System accumulates air......

Alan (California Radiant) Forbes

This is a new customer that called me who has a beautiful Berkeley Victorian. Three levels, 5,000 [], cast iron rads and a 10 year old Triangle Tube Prestige Solo 175. The bolts on the burner/HX had all sheared off and were sitting on top of the burner/HX, nuts on threaded bolts. The local dealer saw the pictures and sold them a new PA175. I'd like to make the new boiler last more than 10 years and told them to add a low loss header or buffer tank to keep the boiler happy.
The owner told me he has problems with upstairs radiators; has to bleed air every few days. The top floor is 30 feet above the boiler room and I'm thinking that a 15 psi fill pressure is not enough to keep the air out of the system; that we need to install a 50 psi relief valve and fill the system to 25 psi.
Any thoughts?

JohnNY

I find persistent air problems are often related to an undersized or ruptured expansion tank. When the system cools down, if it can't draw from the expanded diaphragm/air cushion, the negative pressures draws air in from the vents, pump seals, etc.

JohnNY

Mounting an expansion tank on its side like that greatly reduces its effectiveness.
Effectiveness. Is that a word?

leonz

I know I am a novice but:
Are all the air vent caps fully closed and tight?
If an autofill is used is it open and the boilers water filling line open?

If your bleeding air its an air problem not a balance problem following Dan's advice. Air is being drawn into the system somewhere.

I would make sure the vent caps are closed first. Then I would make sure the bladder expansion tank was sized correctly as it sounds like the new boiler was just that, a replacement without measuring the radiators and the size of the residence rooms and windows.

Gordy

If the air sep cap is loose could that circ differential be sucking air into the system through the air sep? I like the PONC location, but not the air sep so close to the suction side. Maybe I’m wrong hard to tell on phone screen.

Jamie Hall

15 psi will give you 33 feet from the point where that is measured to the point of 0 pressure. That may not be enough in that house.

Zman

Triangle Tube should have given them a new heat exchanger. That was a problem with a bunch of those boilers at the time.

You should be fine with a fill pressure of 18-20 psi which will work with a 30# relief valve.

Is the circulator pumping away from the PONPC?
I have seen issues like this where the relief valve occasionally pops due to a failed or undersized expansion tank and the owner does not notice. The system is then refilled with oxygenated water.

Rich_49

Gordy said:

If the air sep cap is loose could that circ differential be sucking air into the system through the air sep? I like the PONC location, but not the air sep so close to the suction side. Maybe I’m wrong hard to tell on phone screen.

What number circ is that Alan ? Looks like it's one of the bigger boys and Gordy may very well be onto something

Alan (California Radiant) Forbes

It's a Taco 0012 which has a flat curve. I was surprised when I first saw it, but I don't think it's over-pumped. It's pumping away, so it is adding its pressure differential to the system; all good for keeping the air out, but it's not.

JohnnyNY: "Mounting an expansion tank on its side like that greatly reduces its effectiveness." Please explain as that is blowing my mind.

Jamie Hall: I agree!

FranklinD

I think it’s because it’s harder to get the initial air bubble out of a tank mounted on its side...but I’m probably wrong about that one.

I was surprised by how small that expansion tank is, given cast iron rads & 5000 sq ft...that says “lots of water volume!” to me. My installer had used a tank that size on my system - which is only 425 edr - and it turned out to be too small by half.

Gordy

FranklinD said:

I think it’s because it’s harder to get the initial air bubble out of a tank mounted on its side...but I’m probably wrong about that one.



I was surprised by how small that expansion tank is, given cast iron rads & 5000 sq ft...that says “lots of water volume!” to me. My installer had used a tank that size on my system - which is only 425 edr - and it turned out to be too small by half.

The problematic air in the system could be enhancing the effectiveness of the xtank.

How much air is released every few days?

What’s the system temperature running at?

If this is a continuous problem I think it’s fresh make up water being added, and when it hits the lower system pressure at the top it comes out of solution. If the system uses hotter supply temps that enhances air coming out of solution also.

