How is air getting in my system?

ptj

I have a small 1924 home that was originally a gravity system with a cast iron radiators. I recently finished part of my basement and added an aluminum radiator. Along with that, I replaced some original pipe that was in the way with pex and replaced one original radiator with a vertical aluminum radiator. This new radiator is at the highpoint in the system and is designed such that a small amount of air cuts off circulation to a significant portion of fins. I'm having to bleed the vertical radiator every day and it seems to be gradually getting worse.

My suspicion is that this is corrosion putting off gas. Old cast iron and left empty and damp for a while I worked on the pipes possible made things worse. I've added inhibitor. I have already added a couple extra doses, but it didn't seem to make any difference. I recently got the test strips from hercules and it showed me a little low in reserve alkalinity. Is there any harm in doing too much inhibitor?

Another theory is the taco circulator is forcing bubbles in somehow. It is circulating fine now, but I did accidentally turn in on for an hour or so when then system was empty. Possibly ruined some seal?

Is there anything else I'm missing and should be looking for?

some notes about the system

- Weil-Mclain boiler about 15-20 years old
- New PEX is oxygen barrier
- Steel drum expansion tank
- Single zone
- Maintaining water pressure for the most part. There is no automatic water feed

The next step is probably add a spirovent if there's not a clear answer

Ironman

How about posting some pics of the boiler and the work that you’ve done?

You definitely don’t want to put a SpiroVent on the system if it has a compression tank. You should have some means of moving air into the tank to maintain its air cushion.

Jamie Hall

First, corrosion is very unlikely to produce enough gas to cause a problem.

Second, if you have a compression tank and not a bladder style tank, absolutely the LAST thing you want to add is any form of Spirovent or other air eliminator. That type of system is specifically designed -- usually with a fitting called an airtrol -- so that any possible air gets directed to the compression tank. Adding a Spirovent or some such gadget will bring on a host of issues which you don't have... now. And don't want.

The bottom line, though, is that air can only get into the system if there is somewhere in the system which is below atmospheric pressure, and there is a way for it to get in at that location. Where that might be depends, in turn, on how the system is piped (particularly where your compression tank is connected in relation to the circulating pump), what pressure you are maintaining, and how high above that pressure point the highest point in the system is.

HVACNUT

No auto feed (pressure reducing valve)?
Where is the "highpoint" radiator, third floor?
What pressure is on the boiler?
Is it a mono flo system? A perimeter loop around the basement with branch loops to each radiator?
Did you change the piping configuration or remove any Tee's?

EBEBRATT-Ed

@ptj "Maintaining water pressure for the most part."

What does this mean?

ptj

Here's the rough schematic of the whole system.

Really appreciate the feedback so far. Good to know that is probably not corrosion and to expect a little air with the old style compression tank.

To answer questions:

The pipe to the expansion tank comes out of the top of the boiler. The boiler is weil-mclain gold cga series 1

No autofeed. Just a manual valve and a backflow preventer.

The boiler is in the basement with two floors above it. I would estimate that it's about 20ft vertical between the highpoint and the boiler.

Right now at the boiler I get about 16 psi cold and 18 hot. by "maintaining pressure for the most part" I mean the pressure drop seems proportionate to the air I let out, maybe 1 psi over the course of a couple winter months.

Not a mono flo system. Lots of Tees. I added the basement radiator and the pipes to it, but the configuration for the other radiators stayed the same. It's a small house and fairly balanced in terms of distribution, so I keep all valves open all the time.

EdTheHeaterMan

I am guessing that your diagram is a basement pipe layout since the radiators for the second floor are at the bottom, the middle, and the top of the diagram and at the same plane as the first floor radiators in some cases. So based on that. I doctored your drawing to make it clear that the second floor rads are actually higher than the first floor and basement radiators. The lines on the angle are actually vertical pipes.


Can you pinpoint what radiator is getting all the air?

