR is that hot water isn't getting to the floor zones.

But let's start from the beginning.

The system setup:
1. Radiant heat with a single pump for three zones.
2. In terms of positioning, the hot water out the boiler goes to the hydrostatic mixing valve (the other port of the valve would be the return flow of the floor heating). From the exit of the mixing valve it goes straight to the circulator pump (Grundfoss; a common model) and from there to the manifold with the three zone valves

The symptoms and other observations:
1. Floor isn't heating up despite thermostats calling for heat. All three zones affected equally.
2. Domestic hot water (showers etc) is working fine; burner cycles on as needed
3. When thermostats are calling for heat, the pump is "on". At least, I can hear it.
4. However, no warm water reaches the pipes after the pump; there's a thermometer installed right after the pump and it reads something like 25 Celsius (77 F). Those pipes feel cold, anyway.
5. The pump itself feels very hot, as do BOTH entry ports of the mixing valve.

The history:

1. In 2020 the circulator pump failed and was replaced; no issues after this until...
2. November 2022 the boiler's gas valve needed to be replaced. Fine
3. A few days later, no heat. Plumber came and replaced the circulator pump, again!?
4. Things seemed to be working, but heat was disappointingly low. Plumber came again. Also replaced the mixing valve. Things were working well from then (that was early December)
5. Now again no heat coming, no water circulating. Plumber came. Flushed out the whole system, checked the mixing valve. Looked at the pump (not sure how exactly, but said he took off the cover and saw it moving?) and declared that he was floored and puzzled and stumped. Suggested he talk to his boss or that we might try our luck with someone else...

Questions...
So this is all really weird. If the previous things (brand new pump and mixing valve) hadn't happened, then obviously my first instinct based on the above symptoms would be that the pump was broken because it seems that's where the hotness stops. But it could also just be heat conducting from a blocked or seized up mixing valve. One random thought I had: Could the check valve be broken and stuck open such that hot water from the boiler comes to the cold water port of the mixing valve, making it permanently closed?

Or just based on folks' experience, what other culprits could there be? I don't want to have someone replace the pump and mixing valve again only to have everything break again after two more weeks of operation.

An additional clue (red herring?) is that we had the hardwood floor redone; that involves removing the top layer of the hardwood floor with some heavy machinery. Could the vibrations from that have let loose a bunch of buildup and debris in the piping, causing such massive problems?

Help...
What can I check? What's the equivalent of "did you forget to plug it in" stupid mistake we could be overlooking? Should I stick with the plumber who by now intimately knows our system, or see if someone else knows more about these systems? Should I replace the whole system (except, I hope, the floor pipes)?

The one thing to note is that anything that would be fundamentally flawed in the system should probably have caused issues to come up in the almost 20 years that this system has been in operation already...

Mad Dog_2

Geez. Good detective work so far. Can u post pics of all the piping?   Also. How many years did it work trouble free?  Mad Dog 🐕 

kcopp

What tubing is under the floor?

Jon_blaney

Bad check valve at pump?

hot_rod

What type of radiant tubing do you have? Is it a rubber tube, black or orange in color?

tim smith

The very hot circulator concerns me. Also could be stuck check valve in pump. Gunk in mixing valve inlet?

HomerJSmith

The variety of comments shows that there is a lack of understanding of your situation. I, myself, is having a hard time understanding. You should post pics of your sys.

Not getting heat in the radiant loops indicates no flow. An abnormally hot pump indicates that it is turning, but no flow. So, NO FLOW. Why??? The fist thing that comes to mind is that there is air in the circuits. That there is no flow of supply water to the pump, a restriction possibly, or that there isn't a discharge of water from the pump, restriction (blockage), a misadjustment of the mixing valve. I would want to know more about the mixing valve and how it is piped.

Do you mean thermostatic mixing valve, not hydrostatic? Make and model of your mixing valve and pump and manifold and boiler and, and , and would be helpful.

I take into consideration the homeowner's understanding of the problem, but less so of his conclusions. I am fact oriented and rely upon my own judgements of the facts that I observe. I need more facts.

Lagerbaer

A big thank you to everyone chiming in so far!
Indeed, I have wondered if the check valve might be an issue. Did I mention in the writeup that, when the mixing valve was replaced (early December) that apparently a bunch of gunk had collected (photo attached as well. The one shown is the same make / brand as the new one (Taco brand).

EDIT: I've uploaded the images https://imgur.com/a/dtrQVIF here on imgur so they don't clutter up the whole page and so I can add captions and all.

I have also uploaded a video walkthrough because with all the different pipes it can be hard to trace in the images: https://youtu.be/p_Z5kd1ebmA

The floor heating pipes are those red PEX (is that right?) pipes.