Could try higher system pressure with the 50#relief valve. Might work. Keeping the top under higher pressure might insure the air gets back to the scoop.

That’s all I got, unless there is problematic parts of the system wanting to hold air from initial purging.

JohnNY

Alan (California Radiant) Forbes said:

JohnnyNY: "Mounting an expansion tank on its side like that greatly reduces its effectiveness." Please explain as that is blowing my mind.

Look at it. Mounted on its side, you're creating an air-filled void from the connection thread to the top of the tank. That's a large part of the diaphragm that should be in direct contact with the system water.

Gordy

The big issue is corrosion to the tank. When air is trapped on the system side of the bladder the tank is more prone to corrosion, and failure.

Amtrol recommends a different style tank when non barrier radiant tubing is used. It has a plastic liner to help protect the tank walls.

Of course we could pull all kinds of thread up here on location, and orientation. Having it hooked directly to the bottom of the air sep upside down allows dirt to collect on the bladder possibly causing failure.

Rich_49

0012 is alot of circ for a system that would operate really well with a 30* Delta , especially one with low resistance .
Any idea what highest head circuit is and what the resistance is ? Is the DPBV set properly ?

http://www.taco-hvac.com/uploads/FileLibrary/curves009-0014.pdf

Would a 0015e be a better fit and get rid of the DPBV ?

http://www.taco-hvac.com/uploads/FileLibrary/102-541-0015e.pdf

Steamhead

Another thing to consider: Is this a converted gravity system with large pipes?

How many square feet of radiation does the system have?

Alan (California Radiant) Forbes

JohnNY said:

Alan (California Radiant) Forbes said:

JohnnyNY: "Mounting an expansion tank on its side like that greatly reduces its effectiveness." Please explain as that is blowing my mind.

Look at it. Mounted on its side, you're creating an air-filled void from the connection thread to the top of the tank. That's a large part of the diaphragm that should be in direct contact with the system water.

I always thought that they don't like horizontal because of support. Some installers let the x-tank hang by the threaded connection.

JohnNY

Yeah. I guess the wall brackets are best. I probably rely on the threads as a support device too often too.

EBEBRATT-Ed

30'=12.9psi +5 at the top =18 is what I would run. I was going to say make sure it's pumping away but others have mentioned that.

Wonder if the air pressure in the ex tank matches the system and prv fill pressure. If he has to bleed radiators you may not have an air cushion. I am betting the ex tank is the culprit one way or the other

Harvey Ramer

The tank is not to be mounted on its side because of uneven diaphragm extention and air trapping.

I am going to ask the obvious question. It the air separator working? I have seen them get clogged with flux at the top, preventing any air removal.

neilc

I would also question that looped pipe up to the x tank,
couldn't that be pigtailing ? unless that line is bled up to the iso valve just under the tank while the tank is disconnected,
seems you would want a vent up by the tank also.

HomerJSmith

You have 30' of static pressure on the PRV, which is about 13 psi, ignoring density etc. No anti-freeze. You add 18 psi system pressure and a 30# PRV, I don't think will handle it, of course what do I know. I would go to a 40 or 45 psi PRV.

You guys need at least 10 pipe diameters before water passes thru a micro bubbler and at least 5 pipe diameters on the discharge of a circulator. You may not be getting any air elimination, just a lot of turbulence. The pump may be forcing the float up. Total hd for the 0012 at 0 flow is 14' hd, would that be enough to push the float up? Heck if I know.

I like it when you can click on a jpg and it magnifys the photo, that's cool. Looking at your photos, it appears to me that you are pumping into the expansion tank rather than away from it. Hmmm, strange piping.

It appears the the 0012 is pumping into what seems to be the return to the boiler and a smaller pump is pumping away from the return. That's an indirect that I see?

Somehow you're sucking air into the system, I would think.