To understand how air from the closed steel expansion tank can get into the radiators you may want to look at these recent posts
https://forum.heatinghelp.com/discussion/comment/1767019#Comment_1767019

https://forum.heatinghelp.com/discussion/comment/1767231#Comment_1767231

https://forum.heatinghelp.com/discussion/comment/1767235#Comment_1767235
After reading this info, If you have more questions then please ask them here, or PM me.
Hope this helps.

Mr. Ed

Teemok

Do Ed's reading first, it likely covers my ideas.
What I see is maybe a system with unbalanced distribution flow. An old gravity system likely has good sized pipe. With radiator valves wide open the paths of least resistance might be getting the bulk of flow and your new more restrictive radiator gets much less and collects air because it lacks the proper flow rate to move air out of it. The size of the circ. and piping might help in guessing what going on. Balancing flow by restricting the easiest paths (closest radiators) at their valves might help. Careful you can reduce flow too much. It could be you can't get the needed flow rate with your combination of circ. and all the flow paths. Doing some logic checking math for flow might tell you what's possible to get at your new radiator. 2-5 GPM is the sweet spot flow rate. I'm used to bladder style expansion tanks and efficient air separators. When it's smart to leave the old working systems alone and when upgrading is best is not in my everyday tool kit. I'll let those more experienced with compression tanks comment on that.

SuperTech

I understand that with a compression tank you don't want any automatic air vents.  But would a Spirovent piped to the compression tank  help improve things? I work on a few boilers with old fashioned camel hump air scoops piped to direct the air to the compression tank. 

Teemok

@SuperTech On the supply piping of the boiler maybe. Sponge effect would keep air moving to the tank. Air collection at the top of the new vertical radiator is the problem. Maybe the sponge effect would counter the tendency to collect at the high point but I think proper flow is the first thing to check/confirm/achieve.

Sounds like the system worked well as designed before the radiator add.

neilc

that diagram,
you're pumping towards the boiler and the tank, PONPC,

I had a noisy system, air noise,
NOT pumping away, and one far random rad would always catch air,

I had the circ moved to pump away from the PONPC,
from the moment of that first restart, the system was silent, and I haven't touched that rad sinse.

If you pump to the tank, you'll push the air out of it,
PUMP AWAY.

hot_rod

Pumping away from the expansion vessel, whatever type it is, makes air removal much easier.
Large diameter pipes= low flow velocity. Below 2 feet per second it is hard to keep air moving along with the water back to the purger.

Yes you could add a good air separator and pipe the discharge up to the compression tank. I’ve done this with drainback solar many times.

Caleffi air seps have an optional fitting, as do most others, to change that metric cap thread to NPT, then copper tube to the steel tank..
With the low flow, you may still need to bleed each radiator occasionally. They become high point, low velocity air seps.

This installer runs a 3/8 copper off the high point air vents on the boiler down to the floor. So when they spit or leak they don’t damage the electronics

Teemok

There are so many non pumping away systems that work. Not as well as they could but just like they have for decades. Being the guy that shows up and says, "this all has to change around" even though It might be correct, is often not well received. That said, @neilc is correct moving the pump would better given that the existing pump is sized correctly for all the radiators and the by-pass. Try flow management adjustments first, if that doesn't work. Flow pump assessment next. If it's good, canning the compression tank for a bladder type tank mounted behind/upstream of the existing pump location and adding a good air separator on the supply would my minimum fix effort. If that pump is a tired old B&G100 replace it and re-pipe with new peripherals pumping away.

ptj

Really interesting discussion.

@EdTheHeaterMan That's a good improvement to the diagram. It's A2 that is catching the air. It's a 70" vertical radiator so the top is the lowest pressure in the system. The other two top floor radiators are 36" and can have a little bit of air, but they're thick cast iron fins and the circulation doesn't get cut off with a little air like the vertical rad.

I did get air prior to the conversion but it didn't seem like as much, or maybe just didn't notice because it didn't affect the room temp as much.

