Air in boiler feed water

DIYorBust

Ok lately I've been dealing with a headache with my Navien NHB-150 setup. I keep getting E001 errors. There is some confusion as it appears some Navien models may show E001 as an ignition failure, but the literature and notes on the cover indicate for this boiler it means heat exchanger is over temp. Recording with a camera and in person, I believe this is caused by air entering the heat exchanger. After some frustration, I believe I understand the source of the air. We frequently need to drain down the domestic water system, because of ongoing construction in the building, and sometimes to protect against pipe freeze flood risk when we are away. When we do this, we isolate the heating system by closing a valve downstream of the PRV. Now my theory is that when we repressurize the system, and open the isolation valve, air remains in the pipes feeding the PRV/Heating system, and eventually into the makeup water at the boiler. The air eliminator is encountered only after the boiler primary loop. Sometimes I can hear sounds that sound like air moving through the boiler before the E001. The navien has an air vent on top, but it doesn't seem to vent this air in time. With the valves I have available, and the location of the makeup water connection, it is difficult to purge the primary loop. I replaced the air vent and it did not seem to help. There is a point after the PRV where the makeup water takes a 90 degree turn down, and drops about a foot or so before going horizontally into the primary loop where water moves away from the boiler.

I'm contemplating replacing the 90 elbow with a tee and adding a minical valve on top at this location, hoping to vent the makeup water of air before it reaching the boiler. It is half inch copper. Would this work? Is it a good idea? Should I run some pipe vertically and raise the height of the minical? Would an air scoop or separator be a better choice?

Thanks!

DIYorBust

hot_rod

All mod cons, water tube, fire tube, or variations can be difficult to purge on first fill. Most show a high point vent on or in the boiler itself. That helps but it is not 100%. High point vents are not a substitute for a central air purger.

A good air sep at the immediate supply out of the boiler is the next best place to grab air. Running high temperature for an hour or so drives air out best.

Vertical air seps were developed specifically for mod cons with piping connections out the top or bottom.

But constant draining and refilling is not a great idea. In addition to fresh O2 entering you are adding more minerals every time. Once the boiler is filled and purger it should not be taking on additional water? That could indicate a slow leak in the system.

Even with plain water a fill tank can be a good option to monitor replacement water.

DIYorBust

Thanks hot rod, keep in mind we are not draining and refilling the hydronic system, we are draining and filling the Domestic water system. The problem is annoying in particular because it takes quite some time before the air enters the boiler. Perhaps there are some slow leaks, which I will endeavor to address, and there may be some other problems we need to correct with the system, but eventually makeup water will be needed, and it will draw from the domestic side. If this were not the case, why not save a few bucks and not bother with a prv at all? We have a spirovent air eliminator in the location you suggest, the problem is that when the air passes through the boiler heat exchanger, and causes the boiler to go over temp and shut down, until I come and find a way to purge. I think the primary loop pump can't force the air down and out of the primary loop to where the spirovent is. It doesn't reach the spirovent. Purging through the flush valves can work, but has to be done backwards, and takes multiple attempts because new air come in until the line is clear. We're using the matrix supplied by navien, and I don't see a spot to install the air separator in the primary loop.

Jamie Hall

Given your slightly unusual situation… I'm not sure there is a good way to completely avoid the problem.

However, one thing you can do to at least reduce the nuisance is to make sure that your pressure reducing valve is set correctly and located properly and, if not, close the feed to the boiler and leave it closed — assuming that you have no leaks in the system causing it to lose pressure.

So.

The second approach is the better or the two, and therefore what I suggest is that you close the feed to the boiler — and leave it closed while you observe it for a day or two. It should hold pressure (if it doesn't, you have a leak on the heating side — which is another and different problem which will need attention). If this works, there's no reason to open that valve again until you are done with the work which requires shutting the water off.

However let us consider the piping and valve arrangement. The pressure reducing valve — autofill — should be as close as possible to the expansion tank, and both of them should be on the inlet side of the main circulating pump. The pressure reducing valve should be set to the same pressure as the empty precharge pressure in the expansion tank. If this is so, then there should be no water feed to the system — again, unless there is a leak…

hot_rod

One trick is to raise the boiler temperature, maybe 180 for a few hours. And also raise the fill pressure as high as possible for the same period. Boosting the system pressure basically squeezes the bubbles small and it may be enough to flush them through.