Gordy

Only way to find out is shut valve off and remove xtank

HomerJSmith

I think, I would have piped this as a primary/secondary system. Triangle Tube is a medium mass boiler and they probably don't require it, but, I would have done it anyway.

EBEBRATT-Ed

@HomerJSmith for 30' head you need 2.31'/psi so 30/2.31=12.9psi You only need 4-5psi on the top of the loop so 18 psi is fine.

Keep in mind expansion tanks are sized based on fill pressure, temperature rise, volume of water in the system, and pressure rise to some pressure below the relief valve setting.

Raising the pressure as you suggest reduces the expansion tank acceptance fator.

Raising the pressure temporally will aid in air removal.

HomerJSmith

If you are in fact pumping into the expansion tank you are lowering the pressure at the top flr ? and maybe negative pressure when the boiler is operating sucking air and then when the boiler is not running, you're releasing the air?

HomerJSmith

EBEBRATT, I guess I should rethink that. System pressure and tank air charge should be the same to maximize the acceptance volume designed into the tank, I do get that.

EBEBRATT-Ed

@HomerJSmith , that was my first thought, I have seen what you suggest happen, but he says it is pumping away.

@Alan (California Radiant) Forbes what size is the piping on that 0012? I would think with that boiler you would need to move about 15gpm so everything would need to be 1 1/4"

175,000 x .90%=157,500/10000=15.7 gpm. The oo12 is capable of more than that depending on the head. The boiler probably has a large pressure drop hence the 0012.

I think a low loss header or primary secondary to slow the velocity and remove air would fix this

Alan (California Radiant) Forbes

I'm going back tomorrow and will check the fill pressure of that x-tank. Will also check the tnk volume to make sure it's large enough for the system.

Definitely, pumping away. The pump is located after the air eliminator.

HomerJSmith

It's hard to tell from the picture the direction, I was looking at the hump on the volute as to direction. If it is pumping away that would be consistent with the smaller pump. OK, I looked at it with my magnifying glasses and I could see the pixels on the screen and the pump shadows says it is pumping away from the X tank.

Back to square one.

Since the PRV isn't dripping, EBEBRATT's calculation of system pressure at 18 psi is correct and a 30# PRV is ok. So, where is the air coming from? The 0012 is installed correctly pumping away from the X tank and more importantly the air eliminator, although I like 12 pipe diameters of straight pipe on the input to any pump. Also...

I can't really see your header very well, they look small and I am wondering how you are controlling and balancing your zones.

As we all know air is buoyant in water, so the air could be coming from any level. My thought is that the air is being sucked in thru the air eliminator. Is it possible that the negative pressure at the eye of the impeller is so high that it is sucking the float in the air eliminator down and pulling air into the system which rises to the top floor? Screw down the cap on the air eliminator and see if you still have a build up of air on the 3rd floor.

If that does solve the issue, consider your pressure losses and your choice of a pump.

HomerJSmith

If the boiler has a high pressure loss thru the heat exchanger, then the 0012 is not the pump to use. It only has 14' hd at 0 gal of flow. I don't know what the minimum flow rate thru the HX on that boiler is, but if it is 175M btu's, I would suspect about 17 gal at max firing rate?
At a flow of 17 gal, the hd would be 12'. Is that enough? I hope, I got that right. I'm just throwing ideas around.

Alan (California Radiant) Forbes

Homer: Thanks for your thoughts! I'll try closing down the air eliminator next time I go out.

The pressure drop through the boiler is minimal; 2' @ 15 GPM.

Steamhead: This is not a gravity conversion. The previous owner installed the system within the last 10 years.

Sorry, I don't have the EDR number; probably somewhere close to 1,000.

neilc

I'm sure you're on this, but,
while the tank is off for checking and setting pressure,
crack that isolation valve and prime that line up to the tank

Alan (California Radiant) Forbes

Probably another good reason to use a potable expansion tank.

