System keeps getting air bound

wchevron

We've been having repeated no heat issues at a 60 unit apartment building. There are 2-750,000 btu condensing boilers installed with primary/secondary piping. There are 4 zones. Each zone has it's own circulator. The circulators are installed on the return piping. The expansion tank & main air separator are located on the supply side of the circulators.
Talked to Taco about it and the ex tank should be on the return side of the circulators but since are 4 individual zones, one tank wouldn't handle all four zones. They recommended closing any automatic air vents and boosting pressure in the system. Closed all the air vents in the mech room and any we found at the end of the mains. Can't get into most of the apartments to check the baseboard. Changed the relief valves from 30 to 40 lb and boosted the pressure up to 35psi. Just got a call about 6 units were air bound again yesterday. At a loss of what to do next. Everytime the system gets airbound, it always seems to be the same zone. Can't figure out where the air is coming from. Any thoughts.

Zman

What model boilers?
What pressure are the boilers rated to?
How tall is the building?
Which Taco Circs?
Do you have pictures?

You almost certainly need to do some repiping.
If you read the book "Pumping Away" available on this website you will understand what is going on.

Carl

wchevron

2-Weil McLain 750,000 ultra boilers. Boilers are rated to 70psi
Building is 3 stories tall, in an L shape configuration.
The longer of the L shape has 2 zones with a Taco 1610 on each zone.
Shorter side has 2 zones with Taco 0014 circ on each zone
Don't have any pictures.
I'm assuming putting the circ's on the supply side would fix the issue. Lots of work involved doing that though. Was hoping increasing the pressure would help.
My thought was to go there tomorrow and open all the taco zone valves (1 per apartment) manually and start checking the return pressures on each zone to see what we are dealing with first.
Then was thinking of isolating the 2 boilers and cranking up the pressure in the system until I'm getting positive pressure at the farthest point, or where ever I find my lowest pressure. My thought is this would be the pressure the system needs to run at to stop the air infiltration. Then change the boiler reliefs.

Tinman

Any chance you're losing water somewhere via leak? If so, you'll be continuously introducing fresh water with fresh amounts of air/oxygen.

GW

Air can't go where there is pressure, can you possibly check pressure at one of those air bound units? Take an air vent out and screw on a temp hose and guage? What floor are you talking?

Jamie Hall

GW is right, no surprise there. Keep in mind that with the setup you have, although you may have 35 psi at the boiler, you have less pressure at the inlet to the pumps -- and even less pressure at the end of the radiation, since it is the pump inlet pressure minus the elevation difference. It is quite possible that on the trouble zone you have low enough pressure when the pump is running -- despite having enough static pressure -- to suck in air.

SWEI

Any chance you might be able to simply unbolt and reverse the circulators?

Zman

Are the circulators in the boiler room?
Is the boiler room at the lowest point?
Neither of those circs could suck air at 35 psi. If they are higher in the building, perhaps...
My guess is that once you get the air out, it will run OK until the next guy drains it.
A repipe would be the right thing to do...

Carl

Gordy

Pump away from the PONPC and your troubles will be gone. This is where the difference is in pumping away where in a residential smaller application it may not make a difference as much as a larger application like this.

Also air seperaters do a better job of removing air on the supply side of the boiler where the water is the hottest, and air more readily comes out of solution to be captured by the air seperator.

Snowmelt

Who installed it, sounds like a piping problem. Does the apartments with the air problem, do they have an air bleeder, I would experiment by putting a bleeder in each room. That's a temp fix that will work but ultimitly you want to pipe it out correctly.

neilc

Gordy said:

Pump away from the PONPC and your troubles will be gone. This is where the difference is in pumping away where in a residential smaller application it may not make a difference as much as a larger application like this.

I recently had a circulator relocated from pumping to,
to pumping away from the tank.
Made a world of difference to a system that would be noisey, need bleeding several times a season, would water log the tank, etc.
The rad that collected the air doesn't even think to trickle any more, is solid heat all day long, and in my bosses office.

Pump away from the tank or PONPC ! ! !
and read the book.

icesailor

35# PSI pressure at the gauge on the boiler will support a column of water 80' high. Is the highest emitter 80' above the boiler?

