Increasingly wild pressure fluctuations as hydronic system increases temperature

np_mdbr

Hi folks,

I have a hydronic heating system that serves both a hot water tank, a heat exchanger coil at a blower, and an in-floor heating system. The boiler is a high-efficiency HTP brand running off propane. The system is not connected to city water. It has a feeder pump with a tank. The fluid is a mixture of distilled water and hydronic glycol solution.

This heating season, things started out OK - domestic hot water and heat were working pretty well. Then one night, I was woken up by horrific banging sounds in the boiler room. I went in and observed that the in-floor heat and domestic hot water were calling for heat. This is normal, but the pressure gauge nearest the boiler was bouncing up and down wildly between 0 PSI and ~15 PSI. This was the cause of the banging, and the HTP boiler was in process of shutting itself down due to a low pressure condition.

Talked with a local heat guy and he immediately suspected the expansion tank, an Extrol #90 14 gal. model. I did the tap test on the tank and it sounded good. I then depressurized the system down to 0 PSI and checked the pressure on the tank. The pressure was below spec (5-8 PSI instead of 12PSI). I re-pressurized the tank to just slight over 12 PSI and re-pressurized the hydronic system to 22 PSI.

At this point all of my pressure gauges across the system agreed with each much better than they had previously (some were showing 22PSI, others 15PSI, others 18 PSI). I fired up the boiler and it got a call from the water tank for heat. I let that run and observed gentle pressure fluctuations, kind of waving back and forth between 20 PSI and 23 PSI.

I let the hot water finish. Then I turned on the main heat. That started out nicely and I was not seeing any pressure fluctuations. However, within about 7 minutes, I started seeing slight pressure fluctuations again, between 20 PSI and 23 PSI. The fluctuations started getting more abrupt and pressure started dropping. I let things keep running just to see if the system would correct itself. At this point, most of the noise was coming from the boiler, and it sounded like gurgling/tapping in the lines either at or inside the boiler.

I let things go until pressure fluctuations were very wild and the noise was as it had been the first time it had woken me up the first night. Now, pressure was jumping wildly between 5 PSI and 15 PSI, probably back and forth between the two multiple times per second.

I had the wife shut off the main heat. At this point, the pressure gauge spiked up to 28 PSI which caused the relief valve to rather violently expel fluid. Immediately after the spike, the pressure stabilized back down to 23 PSI and no more fluid was expelled from the relief valve.

Could this be a bad expansion tank? I checked the pressure again at the tank and it was still holding pressure (now at 22 PSI after I had filled the system). Can these tanks slowly go bad, and if so, does the pressure fluctuation correspond to a tank that is going bad?

I'm trying to think of other things it could be, but am turning to this forum as I have already gleaned much good info from reading posts on this, and other topics.

Thanks much.

Ironman

The air pressure in the tank should be the same as the cold static fill pressure setting of the fill valve. 12 psi is only good for a one story house.

However, what you’re describing sounds like steam flashing in the boiler due to insufficient circulation.

hot_rod

Try setting the tank a few psi lower than the fill pressure. That will assure fluid can flow into the tank, that the diaphragm is not stuck against the nipple.
The tank needs to be isolated to do this.

I suspect you have a partial air lock in the boiler. Both fire tube and water tube can be tricky to purge. I've learned that holding the relief valve open for a few seconds works best on fire tube. Plenty of purge pressure with water tube.

Your challenge will be getting a good purge while reclaiming your glycol purge water. You almost need a 1/2 hp purge pump or cart to recirculate the glycol

Does it have a primary pump on the boiler, be sure it is actually spinning and moving fluid, sounds like Bob may be on to the lack of flow. What size pump on the boiler the higher head, higher rpm type are more prone to cavitation issues.
So either an air lock or non pumping pump

Assuming it is piped properly.

STEVEusaPA

np_mdbr said:

...I did the tap test on the tank and it sounded good.

"Tap test"...lmao

HomerJSmith

Yes, flashing! Could be because of over firing. Combustion analysis is needed?

hot_rod

Does it have a zone valve under the hood to switch from DHW to heat? Maybe it is not allowing flow, or enough flow in heat mode.

np_mdbr

Thanks for all of the replies!

I have some questions about what to try, and in which order.