At some point the system should get to a 100% air free condition, and stay that way for years. Unless air, or fill water with air in it is being introduced.

P/S can make it harder to get all the air out. Indirect tanks sometimes suffer the same issue, the coils never flush 100% clear.

The tanks with large coils 1-1/2" for example the flow velocity can be so slow it doesn't move air with the water to the vent locations.

Purge valves are fill and start devices, you should not need to use them on a regular basis. As you do, you bring in O2 laden water.

Playing with this small piping demo. I can pump a full cylinder of air into maybe 3 gallons of water content and within minutes all the air is eliminated. At room temperature! There is no question to the 97% efficiency of microbubble air separators.

I have been building and demonstrating air elimination hands on displays for maybe 20 years now. They are always a crowd pleaser when teaching hydronic techs.

DIYorBust

I think I did make some progress here, I haven't installed any new air vents yet, but I believe at least some, and perhaps all of the time, the problem occurs not because air is entering the heat exchanger, but because airlock is occurring the the primary loop pump, and lack of flow causes water to boil in the heat exchanger. Bleeding the pump frequently seems to help get the system going. Is there anything I can do to help prevent air from getting into the pump? Is there an autobleeding pump I could use for this purpose?

hot_rod

You need to get, and keep air out of the system. A combination of manual, automatic and system air purgers is what it takes sometimes.

A pic or drawing of you piping may help.

Some circuits can be very hard to get air out of, and it will find the high spots to congregate, in many cases the top of the boiler.

The auto/air purge functions on some circs, just speeds up the circulator to try and force air back to the purger. It doesn't have an air removal mechanism in the circulator.

Some times you can boost the fill pressure to maybe 25 psi and it helps get problematic air out. Then reduce it back to normal operating pressure.

Screenshot 2025-03-07 at 3.21.56 PM.png

mattmia2

Any chance you have an expansion tank issue that is causing it to hit the relief valve pressure every cycle and squirt a little water out that then gets replaced by the prv when they system cools?

EBEBRATT-Ed

I would be more concerned as to why the boiler even needs MU water. The expansion tank/relief valve could be an issue as @mattmia2 mentioned

DIYorBust

https://forum.heatinghelp.com/discussion/comment/1847182#Comment_1847182

Yes, sort of, but I've managed that issue with minical vents at the top of this system. So the TPR valve doesn't trip. I think we take in air at the top of the system through the auto vents, but most should also vent out up there rather than come down 35 feet to the boiler. The problem isn't tripping the relief valve so much as keeping the convectors at the top of the system full. Soon I will try adding a larger expansion tank, but this newer issue seems to happen when we drain down the domestic side.

Jamie Hall

If you have 35 feet in height from the boiler to the highest radiator, your system pressure at the boiler must never be less than 20 psig (15 psig plus a 5 psig cusion). This is higher, naturally, than most systems run in buildings which are less tall.

What this means is several things. First, note that that is lowest pressure, which will be at the inlet to the main circulating pump when the pump is running. If the expansion tank and autofill pressure reducing valve are located there, that is the pressure they must be precharged to (the expansion tank) or set at (the autofill pressure reducing valve). It the expansion tank is located somewhere else, then it will have to be precharged to a high enough pressure to produce that pressure at the pump — which, if it is on the outlet side of the pump, may be quite a bit higher. This also applies to the autofill pressure reducing valve. Second, note that with this higher precharge pressure, you will need a larger expansion tank to avoid exceeding the 30 psig limit of the system. Probably one size larger than might otherwise be necessary for a similar system.

Now I note in your comment that you seem to notice this problem may be related to a draw on the domestic water side. This suggests to me that you do not have a reduced pressure zone backflow preventer between the domestic water and the boiler, or, if you do, that it has failed. A single check valve is completely unsatisfactory in this application and, while a double check might be permitted, I can't recommend its use.

hot_rod

does the fill valve for the boiler have a backflow preventer? If not boiler water could go back into the domestic side when you drain down. A pressure gauge on the boiler would be one way to know that

Draining the domestic side should not have any effect on the boiler water content ir pressure

EBEBRATT-Ed

@DIYorBust

The problem is if air keeps getting in the system it contains oxygen and that causes corrosion.

hot_rod

yes it is possible to pull air into an auto air vent

Usually you see that possibility on systems where you are pumping at the expansion tank, and have a low fill pressure or low to no pressure at the high point where an air vent may be.