I'll play with flow rates to see if that makes any difference. I'm mostly just relieved it's not a corrosion issue causing permanent damage

Adding an air separator piped to the expansion tank is an interesting idea. I'll try and find some examples of that.

That other thread mentions the airtrol tank fitting. Is that an option here? It seems like I would need to get an airtrol boiler fitting (doesn't seem to be for sale any more) or similar to collect air

EdTheHeaterMan

Many Weil McLain boilers have that "Airtrol-style" boiler fitting built into the design. Read this: https://forum.heatinghelp.com/discussion/comment/1767315#Comment_1767315 What Weil McLain Boiler do you have?

EdTheHeaterMan

Then if you look at the piping in the basement with an accurate level, you will find that most of the air bubbles created by the boiler (because hotter water can not hold as much dissolved air in suspension) will find their way to the top of what ever vessel they are contained in. as you look at the pipes you will find that most of the air bubbles are following this path.
ending up in the problem radiator. If it is easy to place the circulator on the supply pipe of the boiler, and pipe the expansion tank on the proper tapping on the boiler, then your problems will vanish.

Teemok

Hot_rod has a good point that even if you could get the air out of the high point of the radiator with good flow and get it into the return piping, it might just get stuck there due to the low velocities of the larger piping and the air wouldn't make it back to the boiler to refill the tank. When the flow stops most of the air rises back to the high point.
This is a good argument for a well placed air separator and a bladder tank. This new set up would create the conditions for air elimination vs/instead of an air circulation/ tank management skeem. The hot water leaving the boiler can't hold as much gas when it's hot so bubbles form coming out of solution and the air sep. extracts and vents it. That just vented gas had just been absorbed into solution at the cooler high points of the system. Soon there is very little air in the system at all.
The pumps flow rates with the old radiators dynamics might have been just right to keep air moving around but the new radiators have higher head and higher air traps resulting in lower flows and the air collection condition.

mattmia2

Is there no airtrol fitting? Is the tank getting more water in it? I assume the air is migrating out of the tank and ending up in the radiator.

I think this is the replacement for the airtrol boiler fitting. The instructions sort of explain the system.
https://www.supplyhouse.com/Bell-Gossett-112119-1-1-4-Inline-Air-Separator-8601000-p

@DanHolohan explains the airtrol system here:
https://www.youtube.com/watch?v=y15SVvwn4sI

neilc

I thought the airtrol fitting was the piece that makes ceiling tanks work, eliminating the air exchange up there,
tanks work when fitted, and systems piped, properly

neilc

https://www.xylem.com/en-us/products--services/heating-ventilation-air-conditioning-hvac-plumbing/expansion-tanks/tank-fittings/airtrol-tank-fittings/

leonz

Pictures of your system would help us/me a lot.

By what you have described your newer boiler has an air baffle in the steam chest to capture air bubbles and divert them to the steel tank.

If you do not have an airtrol valve in the base of the steel tank it will be difficult for cooler water in the tank to fall back into the boiler. The airtrol valve is cast with 2 passages so that it has one passage for hot water to rise into the boiler and the other passage allows cooler water in the tank to sink and pass the hot water rising into the steel tank and drop back into the water flow.

I have a B+G Internal Air Separator to pull micro air bubbles out of my system and a B+G ATF-12 airtrol valve in the base of my 15 gallon steel compression tank to strip the air bubbles out of my system and it works very well with my 225 foot single loop of baseboard(which I hate).

Ideally to correctly make use of the point of no pressure change you need to be pumping away from the boiler and pushing the hot water through an Internal Air Separator to your steel compression tank with a 3/4" line coming from the Internal Air Separator to the ATF-12 airtrol that is in the base of your steel compression tank.

Adding automatic air vents of any type will cause your system to waterlog and you will lose the point of no pressure change.