@tim smith Indeed, there was gunk in the mixing valve when it was removed; also at the recent plumber's visit there was gunk and he washed it out. He flushed the whole system for a long time and says he was able to get water through all the parts and zones.

@HomerJSmith Indeed, thermostatic mixing valve. Photo of it is attached. Taco 5000/3? Pump is the Grundfoss 15-58. The attached photos hopefully show you all you want to know about the piping.

@Mad Dog_2 The whole house had a big reno in 2005. No issues with the floor heating that I know of. My mother-in-law was very diligent about keeping all sorts of records on repairs.

If there was gunk on the mixing valve, where else would there be a detrimental buildup of gunk? Could that just cause the check valve to be stuck open, letting hot water from the boiler hit the "cold" side of the mixing valve?

Thanks everyone, and a happy new year 2023!

hot_rod

Really no need for check valves when you use a mix valve in hydronics. Looks like the one side is double checked, both in the valve and an upstream swing check?
Has the swing check been opened and cleaned?
Sometimes small strainers are included with checked thermostatics, usually in the tailpiece somewhere.

A very hot circ is either locked, or the loop is air bound. You could loosen the 4 Allen bolts, see if air, or steam burps out, carefully.

If the system keeps bunking up, maybe run a hydronic cleaner for a few days, flush and refill.

Lagerbaer

@hot_rod Indeed, there's a swing check here. Most visible in this picture https://imgur.com/MwNXHXg

It's to the right of the mixing valve and to the right of where the return flow comes back. My naive understanding is that this prevents hot water from the boiler hanging out at the cold port of the mixing valve.

It has not been opened and cleaned; looks like it's awkward to get isolated based on where it's sitting in the whole circuit, but it might need to be done...

Just for my own education, how would a failure at the check valve, either being stuck in one or another position, or just being gunked up, lead to the symptoms I'm experiencing?

hot_rod

If the mix valve is an ASME 1070 listed valve it has to close flow off if it loses H or C supply. A scald feature. So if somehow flow is stopped on either port, nothing flows from the mix port.

That could explain the hot circ motor also

HomerJSmith

What kind of gunk are we talking about? I had a job where the distribution pump was so oversized that when one zone valve was open the flow exceeded the gpm for that sized pipe and pipe erosion occurred filling the whole sys with eroded copper gunk.

I had to install a new boiler with a proper P/S sys, but before I could commission it I had to flush the old sys, pex and copper tubing, 5 times and 2 times with cleaner before I filled it with conditioned water. So, you may have to flush it more than once.

Not all mixing valves are equivalent. CV of the mixing valve is important. That is the flow thru the fully open valve. Also, important is some mixing valve have to have a temperature differential between the hot and cold ports to function properly. I prefer a valve with a high CV and sensor controlled. Thermostatic mixing valves are a cheap, dumb valve. In some applications that's all one needs.
Perhaps your mixing valve is out of adjustment, remove the knob for adjustment.

One should pump away from a mixing valve, not into a mixing valve.

hot_rod

Trade offs with thermostatic mixing valves as @HomerJSmith mentioned Cv

To low of a Cv and you get pressure drop beyond what the circ can over come, 3 Cv is good for maybe 6 gpm

Thermostatic mix valves also have minimum flow requirements. So if you use a high Cv, maybe a 7 or 10, you may need 3 or 5 gpm for the valve to mix properly. So read all the spec. On a zoned system, a high Cv may not mix at low flow conditions

Also most thermostatics need at least 20 degree hotter water on the H then the desired outlet to work properly. Some brands want 25- 27 delta

Lastly the pressure ratio from H to C needs to be within range.

So it is really a smart, but simple valve to be able to know and adjust to those 3 conditions.

A dumb valve just has a handle, no sensing component. In my way of thinking🧐

Most commonly the cartridge inside plugs or sticks from water conditions

tim smith

I looked through photos and video. Two things, sprirovent screen may be fouled to such a point it’s impeding flow. 2nd , I think the pump is a 1558fc, couldn’t quite read the model. I would yank out the integral plastic spring check out of outlet of pump. I have seen these hang up on there plastic piston and or on the face of existing flanges. Does not look like spring check needed at all.
Hope some part of this helps.

Jon_blaney

The tech installed a pump with a check valve when no check was required. That valve is fouled. Pull the pump and get ride of the check valve. The 2020 pump may have also had a check when it was not required. You serve guy is a parts changer.

hot_rod

it could be an air bubble also, after opening the system so many times some loops could be airlocked, especially if it worked in the past 

sounds like you have 4 check valves now? You really don’t need any of them for the system to work. They are maintenance issues

i would purge the radiant loop next, before replacing more parts.