A pic of the near boiler piping would help us determine that

DIYorBust

https://forum.heatinghelp.com/discussion/comment/1847202#Comment_1847202

Thanks Jamie. I do have both a #30 and a #60 amtrol tank installed. I have a multiyear thread discussing that problem in another post, and I think I am going to try installing a 20 gallon tank. 300 bucks for one of these, but if it works it works. The original installer had only the #30 tank.

https://forum.heatinghelp.com/discussion/187608/psi-at-boiler-exceeds-30psi-when-hot-in-4-story-house

I do not believe the problem I have is with heating system water being pulled into the domestic side. I believe the installer used a double check valve, I will check this next time I'm down there. What kind of backflow device do you recommend?

At any rate, drawing heating water into the domestic water would certainly be a major health and safety problem. Fortunately, I don't believe this is the case. The issue is happening not when we draw on the domestic water, but some time after we drain down the domestic system for service. When we drain down, we close a manual ball valve after the prv, which is opened again after the pressure is restored on the domestic side. I suspect the pumps and boiler may ingest residual air coming through the makeup water supply. Much like we get air sputtering through all the faucets when we first open them, I figure that air must go into the heating side until it reaches a vent, which it does not until it passes through the boiler and primary loop pump. Today I spent several ours purging the system to get it working. The purge loop does not seem to push much or any water through the boiler and primary loop. It was frustrating, but ultimately bleeding the pump for a long time got me going again. Not 100% sure this will work every time but a lot of air came out of that pump. More must have been going in because normally just a few seconds would seem to unlock a pump.

mattmia2

you probably need another valve or 2 to force the purge through the primary loop alone. manual bleeders or at least closing the caps on the automatic vents at the top of the system once you get flow established is probably a good idea. if anything produces a pressure differential in the system they could suck in air. if they are sucking in air that would cause the pressure in the system to rise. if they are sealed rather than automatic vents the static pressure being enough to push the water up there becomes less important.

DIYorBust

https://forum.heatinghelp.com/discussion/comment/1847221#Comment_1847221

Yeah I think that may be right. I think if I remove the check valve from the primary circulator, I might be able to do it, any downside to this?

Regarding the auto vents, they will installed to replace coin vents, because the radiators did in fact become air bound constantly. This solved the problem of the top floor always being cold, but I'm hoping that maybe a bigger expansion tank will help keep the pressure high enough to maintain water up there. How did the air get up there with coin vents? Not totally clear, but air got in there all the time, and was a major headache.

mattmia2

If air is getting sucked in to some other auto vent because of pressure differentials in the system that could increase the volume of the system enough that the expansion tank can't accommodate it if water is getting pushed in to the expansion tank to suck that air in. Air shouldn't just appear in the system, especially with a microbubble air separator.

Jamie Hall

At the risk of the obvious: if air is getting into the system, it can only come from makeup water or from there being a location on the system where the pressure is less than atmospheric and there is a small leak.

Problem one can be solved by reducing makeup water to a minimum — it should be zero, of course.

Problem two can be solved by ensuring that the pressure everywhere in the system is above atmospheric…

hot_rod

put a temporary pressure gauge on the highest radiator. It should always read 5 psi or higher

that test would eliminate the question of negative pressure

At the least you want a dual check with vent port on the boiler fill. The vent port lets you see if the check has failed, as well as preventing boiler water from entering the potable side. A double check can fail and allow boiler water into the potable side

LRCCBJ

Now I note in your comment that you seem to notice this problem may be related to a draw on the domestic water side. This suggests to me that you do not have a reduced pressure zone backflow preventer between the domestic water and the boiler, or, if you do, that it has failed. A single check valve is completely unsatisfactory in this application and, while a double check might be permitted, I can't recommend its use.