My system was throwing fits and had gone into a vacuum condition until I found that the packing nuts for the water gauge sight tube were not tight enough and tightening them solved this problem and it has been fine ever since.

hot_rod

Sometimes the confusion is that the air never got removed 100% from day one. Not necessarily air that is getting in constantly.

Air comes out of solution as water heats, at that condition it needs to be removed. If not as the system cools that air goes back into solution. So you hear the air noise again on start up, when the fluid is cool.

The combination of a properly placed, working microbubble device and critical high point vents should make and keep mall system noise-free and efficient.

The big problem becomes fixing complicated legacy systems. How much needs to be repiped or modified to get the best air removal working for you.

Cavitation and flashing can sometimes be confused with air noise also. Different cause, different solution needed.

Then there is non barrier systems they do in fact present ongoing problems. It's not a myth.

hot_rod

mattmia2 said:

Is there no airtrol fitting? Is the tank getting more water in it? I assume the air is migrating out of the tank and ending up in the radiator.

I think this is the replacement for the airtrol boiler fitting. The instructions sort of explain the system.
https://www.supplyhouse.com/Bell-Gossett-112119-1-1-4-Inline-Air-Separator-8601000-p

@DanHolohan explains the airtrol system here:
https://www.youtube.com/watch?v=y15SVvwn4sI

That is just another way of doing central air scrubbing, using centrifugal force instead of a media. One concern with that device is variable speed pumping. Does the flow drop to a point where you no longer "spin" the air out?
I built one of those out of clear plastic pipe once to see what happens at various flow rates. Properly applied they work fine.

mattmia2

leonz said:

If you do not have an airtrol valve in the base of the steel tank it will be difficult for cooler water in the tank to fall back into the boiler. The airtrol valve is cast with 2 passages so that it has one passage for hot water to rise into the boiler and the other passage allows cooler water in the tank to sink and pass the hot water rising into the steel tank and drop back into the water flow.

The airtrol fittings are designed to prevent this. They mostly stop gravity circulation to the tank. The gravity circulation causes much more water to circulate through the tank than will circulate as water enters and leaves through expansion and contraction of the water in the system. The gravity circulation brings dissolved air with it which then causes the air charge in the tank to migrate to other parts of the system.

jesmed1

mattmia2 said:

The airtrol fittings are designed to prevent this. They mostly stop gravity circulation to the tank. The gravity circulation causes much more water to circulate through the tank than will circulate as water enters and leaves through expansion and contraction of the water in the system. The gravity circulation brings dissolved air with it which then causes the air charge in the tank to migrate to other parts of the system.

Thank you for that explanation, as I was unclear how the Airtrol was supposed to work.

The Airtrol tank fitting instructions also give detailed instructions about how to eliminate most/all of the air at startup, including first running the circulator with the boiler off to direct air bubbles back into the expansion tank, then heating the boiler up to 220 degrees with the circulator off, then running the circulator to let the air bubbles separate out at the system high points, then bleeding/venting the high points to get that air out.

None of which I've ever done, because I don't have an Airtrol and didn't know to run the boiler temp up high to get the air out. But I'm hoping to install one and will try this procedure.

EdTheHeaterMan

OK Ok this is a point that no one has covered. When the steel closed system tank is not properly piped, the circulator on the return as so many are and other poor designing, the air in the expansion tank will get absorbed into the colder water (that is Boyle's Law dan referred to in the video) then that cold water finds its way into the system and releases the air. that air that is no longer in the tank will end up in a radiator.

When you pump away from the compression tank and the circulator is on the HOT supply pipe, that is where the air will separate from the water. If you place the air scoop/vent/microbubble catcher thingie at that location and pipe that air into the expansion tank then the amount of air in the tank will increase over time.