Lagerbaer

But the swing check, isn't that supposed to keep the super hot boiler return water from mixing with the not really hot radiant heat return water, which would confuse the mixing valve?

Lagerbaer

Sorry, wrote the last one on mobile. A few more points. I think at this point my best bet is likely to get a dedicated floor-heating-hydronics expert in because it seems in addition to the acute issues there might be long-term issues (though again, it worked fine for almost two decades with what it had...); and I don't trust myself doing a hydrostatic cleaning properly.

@hot_rod The last time our plumber worked on the system, he did a full flush of everything so while it's not impossible that there's air, I would suspect a different culprit for now, or at least in addition. He says he was able to push water through all the parts; mixing valve, pumps, radiant loops. It's just that no hot water gets actually pulled.

@HomerJSmith I don't think the pump is in the wrong way; the motor part is just rotated on the case to get the terminal box to a more accessible location than the standard orientation would have. Or would there be an easy way to check which direction it wants to flow?

HomerJSmith

Look at the volute (back). The bulge of the casting is the input to the pump.

SgtMaj

Looking at those photos, I just had a glimpse into the brain of the installer. It gives me a headache.

hot_rod

Lagerbaer said:

Sorry, wrote the last one on mobile. A few more points. I think at this point my best bet is likely to get a dedicated floor-heating-hydronics expert in because it seems in addition to the acute issues there might be long-term issues (though again, it worked fine for almost two decades with what it had...); and I don't trust myself doing a hydrostatic cleaning properly.

@hot_rod The last time our plumber worked on the system, he did a full flush of everything so while it's not impossible that there's air, I would suspect a different culprit for now, or at least in addition. He says he was able to push water through all the parts; mixing valve, pumps, radiant loops. It's just that no hot water gets actually pulled.

@HomerJSmith I don't think the pump is in the wrong way; the motor part is just rotated on the case to get the terminal box to a more accessible location than the standard orientation would have. Or would there be an easy way to check which direction it wants to flow?

Should be a flow arrow on the pump volute

Lagerbaer

So the arrow I get, but I remember in the pump instructions it talks about rotating the terminal box. So if the arrow looks wrong on the pump casing but the motor is installed correctly, would that be okay? 

hot_rod

Yes, the motor spins the same direction regardless of orientation. The arrow cast into the body doesn’t lie.

HomerJSmith

The arrow on the volute is the direction of flow as hot_rod says.

Lagerbaer

All right, I probably owe an update on this

So, previous technician left perplexed and told us he'd get back to us after consulting his boss, but never did. Called in another company, they sent their rookies, they flushed the system (again) but that didn't help so they left, perplexed.

Finally got someone in. Owner of his own company, old guy who knows his stuff. He goes to flush the system, immediately realizes that this one section (between the swing check and the radiator) can't be flushed based on the pipe layout. So he cuts the pipe and installs an air vent, checks swing check in the process, all good there. Notices that the mixing valve wasn't set properly either. Then notices that the pump is broken due to a manufacturing defect where a cable came loose. Puts the old pump in (that one of the previous techs replaced with that new, defective one...) and now everything has been working for almost a month 🎉

ratio

Glad you got it resolved, & thanks for the follow-up.

Too bad it took so long to find the right guy, but it sounds like you've got a keeper now! Did you ask him if he reads heatinghelp.com?

archibald tuttle

For future reference, in following the thread, I think you said the pump was 'making noise', although it sounds like your consciencious contractor found a connection in the onboard wiring that had failed . . . so would anticipate that the pump wasn't actually working and maybe you were hearing one of the other pumps or even a transformer vibration or something. It can be very hard to tell if a pump is actually running, as even when that pump is energized the rotor can be locked. On large systems I put a pressure gauges up and downstream and watch for them to change when energizing the pump. But, noninvasively, you can use an inductive ammeter to see if the pump is actually drawing power, as in this case, just testing the 120V was going to the pump terminals wouldn't necessarily tell you the pump was on if an internal wiring connection had failed. The ammeters are relatively inexpensive and you just need the cover off the pump so you can clamp over one leg of the 120.

Final observation, while I can't absolve your early servicemen for failure to accurately diagnose and fix, that installation seems efficient in size but no picnic to access. Not really sure there is anything to do about that and since it all worked for some time, one can hope you don't have to get back into it anytime soon, but if anyone is actually doing anything to the system (like say removing redundanct checks), etc. thinking of a layout that doesn't pinch the mechanicals between the boiler and the indirect tank would make it potentially easier to get to components and diagnose.

Lagerbaer

@archibald tuttle

Haha, fair enough on the layout. Maybe the right thing to do, when the whole system craps out or another big repair would be required, is to rip it all out and put a much smaller on-demand boiler system above ground.