@Ironman and @hot_rod suggested that I increase the pressure on the expansion tank to match, or come close to, the fill pressure of the system, which is 23 PSI currently. Would I need to first bring the system down to zero again and then fill the tank to 23 PSI, or can I use the ball valve to isolate the tank and then add air?

@hot_rod also suggested purging the boiler to remove any trapped air. Are there any good walk-throughs on doing this? I do have a 1/2 HP cast iron pump that I've been using to drain and fill the system and it works well.

There is a primary pump for when the boiler is running, then there are secondary pumps for the floor zones, the heat exchanger, and the DHW. I checked that the primary pump was running by listening to it using a screwdriver on the housing and then up against my ear. It sounded pretty clean (no cavitation, etc.)

@HomerJSmith I'm pretty sure a combustion analysis is beyond my scope of expertise and the equipment that I have, but I will say that this unit hasn't really been given much attention at all since we moved in, which was 4 years ago. I'm also not sure if the previous owner gave it much attention either.

Would opening the unit up and cleaning the exchanger coil, just doing a basic inspection be prudent in this case?

@hot_rod, yes, there's a switch that senses if the call is from DHW or heat zones. If DHW calls, the pump for the water tank starts up but the other pumps continue to run (I *think* this is the case but I'd need to look at the schematic or manual to verify).

So, should I try things in this order, or another?

1. Increase pressure on the expansion tank as some suggested, to nearly match system pressure, which is currently 23 PSI.
   
2. Do a boiler flush.
   
3. Clean the heat exchanger?
   
4. Do a combustion analysis.
   

Ironman

I’d get the pressure right in the expansion tank and confirm that there’s proper flow through the boiler. If not, it could be air or a bad pump.

23 psi is a high cold fill pressure. How many stories is your house?

Jamie Hall

The "tap test" is totally useless in a bladder type expansion tank. Even if it turns out to be flashing in the boiler, which it certainly sounds like, you should test and recharge (if it's OK) the tank correctly.

np_mdbr

@Ironman, the boiler and expansion tank are on the ground floor (let's call it "first story"), and there are two stories above that floor. The in-floor piping on the third story is approximately 16 ft. above where the expansion tank is installed. The expansion tank is installed ~9ft above the surface of the first story. The boiler sits on the floor of the first story.

np_mdbr

@Jamie Hall, thanks for that tip. I have seen a number of videos that rather cavalierly claim that the tap test "means it's good!".

EBEBRATT-Ed

So you have the highest piping 16' above the expansion tank. 2' =1psi so you only need 8 psi Plus a couple of psi at the top of the system but fill it to10- 12 psi. Make sure the expansion tank is pressurized to10- 12 on the air side when disconnected from the system. Over pressurizing the system is like operating with to small an expansion tank.

Everything should be set at10- 12 psi. Air pressure in tank, glycol fill/makeup water and boiler fill pressure should all be 10-12 when it is cold. When the system is hot it may rise to 22-24 psi. If it goes over that you need a larger expansion tank. (I am assuming the boiler relief pressure is 30 psi

mattmia2

It sounds a lot like there is air in something that hadn't had a call this season until when this started(and maybe the pressure in the tank is a bit of a red herring, or maybe a slow leak in the diaphragm was the source of the air). What kind of air elimination is in the system, is there an automatic vent and separator at the boiler? If there is, try circulating the heating zones one at a time without firing the boiler and see if you get air coming out the air eliminator and a drop in pressure.(this may have caused something to become air bound so it won't circulate.)

np_mdbr

@EBEBRATT-Ed, I will try this tomorrow after having the system shut off overnight. Hot water production is the only thing I'm allowing the system to perform at this point, and that's been going fine so far. Yes, the relief valve is 30 PSI.

@mattmia2, there are air bleeders all over the place in the boiler room, probably 10-12 of them. There is one at the boiler, yes. An air leak from the expansion tank was one of the thoughts I had as well. When you say "circulate the zones without firing the boiler" do you mean hook the pump up and circulate that way? As it's set up right now, the in-circuit pumps won't fire up unless they get the command from the boiler. So I assume that means the boiler needs to be firing and sending the signal at the same time.