Here is the part that no one is thinking of... The air in that radiator will get absorbed into the water. (Steel tanks are not the exclusive air losers in a system). Since the radiator water is cooler and at a relatively lower pressure, the radiator air now gets absorbed into the water. Eventually you will get a water logged radiator. By the way... you want a waterlogged radiator... they are more efficient than a radiator with some air in it. So by pumping away from the tank and having that tank on the hottest pipe in the system, you have created an air removal system that will charge the steel tank with air, not the other way around.

jesmed1

EdTheHeaterMan said:

OK Ok this is a point that no one has covered. When the steel closed system tank is not properly piped, the circulator on the return as so many are and other poor designing, the air in the expansion tank will get absorbed into the colder water (that is Boyle's Law dan referred to in the video) then that cold water finds its way into the system and releases the air. that air that is no longer in the tank will end up in a radiator.

I think I understood most of what you said, but you lost me here. Isn't it true that the expansion tank pressure is unaffected by the circulator? So the pressure in the tank is the same whether the circulator is pumping away or pumping towards the tank, because the tank is the point of no pressure change.

And if the pressure in the tank is the same regardless of where the circulator is, how does pumping towards the tank make more water get absorbed?

Maybe you are saying that pumping towards the tank doesn't necessarily cause more air to be dissolved, but it does cause more air to be released at the high points due to lower pressure?

I'm sure you are correct, as always (!) so am not doubting, just trying to understand.

Teemok

His system uses the boiler itself as the air pickup. The tank is not on any pipe. Apparently the section casting is made for it. There may or may not be a tank airtrol. The pump is without a doubt on the wrong side. My suspicion is that the rate that air is either absorbed at or carried away in flow out of the high point radiator is slower than the air leaving the compression tank gathering in the radiator. Why? Pump position, likely. No tank Airtrol? If missing defiantly. Low pipe and radiator velocities unable to move the air back to the tank? maybe. Is there a 100% fix available? Yup. What temp. is this running at? If it's low-ish it will be even harder to get air out old tech style.

Teemok

Air will get absorbed in cool parts of flow IF there is effective venting at hot point (need bladder tank) or the air captured by the existing tank boiler combo can keep the air in the tank. Otherwise it's just air moving around collecting at the wrong place. I'm guessing the compression tank would get water logged if the radiator was just vented and new make up water is added to replace the vented air.

mattmia2

as you listed in your previous post and others have said there are ways to encourage the point where the air comes out of solution to be at the compression tank connection. that will encourage the air to dissolve in the cool water in the radiator and get move back to the compression tank. it is slightly more difficult to do it with a compression tank than a bladder tank but it can be done without retrofitting to a bladder type tank.

Teemok

I used the word OR in that post. Very hard to do it (Keep air in the compression tank) with no tank airtrol and the pump pushing at the tank. My thought was the new vertical radiators present a system change that might be best dealt with by switching to a vent the air out scheme instead of shepherding it to the tank. By the time you move the pump, that may or may not be sized right and add an Airtrol to an old tank you could do it using current tech. No doubt the old set ups do work when right. Some of the old set ups done wrong worked well enough that they are sill operating.

Steamfighter49

Seems well covered. Pump away. Bladder tank. Airtrol. 

 One other thing - NEVER mix radiator types or material ie alum vs cast iron. They heat and cool at different rates and times. 

mvickers

To simplify all the complicated answers; Are you sure that you don't have a very small leak, evaporates before seeing it, which when the pump kicks on & off, it's pulling in air?

joebinter

I'm currently working on a similar problem, except in my case I removed one small radiator, rather than adding one, and another small and kooky radiator Tee'd off it stopped working.
My system has no pump or circulator, pure gravity. One thing I noticed in my system is that one twenty-five-foot baseboard radiator is very hard to bleed completely because it's level, rather than being angled with the return a little higher than the feed. I have to bleed it with the valve barely open, and it sputters an air-water mix for minutes before it finally delivers a stream of pure water. I'm thinking maybe you ran some pex so it's perfectly level or even angled a little retrograde, creating a small air reservoir. If that's it, it should eventually work the air out as EdTheHeaterMan explained above.
Thanks to everybody posting to this discussion. I have learned much by reading the posts.