Pumped Condensate backing up prior to reaching Deaerator

maddymorg

Pumped condensate is not returning to the deaerator (set to 6 psi). It may not have been returning since the DA was installed to replace a feedwater tank, but I'm not sure. The owner has to open a manual valve to atmosphere on the pumped condensate line (near the DA) and drain out all the condensate various times throughout the day to keep the system running.

This is a process steam system where 75 psi steam heats up various kettles, modulated by control valves. F&T traps collect condensate and send it to local electric condensate receiver pumps (8 to 10 pumps total). The pumps are tied into a pumped condensate return line, which extends to the deaerator pumped condensate connection (also tied to the make-up water connection). High pressure returns are tied into the deaerator separately.

There are a couple things I believe are happening:
1. The local condensate pumps are not large enough to handle the back pressure with the DA. Each pump is approximately 25 GPM at 50 FT head with a 6-gallon tank. The DA is at 6 psi (12 ft), the pumps have to lift approx. 10 psi (20 ft), and there is a long run of piping (in some runs it can be ~10-12 psi).
2. There may be a vacuum forming when the pumped condensate falls to the DA. Or when the control valves on the steam side close?

Solutions:
1. Long term solution in my mind is to install a surge tank. This will be atmospheric, so the pumped condensate can gravity drain to the tank.
2. Short term solution... Install a vacuum breaker at the high point of the pumped condensate line, just before it falls to the DA. This would allow atmospheric pressure to overcome any vacuum forming in the pumped condensate line. I believe this would overcome the 6 psi DA pressure since 14 psi absolute is higher.

All thoughts and suggestions would be greatly appreciated!! Thank you.

Jamie Hall

Are the local condensate receiver tanks vented? If not... they need to at least have vacuum breakers on them. Your surmise that a vacuum forms in the lines to the condensate receivers -- and this will happen if there are no vacuum breakers any time the valves close. The F&T traps can't be counted on to do this for you.

Also your pumps are marginal. You have the potential of 50 feet of head on the pumps at that 25 gpm flow, which is what they are rated for, but any vacuum on the tank side will reduce your margin.

maddymorg

Jamie thanks for the quick response. Yes, each local condensate receiver tank is vented. They all pump up about 20 ft and tie into the top of the return line that runs back to the DA.

EBEBRATT-Ed

Sounds like not enough pump.

Since it is not a closed loop you have to overcome the lift, the DA pressure and the pipe and fitting resistance as you already know.

The other question is how hot is the condensate? You may need low NSPH condensate pumps, you could be cavitating at the pump inlet.

How much head is on the condensate pumps? You may not have enough head for the condensate temperature especially with HP steam and atmospheric pressure in the receiver.

Are all the condensate pumps doing this?

maddymorg

The condensate pumps are operating within their temp range (using a temp gun); it seems the control valves and the FT traps are allowing the condensate to cool. The pumps are all floor mounted, with kettles raised up about 3FT above them.

The furthest pump is about 500 ft from the boiler room. Is there a point when a boiler plant expands so much that you start to look at bringing everything back to an atmospheric surge tank (condensate receiver)? Just seems like a lot to potentially upsize all 10 local condensate pumps, especially when the piping runs are that long.

DanHolohan

I like your solution #2.

Pumpguy

@maddymorg You say the condensate pumps are operating within their temp range. Could you tell us what that operating temperature is?

If above 190*F. or so, you may want to consider using special 2' NPSH pumps which have both centrifugal and axial prop impellers and are specially designed to handle high temperature condensate.

retiredguy

When the condensate pumps run do they make noise like they are pumping marbles, vibrate or run at a higher than normal amp draw, if so the pumps could be cavitating? Cavitation can cause excessive pump and impeller wear, seal failure and bearing failure and higher than normal current draw.To help with trouble shooting the pump's problems you could install pressure gauges with snubbers or orfices on the discharge piping to check the discharge pressure when the pumps not performing as you feel they should. You could also add a discharge valve to control the pumps discharge flow when cavitation is suspected, or switch the pumps to a low NPSH type as @EBEBRATT and @Pumpguy stated. Do all the pumps seem to have this problem or is it just certain pumps? If you can determine a certain pump as the culprit check the equipment that that pump serves and see what is happening with just that piece of equipment. For example, if a kettle is calling for steam does this happen when the steam valve first opens or when it is calling for max steam flow.

If this problem has just surfaced since the DA has been installed, the height of the DA plus it's operating pressure may now exceed the pumping capacity of your pumps. The installation of a large condensate dump tank (collection tank) at ground level may be necessary or they can re-engineer the condensate pumps that are causing the problem. By the way, just as a matter of reference what is the model and MFG of the DA tank?

STEAMFITTER597

Remember condensate leaving those traps is still 75psig hot. Cavitation could be issue. Can’t flash to steam until pressure is reduced. Does each have a pump discharge check?  Back flow is killer. Receiver tanks should be vented. Most nor rated for pressure. Kettles are modulating.  They should have vacuum breaks.  Your deairator pressure/temperature is what?  Your pump head is a bit low. All combine pressure can pump handle?  When he drains to get pumping. At first does it suck or blow?  

mferrer

Probably not an issue, but just I case:
Are all of the check valves properly operating on the pump discharge(s)?

Preventing one pump discharge from affecting another.

retiredguy

One thing I forgot to mention is that all the condensate from the high pressure devices is normally piped directly to the DA tank unless it is contaminated by some other device or vent in which case it would be piped into a vented flash tank and then pumped into the receiver side of the DA.