Br. Jamie, osb

The use of an RPZ valve for a boiler is completely undesirable for the following reasons:

1. Cost. They are typically 6X the cost of backflow prevention valve.
2. Flooding. Anytime the street pressure falls below the pressure in the boiler, all the water on the boiler side is dumped………..typically into the basement. These devices are not recommended for indoor use for that reason.

The dual check backflow prevention valve is perfectly satisfactory for any boiler installation. Of course there are municipalities who will demand more……………because they can!

The only downside to the dual check backflow prevention valve is that it is rarely checked. However, the manufacturers typically provide the proper test procedure in their literature. It does require some additional plumbing at installation to perform the test…………..something that few will accomplish.

Jamie Hall

Ah… an RPZ will not dump water anywhere unless it is failed. And I agree that they are expensive. The places I used to requite them (and I did, @LRCCBJ , back in the days when I was one of those mean horrible nasty obstructive building inspectors) was when there was a possibility of health hazard contamination present in the water on the downstream side and where the potable water side could reasonably be expected to drop below 25 psig. Otherwise I was quite happy with using a Watts 9D or equivalent.

hot_rod

If you want that extra level of protection an RPZ is the best. Legally they should be tested yearly by a certified backflow tester. That could be 150 bucks or more.

And no a RPZ will not dump the boiler water if the main pressure drops. The second check is what keeps the water in the downstream side.

The 1/2" size is not anywhere near 6X the cost.

Caleffi will be introducing a testable dual check soon. Inspectors are asking that boiler BFDs be testable, in some areas.

Jimmy does a good job explaining RPZs

https://www.google.com/search?q=how+a+rpz+works&oq=how+a+rpz+works&gs_lcrp=EgZjaHJvbWUyCQgAEEUYORiABDINCAEQABiGAxiABBiKBTINCAIQABiGAxiABBiKBTIKCAMQABiABBiiBNIBCDU3NTlqMGo0qAIAsAIB&sourceid=chrome&ie=UTF-8#fpstate=ive&vld=cid:4afb40aa,vid:W3wg11mMfb8,st:0

A fill tank eliminates the concerns about bolers back flowing also.

DIYorBust

Can you remove the press fitting from that 574 and use npt sweat fittings? I know this forum has a policy of no discussion of pricing, so let's just say the press version is more widely available. It looks like maybe you could?

hot_rod

https://forum.heatinghelp.com/discussion/comment/1847561#Comment_1847561

yes it can be sweat, thread or press, different tailpieces is all

I think it is okay to post prices of products that are online, Supply House should have all versions

mattmia2

note that a lot of the items shown as backordered on supplyhouse.com will show up in a couple days, not the weeks the backorder date says

mattmia2

I think they are looking at this:

https://www.supplyhouse.com/Caleffi-574004A-1-2-Threaded-574-Brass-RPZ-Backflow-Preventer-Low-Lead

and this

https://www.supplyhouse.com/Caleffi-574064A-1-2-Press-574-Brass-RPZ-Backflow-Preventer-Low-Lead

and seeing that the press version is about $50 less and asking if the the press version is just the npt version with press adapters pre-installed.

hot_rod

The 2025 pricebook is online. It has list prices, not necessarily the suppliers selling price.\

https://www.caleffi.com/sites/default/files/media/file/2025%20Caleffi%20Plumbing%20and%20Hydronics%20Catalog.pdf

\

The 574 combo might be the best way to go. It has more connection options.

Yes, the press adapters are screwed into the FIP.

DIYorBust

Cool thanks Hotrod, I think I'll consider adding one of those for peace of mind. Particularly since I had another thought as I was down there again today purging air, that I might want to replumb the makeup water connection. Could I reroute the makeup water line into the bottom of my air eliminator?

I figure any air or water must be moving slowing in the makeup water line, except when using the purge loop, and even then I bet it would help a lot. I'm also considering whether to add an air eliminator on the return side of the primary loop.