HomerJSmith

Alan, here's another idea for you. I already stated where the air is getting into the system, but you should be aware of this:

Your Discal air separator may not even remove air from the system. You are using a 3/4" Discal which may be too small for the flow thru it. So, the micro bubbles aren't able to
coalesce and float to the vent as they just get pushed by the flow out to the pump. I'm not going to get into how temperature and pressure affects air removal, but the Discal should be sized on the flow thru it and not on pipe size. There IS a maximum flow limit for it to work effectively.

I would have plumbed the Discal differently. I would have uses
1" or more likely 1 1/4" piping from the boiler, regardless of the boiler's output pipe size, to a 1 1/4" Discal. That would have slowed the flow and would have allowed the Discal to do a better job of removing micro bubbles. That would have been better for the pump, too, as the piping inlet to the pump is so short.

I'm an advocate of generous headers, larger pipe sizes, smaller pumps= lower energy costs= greater efficiencies.

Alan (California Radiant) Forbes

The air eliminator and piping is all 1", but I agree with you. It should have been 1¼".

hot_rod

HomerJSmith said:

Alan, here's another idea for you. I already stated where the air is getting into the system, but you should be aware of this:

Your Discal air separator may not even remove air from the system. You are using a 3/4" Discal which may be too small for the flow thru it. So, the micro bubbles aren't able to
coalesce and float to the vent as they just get pushed by the flow out to the pump. I'm not going to get into how temperature and pressure affects air removal, but the Discal should be sized on the flow thru it and not on pipe size. There IS a maximum flow limit for it to work effectively.

I would have plumbed the Discal differently. I would have uses
1" or more likely 1 1/4" piping from the boiler, regardless of the boiler's output pipe size, to a 1 1/4" Discal. That would have slowed the flow and would have allowed the Discal to do a better job of removing micro bubbles. That would have been better for the pump, too, as the piping inlet to the pump is so short.

I'm an advocate of generous headers, larger pipe sizes, smaller pumps= lower energy costs= greater efficiencies.

I would suggest sizing piping and components as accurately as possible. For piping the generally accepted number for hydronic pipe sizing is 2- 4 fps velocity, plenty of online calculators to show you those sizes.

Some suggest 5 fps is acceptable for continuous flow hydronics, it is on the edge of noise especially if you flow through valves or reduced port restrictions.

At least one pex manufacturer is promoting 8 fps, that is excessive in my opinion, an effort to meet copper tube size flow rates possibly?

Any valve manufacturers should have a Cv number assigned. Reputable companies will run actual flow tests to establish that number. i'm not sure all imports are actually flow tested

Balance and check valves really need to be sized with Cv not pipe size to perform properly.

if you oversize a check valve for example the flow rate may not be adequate to push the flapper or disc completely open. noise and cavitation are the result.

So bigger is not always better.

HomerJSmith

Right on, "hot rod". Your perspicacity is unmatch. What would be the maximum flow rate thru a 1" Discal before coalescence radically slows?

Alan, it's hard for me to judge the pipe size from looking at a photo. Photo's with magnification helps. Glad to hear it's 1".

HomerJSmith

I have heard that Europe is going to larger piping and smaller pumps to lower energy costs because energy are so expensive.

hot_rod

HomerJSmith said:

I have heard that Europe is going to larger piping and smaller pumps to lower energy costs because energy are so expensive.

??, maybe
I know when you visit old buildings, hundreds of years old, that have had hydronic heating systems added you see a lot of 10mm copper and pex home run piping systems to panel rads.

Remember too ECM technology has pretty much cut pumping power in 1/2 over the last 20 years. The EuroPump standards make PSC circulators obsolete over there, and they keep raising the bar on pump efficiency requirements.

How many jobs do you see now a days that can't be pumped with 37W or less with and ECM circulator. You should be able to move 10- 12 gpm with under 37W these days even in 1/2 properly designed pex loops.

Wilo had some I think 5- 10W mini circs on display years back that mount right on the panel rads.

Gordy

So long as the system is low headloss. Get in the 11.5' range with 15 gpm not much on the lower cost ECM circs to fit that slot.