When to use a make up water feed tank?

heatfever

For single pipe steam, at what size boiler do we start using or offering a water make up tank?

Thanks

nicholas bonham-carter

I think that the only use for a condensate pump and tank would be for a multi building system, served by one boiler.
As long as the pressure is kept low, and the main venting generous, there should be no trouble in getting the water back to the boiler by gravity alone, in a single building.
Most modern boiler have been designed to hold enough water in the steaming water level range to fill a system whose EDR matches their heating output, and the old ones certainly were.
The pump and tank are additional complications in terms of long term reliability, (although some may say "service opportunities"!).--NBC

heatfever

Yes, I am mainly speaking of multi unit apartment buildings. And not a condensate pump, I understand they are mainly used to return condensate where gravity cannot. But a feed tank/ pump that has make up water. Are there instances where a typical water feeder can't​ keep up with the demand for water ? And this type of set up would be used.

MilanD

We have a 50 gallon condensate return tank with a feeder motor on a Weil McLain LGB 7 boiler.

I think it has more to do with wet-return piping capacity and volume of water it will hold. We have virtually no wet returns, as our 3 loops start and end in the boiler room, with no dry returns on 2 and 1 35-40 ft dry return on the 3rd loop. We have 2 traps on the end of each of the 3 lines, one is FT (Hoffman 55), and the other one is Warren Webster 78 (a thermostatic trap), parallel with Hoffman 55.

If you have enough room to run some sort of dry/wet return combination with pipes alone, that will keep enough water for make-up in the wet return as the system is heating, esp. thinking of capacities for the initial start-up or any deep set-backs heating cycles, you may not need the tank. This is where math comes in: volume per pipe diameter x ft of pipe needed to provide enough condensate for make-up.

I'll let the experts give more details as to when and how this is.

EBEBRATT-Ed

It has nothing to do with the boiler size. It has to do with the time it takes for condensate to return to the boiler.

Picture this, a vertical building (high rise) will return condensate faster than a long low spread out building. gravity helps

Under normal operation you don't want the make up water feeder to add any water to the boiler. excess make-up causes corrosion of piping and boilers.

If you start the boiler with a normal water level and the make up water is activated and condensate has not been returned fast enough then you need to find the problem.

if it used to work and doesn't now you have a trap or venting, or plugged return problem or piping problem.

If the system is clean and there are no issues causing the condensate to be "hung up" and it doesn't come back fast enough then you need a boiler feed tank which will have make up water and boiler feed pumps

Jamie Hall

And just to be clear -- if you do elect to, or need to, use a condensate receiver tank or make up water feed tank or whatever you want to call it -- the pump must be controlled by the water level in the boiler, not the water level in the tank, and any makeup water (there should be very little, unless you are dumping condensate somewhere -- which in a heating system you should not be doing) is added to the tank, when it's level gets very very low, and not to the boiler.

jumper

A tank ensures you won't needlessly add water. If you can have a tank high enough to not require a feed pump that is another plus. Also then a condensate pump won't flood your boiler. The ideal situation is when a boiler is low enough that condensate returns by gravity to a high enough tank and then you don't need no stinking pumps.

nicholas bonham-carter

Weil Mclain call the high unpumped condensate tank a reservoir tank, and show its configuration in their installation instructions.
It's horizontal center-line must be at the operating level of the boiler, and connected to the steam side, and the wet return for pressure equalization.--NBC

Pumpguy

As others have said, need for a boiler feed pump set depends on if your boiler has enough water capacity to fill the entire system with steam, and condensate returning soon enough to provide make-up water before the LWMU kicks in.

Storage capacity determines steaming time available from boiler feed receiver. Some folks say 10 minutes is acceptable, while others recommend enough storage capacity for 20 minutes steaming time.

Something else to keep in mind is the effect of vacuum of condensation in the header drawing water from the boiler feed receiver, flooding the boiler and header.

During mild weather when boiler is only firing part time, steam can condense in the header and boiler. When this happens, the steam collapses and a high induced vacuum will occur which will suck feed water from the boiler feed tank and flood the boiler and header.

This happens because the boiler feed tank is vented to atmosphere, so atmospheric pressure pushes the water from the boiler feed tank into the boiler and header causing a flooded condition. The boiler feed tank is now low on water, so the makeup valve opens, adding extra water which floods the system and needs to be dumped when the boiler fires again.

The real solution to this problem is to have a positive closing valve on the discharge of the boiler feed pump. This valve should only open when the boiler feed pump is operating and feeding the boiler.

I suppose vacuum breaker(s) on the steam header could also solve this problem, but vacuum is usually a desirable condition, and vacuum breaker(s) would prevent that from occurring.

jumper

Pumpguy said:

Something else to keep in mind is the effect of vacuum of condensation in the header drawing water from the boiler feed receiver, flooding the boiler and header.

During mild weather when boiler is only firing part time, steam can condense in the header and boiler. When this happens, the steam collapses and a high induced vacuum will occur which will suck feed water from the boiler feed tank and flood the boiler and header.

This happens because the boiler feed tank is vented to atmosphere, so atmospheric pressure pushes the water from the boiler feed tank into the boiler and header causing a flooded condition. The boiler feed tank is now low on water, so the makeup valve opens, adding extra water which floods the system and needs to be dumped when the boiler fires again.

The real solution to this problem is to have a positive closing valve on the discharge of the boiler feed pump. This valve should only open when the boiler feed pump is operating and feeding the boiler.

There's positive closing float valves and equivalent but they must have there own issues or somebody would be using them.