I do understand the consensus here is to get the air out, and then just close the system off. But at some point we will need to install some radiators, and make other changes to the heating system as part of our renovation work. It does not seem so crazy to want the system to self purge on fill, at least well enough to avoid seizing the pump or boiling water in the heat exchange, because when the system shuts down, the air eliminator doesn't get a chance to do it's job. It's a big waste of time to sit there trying to get things going and get the air out, but also in the winter this can present a risk of serious damage to the plumbing or building if the boiler shuts down and I am away for a few days.

mattmia2

normally the air that comes in with the fresh water is dissolved in the water and doesn't come out of solution until it is heated in the boiler.

once you get the water circulating after replacing a part if you have a microbubble air eliminator in the right place it should get rid of the last of the air. even if the makeup water feed is feeding air, it isn't any more air than was just eliminated so it shouldn't be enough to stop flow. if the system is gaining air, something else is going on.

SuperTech

I can't stress enough how much of a bad idea it is to replace coin vents on radiators, convectors and fan coils with automatic air vents. So many times I have seen the air vents either fail and cause flooding, property damage, ceilings falling apart, etc. Or they cause air to be sucked into the system when the boiler pressure isn't high enough. That sounds like it might be a possibility for the cause of your problems.

DIYorBust

https://forum.heatinghelp.com/discussion/comment/1847925#Comment_1847925

Yes I believe the problem is the makeup water line is full of air from when the domestic side is drained down, not just the normal small bubbles that come with the street water.

DIYorBust

https://forum.heatinghelp.com/discussion/comment/1847928#Comment_1847928

Imagine you have tenants on the top floor of your building. Now every time the radiators become air locked they complain the heat is out. They are unhappy. Now you have to request access, coordinate a time with them to go and bleed a coin vent. Particularly annoying if you need to do work on the heating system somewhere else in the building. But if you install a Caleffi minical with a hygroscopic safety cap to seal the valve if it leaks, you never have to go up there to bleed the radiators, the tenants remain warm and happy, and work can proceed on the system. If the pressure at the top floor is too low, the radiator can ingest air, but I highly doubt this is the source of my issue because I operated this way for years, and the airlock at the boiler location only began when we started doing work on the domestic side. Fortunately at the moment the building is vacant except for the unit I occupy, so I have access. Also have a 20 gallon amtrol thermal expansion tank on the way, I could also plug that in to a tee off a makeup water connection to the spirovent. Is that a good location for it?

mattmia2

https://forum.heatinghelp.com/discussion/comment/1847961#Comment_1847961

It still won't be more than what was vented out of the automatic vent.

mattmia2

https://forum.heatinghelp.com/discussion/comment/1847967#Comment_1847967

If that happens more than like once a decade you have some other problems to solve.

hot_rod

Tee a thermal expansion tank into the fill line to the boiler, just before the PRV fill valve. Reduce the thermal expansion pre-charge nto 20 psi, turn on the fill valve.

So the building pressure, call it 45 psi, partially fills the thermal expansion tank. Now you have a reservoir of water, maybe 3 gallons on the fill. A check upstream assures you don't lose any back into the plumbing.

We do this on solar thermal systems to handle cold temperature contraction. Viessmann solar calls it a safety seal"

And a BFD device on the fill valve to keep the boiler water from backing out.

This is waaay over-kill in my mind, but it seems you have exhausted other options that generally work. Like a check or BFD on the boiler.

DIYorBust

Ok thanks folks, this had been very helpful. I'm going to think about it and decide what to try when my new components arrive. It appears that I have the following problems:
1. A taller home than typical application for this consumer grade boiler requires me to keep my pressure above about 20psi, but not exceed the boiler specification of 30psi, which is a somewhat tight range given that the 20psi is based on physics, but the 30psi is based on the sensitivity of the T&P valve, which is conservatively set.

2. Ongoing work in the building requires makeup water to be periodically added to the system, and that water may contain a very large amount of air.

3. Air is not eliminated from the system prior to being introduced into the primary loop and triggering the high limit aquastat in the boiler, or seizing the primary pump.

It seems like adding the 20 gallon expansion tank, which will more than double my capacity, is a no brainer here, as is plumbing the makeup water directly into the air purger, a strategy recommended by some BFD manuals I looked at.

Other options include using a BFD pump, an expensive choice, but with some advantages such as eliminating the PRV and backflow preventer. My preliminary research suggests this may bring it's own set of problems due to the possibility of the pump adding air to the system during priming, hence the manufacturers recommendation to plug it directly into the purger. However I liked the idea of an air gapped system, but it's also one more electrical component to worry about.