I am also thinking there is air in the system at this point, but I am going to lower the overall pressure, get the expansion tank to match, just a bit below, and see how that goes.

mattmia2

i think there is a circulator test in the installer menu that will let you run the circulator without firing the boiler. you also might be able to set the dh firing rate at 0%

are they just manual bleeder valves or are they automatic vents? are any of them microbubble separators or just a vent? If they are automatic and there is a lot of air you will hear it vent and see the pressure drop as you run the circulators.

np_mdbr

Reporting back on activities from today. I took a lot of pictures, many of which I will attempt to post here, but I wanted to get some words down on what happened first.

I drew the system down to 0 PSI again and checked the PSI at the expansion tank. It was 12.5 PSI. I then filled the system to 16 PSI. This was lower than the system was previously filled to, which was ~23 PSI.

Switched the boiler on and as expected, it got a call for DHW. That ran perfectly. No pressure fluctuations at all. I did notice some noise at the primary pump and took a sound recording of that, which I'll post. The cycle finished within 10 minutes. On shut off, pressure remained exactly as it was previously, right at 16 PSI.

Then, I turned on the thermostat for the main room, which includes both the blower system and one in-floor zone. As soon as those pumps became active, pressure dropped to 10 PSI. And then the boiler came on, and pressure dropped to 5 PSI, and then soon down to 0 PSI. At this point, the boiler threw a "low water pressure" code and turned itself off.

This makes me think that the previous system pressure of ~23 PSI was deliberate in an attempt to override this boiler shut off due to low pressure.

Open to ideas and suggestions. At least we are getting hot water. But the house is cold. We have a full week of 15F outside, but we're just bundling up.

Zman

I think we need some pictures of the piping. What model boiler? Did the system run correctly for a period of time and is now malfunctioning?

mattmia2

After the pressure dropped to 0 did it go back up when the circulator stopped? If yes it is a design problem or something is closed or clogged. If it stayed at 0 (assuming there is no auto fill) then it is air.

hot_rod

Kinda sounds like the expansion tank relationship changes when it goes into heat mode. Somehow you are pumping at the PONPC in heat mode? An as built piping drawing may help.

np_mdbr

@Zman the boiler model is an HTP EL-150N.

@mattmia2 yes, after the circulator stops, the pressure returns to the same value it was at prior the circulator going active.

I am going to put some pictures together here soon and will post. I will also attempt to get a diagram together.

mattmia2

Then it is what @hot_rod says, it is pumping in to the expansion tank and increasing the pressure there and decreasing the pressure on the suction. Could also be pumping in to an air pocket somewhere.

Zman

Zman said:

I think we need some pictures of the piping. What model boiler? Did the system run correctly for a period of time and is now malfunctioning?

The piping configuration on that type of boiler is super important. Pictures and when did it start?

np_mdbr

@Zman This all started after we had new A/C system installed over the summer. What I think may have happened was air was let into the system and was never purged. However, previous to this event, our heat never really worked well, but it was never at the point of making the insane pounding and pressure issues that we're seeing now.

As promised, here are some photos. Please let me know if you need more detail in any area. I'll be happy to provide additional visual input.

np_mdbr

I am putting together a full schematic of the system just in case the pictures don't do justice.

hot_rod

Where does the line coming down from the expansion tank connect into the piping? Looks like it goes down to the fill tank.

The motorized valve, is that a radiant cooling component?

np_mdbr

@hot_rod Here are some more pics of the expansion tank lines and components near that line. The motorized valve is somehow controlling the lines to and from the floor zones, but I'm not sure how. I included some close-ups of that, and of the control modules that I believe govern how the valve behaves.

neilc

is that an expansion tank, with bladder and air stem? is that feed line vented?
or,
is it an ole fashion compression tank up there?
if compression, then all your air venting needs to be closed after initial fill, and before operating,
air vents defeat the air cushion in the compression tank

mattmia2

You can see in the picture it is a diaphragm tank, it is just mounted a little like a compression tank.

Is that very crusty from condensation auto vent working? There are a lot of places for air to hide in this system.

Is there an auto vent on the boiler that we cant see? Is that auto vent without the separator in the return working? Is the cap loose?

mattmia2

Are you sure the boiler circulator for DH is working?

neilc

mattmia2 said:

You can see in the picture it is a diaphragm tank, it is just mounted a little like a compression tank.

pretty dark up in that corner,

that line feeding the tank should be vented,
it's a big air trap the way I see it now,

neilc

neilc said:

that line feeding the tank should be vented,
it's a big air trap the way I see it now,

dismiss my saying the above,
that tank should go down to the floor, or under the separator, so the above isn't an issue

pedmec

you need some better pics from a side angle so you can see more of the piping on the left hand side of the boiler. can't see how it all ties together.