If you allow 12' per floor, that's still only 36' in altitude or 15# PSI Gauge. Then, allow for the first floor to the cellar. 1600 Taco's pump a lot of water, but not at a high pressure. Once the system is filled completely with water, it doesn't take a big hydraulic mining pump to push it around.

It was running before the boiler change, I assume? So, it worked until the boilers were changed?

Are there any original "as built" drawings? Drawings on how it was supposed to be done?

The problem is how the boilers are connected to the system. That building is too big to have not been designed by some Mechanical Engineer. If it worked before, and the only thing that has changed is the boilers, its the way the boilers are connected to the old system. The old high mass boilers had different flow needs. Weil-McLain Ultra boilers have different flow needs.

Without looking at it, I'll bet that if the two ultra's were connected together so that they did their thing all by themselves, and were connected into the worlds largest hydraulic separator, and the old system was connected to the other side, it would be working just fine.

Those zone valves. They are usually put on a bypass loop so as not to interfere with the pumps flowing. A big commercial job like that usually has circuit setters to balance the system. What are the pressure differentials? Especially at the 1600's.

hot_rod

If a repipe is an option, consider a hydrosep. It would clear up any pump conflicts and provide excellent air separation. You really need to have a working air separator, regardless how you accomplish it.

wchevron

If I can figure out if just the one zone is the causing the air problems, would re-locating the circulator for that one zone to the supply side work if I left the other circulators where they currently are?

Would it be possible to install 4 ex tanks, 1 per zone, in the system before the circulators so I'd be pumping away on each zone. Never seen multiple ex tanks in a system before. Would be much easier to add tanks rather than relocate all circulators.

Gordy

You cant just relocate the X tank you have?

Its hard to say if one on each zone would work with out seeing the piping. It should, provided you get rid of the PONPC you have on the supply side of the circ. Seems like a lot more work, and material than relocating one x tank.

wchevron

The returns from all four zones come into the mech. room at ceiling height, elbow down towards the floor, the circulators are in the drops. On the discharge side of the circulators the four of them tee into a 4" return header which then picks up the primary piping from the boiler then continues and tees back off into 4 zones on the supply side. I would have to pick only 1 zone if I put the ex tank on the intake side of the circs.

Gordy

You cant put the xtank on the primary loop?

wchevron

I could. Would that make a difference?

hot_rod

can you just sketch a drawing, snap a pic with a phone, and post it. It might be easier to just see how it is all piped, and come up with an expansion tank location.

Do you have something like this? It's easy to tie into a primary loop for the expansion tank. Usually the pumps pump away from the P/S headers?

Gordy

As long as all pumps are pumping away from the ponpc ( x tank) then you are all right. Like HR says usually that is the case with p/s piping.

icesailor

Again, if it worked before, what changed to make it stop working and why?

Of there was a hydraulic separator in place of those closely spaced tees, it should be working like the old dead guy that figured out how P/S resolves differential systems.

I always found it more helpful to understand how something worked before and went from there. Not how it should work now but doesn't, to come up with an answer.

What do you say when doing something major, doesn't solve the problem? Try something else?

wchevron

Here is a sketch of what is there now

Gordy

is there room to add an x tank at each circulator inlet. The key to pumping away is having the PONPC as close to the circs inlet as possible so the full differential pressure is added to the system. moving the circs to the supply header gets you there with out moving the X tank.

spinning pumps 180 makes you have to change S/R piping at the primary loop and move those pumps to the supply, and the indirect piping flip flopped.

GW

you may wish to confirm your pressure with a new gauge. I have seen this many times where the gg stops working properly. Ice man nailed it, if it worked once it should work again, unless someone did something to impede operation

edit- also you may want to try getting a pressure reading on the top floors and or where you are having problems. You said you can't get into some units? Are they doing something illegal or what?? They don't own the place do they?

icesailor

Is it piped like page 21, 22, and the rest of this I/O manual for that boiler?