I'm not sure I understand the 20psi expansion tank idea. It sounds interesting, but seems more geared toward thermal expansion in the solar system. Here, the problem which frustrates me is that once air is in the makeup water line, it never leaves until it gets bled out somewhere, and that somewhere is currently the main circulator pump.

I do understand the idea that air should not enter the system for a long time once it's up and running. But on the other hand, why do hydronic system designers include an air separator and a makeup water device at all? Money could be saved by having the installer bring along a boiler feed pump, and an air purger, and just filling up the system once and purging it. There is no makeup connection on a refrigerant line for example. I have to assume the engineers who design these systems include those devices because they see the potential for air to be introduced, and want it to be removed automatically, and not interfere with system operation.

hot_rod

Amtrol did a study years ago and came up with about 4% air in municipal water systems

Some air us eliminated when you fill and purge The entrained air comes out as the temperature is increased. If you don’t gave a good working air sep, that air goes back into solution when the boiler cools down

It just goes in and out if solution until you remove it. I think that is issue one, you are not getting 100% of the air removed

If you keep adding water the air sep should quickly remove it. Issue 2 is you do not have a good sep properly located?

If you tank is sized large enough, pre charged properly you could see only 5 psi pressure increase

You must have 40’ elevation to need a 20psi static fill?

DIYorBust

It is a four story building with the boiler in the cellar, so it's probably about 35 feet or so. I do have a spirovent, it's between the primary loop and the the area where my two zone valves tee off. I'm not totally sure what's going on, but it's no small amount of air I'm dealing with. It could be air getting stuck in the indirect, but I haven't had the issue in the past. Currently the makeup water goes directly into the supply side of the primary loop. The expansion tanks come after the zone valve tees(not in the zone loops, but past the tees). I'm planning to modify it to look more like this installation example:

I'm hoping with this setup, big slugs of air would mostly exit the air separator if they entered with makeup water.

hot_rod

Do you hear air in the primary loop, or in the boiler itself? . If you hear air in the primary loop, I'd disassemble that Spiro and check, it cannot be doing it's job if you keep hearing air going across it. It should remove aALL air within a hour or so of all zones circulating and temperature at designed SWT.

Some boilers, like FT, can trap air up high, I sometimes have to open the relief valve if it is on top of the boiler to remove a slug trapped up there. If it has an auto vent up top as the drawing shows, I would disassemble that also to make sure it is operating properly.

You have the two best air elimination locations covered. No reason to have air problems.

35' X .433= 15.1 psi to reach the top point in the system, add 5 for a safety, so your 20 psi is correct.

This is from the expansion tank location to the top of the system. Not necessarily the basement floor to the top, depending on where the tank is located?

Use the Wessel exp tank sizer if you have an idea of the total system volume?

https://www.westank.com/calculator/

50 gallon system volume?

55° fill, up to 140°

20 psi fill, 27 psi max.

I would go bigger that a #30 if these numbers are close?

DIYorBust

Thanks hot rod. I did replace the boiler top vent with a caleffi minical. It looked exactly like what came with the boiler. I could find nothing wrong with the valve I removed. I think it is a good idea to inspect the spirovent. I think you can the top off it without pulling the unit, like the caleffi version, but I'll have to try. They didn't make it easy for me to replace that unit. It's threaded into sweat adapters on both side, but the gap between the fittings is so small anywhere I can cut the pipe, that I'll have to do a bunch of unsweating and resweating, thanks installers. Should I just plan on replacing the unit? It's about 8 years old. With respect to the expansion tanks, I have a #30, and I added a #60, so I have both. I did set the charge to 22psi. I don't know how much water is in the system, but we have been installing steel column radiators and those hold a lot more water than baseboard convectors. My attempts to estimate the water volume have not been accurate enough to get a sense of the expansion tank capacity, but I hope adding a 20 gallon unit should rule out this problem and give me room to add some more radiators. The tanks are maybe 3 feet below the boiler. I'm not sure exactly where the air is going, but you're right, the spirovent is suspect. I will check it out.