Pumpguy

@Jumper, Positive closing valves on the discharge of boiler feed pumps is pretty much a standard feature on commercial and institutional boiler jobs. I always recommend them whenever there is a chance the boiler will be firing intermittently.

Unfortunately, not everybody takes into consideration the effect of induced vacuum due to condensing steam causing a flooded boiler and header. I've had several recent conversations on this very issue on a new installation in my area (not a job I sold or consulted on), so thought it would be helpful to post this advisory here.

MilanD

We have a condensate return tank with pump operated by a float. Only a check valve as tank sits below boiler. We do have a vacuum breaker on the top of the controls tree (LGB 7 weil mclain). No issues with water being sucked in.

MilanD

But, I am curious, how is a condensate tank different from a wet return with an air vent above it? Both are open to the atmosphere, so why would one suck in water into the boiler and and the other one not? Wouldn't the steam being produced by the boiler just be drawn instead of water?

jumper

>>how is a condensate tank different from a wet return<<

elevation? At least in some circumstances.

MilanD

jumper said:

>>how is a condensate tank different from a wet return<<

elevation? At least in some circumstances.</p>

Not for us... here's the set-up... That's why I'm asking how is it that it would vacuum up the water from a tank, but not a wet return?

https://www.amazon.com/photos/share/xp7smSncmSC5xdkTGSxSKmZGFZS167Tv8dkzReb1OPr

Pumpguy

It all boils down to differences in pressure. As Dan says in TLAOSH, "All fluids flow from high pressure to low pressure; Always"

If the pressure on one side of a water column is lower than atmosphere (at vacuum), and the other side of that water column is open to atmosphere, the pressure of the atmosphere will push the water column into the lower pressure area.

heatfever

Would it be considered an appropriate use of a condensate receiver and pump when you have low returns that don't meet the dimension "A" requirement in a single pipe system? And how would you protect the pump from steam?

MilanD

Pumpguy said:

It all boils down to differences in pressure. As Dan says in TLAOSH, "All fluids flow from high pressure to low pressure; Always"

If the pressure on one side of a water column is lower than atmosphere (at vacuum), and the other side of that water column is open to atmosphere, the pressure of the atmosphere will push the water column into the lower pressure area.

@Pumpguy

I get this dimension A stuff and all that equalizing from high to low pressure. Can't argue about the physics: vacuum will pull water if other side is open to atmosphere. Dimension A makes this easier as it creates pressure on the return side that pushes the return condensate into the boiler, and is needed mostly to overcomes the boiler pressure when the boiler is firing and maintain the water line.

What I don't get is that we don't have an issue with boiler sucking in the water from the condensate return tank, and based on what you wrote above, it should. Therefore, my only conclusion is that there is no collapsing of the steam in the header that's significant enough or long enough to create vacuum strong or long enough that will suck the water out of a wet return, or condensate return tank. IF the boiler is making steam, any collapse of steam as the header is being heated is just replaced with more steam from the boiler (less friction and more of it is coming in), thus, I can't see how the flooding will happen because of this. If it were true, it would happen also with regular wet returns (on the floor) and with the or without the dimension A, and in practice we see that it doesn't.

Have you actually seen this happen? I'm trying to wrap my head around which condition this would happen under... My experience is that this is not happening.

I'll go ahead and remove the vacuum breaker we have on the control tree and see if this changes the behavior of our boiler. I can not remember why this was installed in the first place, but I think it's superfluous.

To answer the OP question: if you have enough wet return capacity, and maintain the proper dimension A of 28 inches, you should be just fine...

See here for some details on various steam systems and when is the condensate used:

https://heatinghelp.com/systems-help-center/a-steam-heating-primer/

jumper

heatfever said:

Would it be considered an appropriate use of a condensate receiver and pump when you have low returns that don't meet the dimension "A" requirement in a single pipe system? And how would you protect the pump from steam?

I don't think anyone makes a low npsh pump small enough for a home steam heating system. And it is difficult to protect them from steam/boiling water unless there's room for a receiver. There are compressed air motivated pumping devices that are immune to steam/boiling water. They can lift condensate to a tank above the A or even B dimension. Holohan once wrote about somebody who dug himself into a deep hole by filling in a pit when changing a boiler. Irony intended.

Pumpguy

@MilanD , Vacuum of condensation, AKA induced vacuum, only occurs when the boiler STOPS making steam, and the steam that it did make that is still in the header and boiler condenses. The volume change is 1/1600, IIRC. Typically this occurs in mild weather when the boiler is only firing part time and shuts down for several hours.

@Jumper, SHIPCO makes a 1/3 HP condensate pump that is a 2 foot NPSH type. They are available with either single phase or three phase motors.

MilanD

Pumpguy said:

@MilanD , Vacuum of condensation, AKA induced vacuum, only occurs when the boiler STOPS making steam, and the steam that it did make that is still in the header and boiler condenses. The volume change is 1/1600, IIRC. Typically this occurs in mild weather when the boiler is only firing part time and shuts down for several hours.

Wouldn't collapsing steam just suck in air from elsewhere in the system, or from all the vents on radiators, or move steam from them out?

Pumpguy

Murphy's law says "Anything that CAN happen, WILL happen"

Dan Holohan wrote "All fluids flow from high pressure to low pressure, always."

To clarify, steam, air, and water are all fluids.

Each system is different. Just depends on conditions at any particular time.

Whenever I am asked to comment on a problem or make recommendations, I always try to keep these two basic principles in mind.