They're pretty specific on what you do to connect them to existing systems.

http://www.weil-mclain.com/en/assets/pdf/550100086_0814Web-manual.pdf

Hydraulic separation in a different box.

Flow through the two boilers is like the three baby bears and the porridge. Not too fast, Not too slow,, but, Just Right.

Gordy

They also want you to pump into the boiler.

hot_rod

Too many mistakes in that piping. If you want to attach your name to it maybe suggest it is re-piped as per the manufacturer and best practices.

Here is an example of what pump toward the exp tank will result in.

icesailor

hot rod said:

Too many mistakes in that piping. If you want to attach your name to it maybe suggest it is re-piped as per the manufacturer and best practices.

Here is an example of what pump toward the exp tank will result in.

Weil-McLain has very specific instructiions on twinning one or more of those 775,000 BTU boilers. They even provide manifolds and instructions too do it.

It appears to this inexperienced eye that the old system was probably piped Primary/Secondary with the old high mass boilers. A large Hydraulic separator with one side connected into a manufacturers provided fittings, with the proper pumps and flows through the boilers, and the other side connected to the system would make the system rock. A reading of the FIM states from Weil-McLain that water flow through the boiler(s) is absolutely critical. Like the three little bears. Not too fast (corrosion), not too slow (overheating) but JUST RIGHT.

Isn't that how Gil Carlson (or whoever it was) how to make the two buildings at Notre Dame University work? The basis of Primary/Secondary where one source was much higher or lower than the other? Unless they were perfectly matched, they wouldn't pump. So, if you mix it through closely spaced tees, it will equalize and mix?

What size pipe do you need to handle 1,550,000 BTU's?

Gordy

4" at a reasonable velocity

Paul48

I don't see P/S. You're attempting to remove air, in the return of a 3 story apartment building?

wchevron

I think we got it fixed. The water fill was only set at 12 lbs. Adjusted it to 25 lbs. Isolated the boilers and each zone. Cranked up the pressure and spent a day making sure each zone had no air left in it. Also manually opened all the zone valves for each zone and noted pressures. Then closed all zones but one and noted pressures.
After adjusting the water make up, put pressure gauges on the end of the main in the basement, on the problem zone. Had 22lbs supply & 20 return. Went up to the first floor hallway, which handles 2&3rd floors, had 20 lb supply and 19 return.
Been back the last couple of mornings and the pressure have been the same and no air bound apts. Going to keep tabs on it next week but it looks like it may be solved.
Thanks everyone for your help.

Funny coincidence but I was at another commercial building yesterday. Wants to replace their large old cast iron boiler with a Viessmann Vitodens. Noticed the 5 zone pumps were on the return, same as this building. I've already figured to move the circulators to the supply side to avoid the same issue cropping up at this place. I wonder with the large mass old boilers the pump/ex tank location wasn't as important?

icesailor

The pressure needs to be the first thing you look at.

12" PSI in a closed boiler/heating system will support a column of water 27.72' high. If you connected a glass pipe to the boiler and ran it up 100' vertically, the water level in the class pipe would be 27.72' up the tube. If the highest emitter in the building is higher than 27.72' from the gauge on the boiler is reading 12#, there won't be any water in it. If you turn on a pump, the level might rise 1/2 the Delta P but drop when the pump stops.

If this is in a 3 story building, take a 16' tape and stand in a stairway, and measure from the top of the first floor to the top of the second floor. Whatever that measure is, X 3. Plus the distance from the gauge on the boiler to the top of the first floor. Multiply that (in feet) by .434. That is how much static pressure you need to get the water into the top floor emitter. You need more for a cushion. I'll bet the old boilers had 40# or 50# relief valves and the boilers were rated for that pressure.

The pressures you relate on the different floors are deceiving. They need to be taken when the pumps are off. You have to have the top floor pressure adequate to the system when the pumps are off. If they aren't high enough, and the pumps come on, the Delta P can drop on the return side to a negative pressure somewhere, and that's where your air will come from.

As I understand it, a 100 story building with a steam system only needs 1# of steam to get to the top floor. The entire system will be 1# of steam once it vents all the air. With a FHW system, 1# gets you barely out of the boiler.