Where does the 1" behind the boiler go. The two lines with the green handle ball valves

mattmia2

neilc said:

neilc said:

that line feeding the tank should be vented,
it's a big air trap the way I see it now,

dismiss my saying the above,
that tank should go down to the floor, or under the separator, so the above isn't an issue

This really isn't an issue. It will trap air. The air will slowly come out as water moves in and out of the tank and as it dissolves in the water. It will get removed by the air elimination. It won't cause an issue with the performance of the system.

Ironman

Please take some more pics from further back that encompass as much of the piping as possible.

hot_rod

The location of the pressure gauge you are reading has a lot to do with what the pressure can be expected to do. At the suction side of a circ, the gauge will drop below fill when it starts. On the discharge, expect it to go above the static fill. Ideally a gauge at the expansion tank connect would be the best indication of the static fill.

So you are looking at static and dynamic pressures in closed loop pumped piping systems. Pressure going above is common, and acceptable. Within reason " Reason being 30 psi!

I will say that determining the best, or a good place to connect the expansion tank into a system with so many circulators can be a head scratcher. It takes some piping calculations to know that exactly.

An example of a system not unlike yours, less pump intensive.

Blue line is static fill. Red dashed is the system curve, shows the delta P, (dynamic pressure) the circ adds.

Notice on one graph, the second attach, the pressure drops below static fill after flow through a couple high pressure drop devices, the boiler and indirect.

Simply moving the tank to the indirect line, cause the system to be at or above static fill. That moves the PONPC to a better location in the piping so the circs ADD their delta P.

In extreme cases with high pressure drop, or high head circs, you could pull sub-atmospheric conditions in portions of the system. If an auto air vent was located in that "pressure zone" you suck air into the system.

Graph are from an article Siegenthaler wrote on Picturing System delta P. Probably online still.

np_mdbr

@hot_rod Thank you for that detailed explanation, I'll need to take some time to ingest all of the data there.

@mattmia2 Yes, the oxidation on the pipes is due to the AC creating condensation on them during the summertime.

Attaching more pics of the system, and hopefully this gives a better idea where the pipes go.

This is the expansion tank connection going into the zone lines. I'm not sure what the component is there, it seems like an air separator that got "decommissioned"?


Here is how the tank is connected to the system. I do not see an air separator anywhere in the line that leads up to the tank.


Here's behind the water tank - the bare copper is for DHW, the insulated lines go to and from the floor zones.


Some more detail of the boiler piping...


One more boiler piping detail pic...

np_mdbr

@hot_rod What you said about pressure changes based on push/pull and location of the gauge makes a lot of sense. It matches up to what I'm seeing when the circulator pumps are running.

np_mdbr

@mattmia2, I'm not totally sure what an "auto vent" is... are you talking about an air separator? If so, there is on on the boiler at the right rear - it's silver in color. I checked the cap and the cap is loose. How would I go about testing an auto vent to make sure it's functioning properly? I am pretty confident the DHW circulator is working because our hot water has been working fine throughout this process. It's just the rest of the system that presents all the weird pressure issues.

@neilc I'm providing some more detailed pics of the expansion tank area. It looks to me that it may not be vented as you were asking.

mattmia2

That is a microbubble air separator. If it isn't working your air elimination will be poor. It has a chamber and a coalescing media that helps entrained air come out of solution and be vented out the vent. The other 2 automatic vents don't have the chamber to encourage the air to collect at them.

mattmia2

If I am tracing the piping correctly there is a second boiler circulator that is just for domestic heat, that is what causes circulation through the boiler on a domestic heat call. It is the circulator to the left of the boiler mounted vertically near where the lines branch off to the domestic heat manifold.

neilc

that's an air separator with auto vent where the tank is connected to its bottom,
whether it works or not, as it looks kinda ugly, is another question,

and you have a couple circs, and vents scattered,
in the last picture, I think I'm seeing a circ, with a vent above it, on the suction side of the circ,
this may be pulling air in,
I would close the vent cap on top of that threaded steel,
and any other circumstances where the vent is before the circ,

you get away with the vent at the tank connection as that is your PONPC, and that should stay at 12, or whatever the tank and fill pressure are set at,

other vents, on suction of side of circs, should get closed.