Condenser cooling water pump, system Q-H curve vs Pump Q-H curve

[ssphogat]

This question pertains to a oonce-through condenser cooling water system, having 2*50% CW pumps/unit. The pumps are designed for low water level at the intake.The pumps supply sea water having a tidal variation of about 5m.

At high water level, the system static head requirement is almost nil ie the system head requirement is for dynamic head only.
In the event of only single pump operating for high tide level case, the pump Q-H curve terminates above the system curve ie the pump curve does not intersect the system curve. Though the gap is small, it remains.
Request your expert views on how the pump (single point operation)and system behave at high tide level situation.

Regards,
S S Phogat

 

[LittleInch]

If you look at a number of similar posts here you'll find the answer, but in short, without anything to stop it (control valve etc) the two curves (pump and system) will always intersect. In your case it means that the pump will go beyond the end of its curve provided by the vendor. It will therefore use more power (higher flow, lower efficiency), but if your motor is reasonably robust and the head / flow curve where the two interesect is not very far from the end point, then in reality you're probably OK. Withou the curves being posted we can't offer much more.

Try this post whicch has some good links and websites which might help

http://www.eng-tips.com/viewthread.cfm?qid=343897

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

Include the control valve in the system curve and they always intersect at the right place.

Independent events are seldomly independent.

 

[LittleInch]

I completely agree, but there is no mention of a control valve here.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[Artisi]

Without some form of control (control valve)to contain the flow within the operating range of the pump, it will probably run way out to the right of its performance curve and beyond and likely cavitate itself to death.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

Artisi - I completely agree, which is why we need data from the OP to see how far it has to go off the end of the pump curve to meet the system curve (he says the gap is "small")

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[Artisi]

Getting meaningful data sometimes is like winning the lottery.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[BigInch]

Lil' ".

Was it not you that mentioned it? --- "(control valve etc)"

Independent events are seldomly independent.

 

[BigInch]

IMO what will happen is that the suction head rises by 5 meters, which forces the pump discharge curve (when superimposed on the system curve) to rise the same five meters. The new operating point will still be at the intersection of those two curves, if they still intersect. It might be such that the pump curve now intersects the system curve at pump runout flow .. or doesn't intersect at some point well beyond runnout, maybe not. If you publish your pump curve and your system curve, I can show you how to superimpose both and we can all find out what it will do.

Independent events are seldomly independent.

 

[LittleInch]

Yes, but it was qualified. I think we are all of the same thought, the pump will go until it meets the system curve. What we don't know is how far that is....

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

Never had any doubt about that.
Just trying to make sure ssphogat is on the same page.

Independent events are seldomly independent.

 

[ccfowler]

As one mechanic stated some years ago, "the pump will always deliver until the motor smokes a bit too much!"

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[DubMac]

If the H-Q curve and system curve do not intersect then nothing will come out of the discharge flange.

 

[LittleInch]

We already know the pump H-Q curve is above the system curve so more than normal will come out of the discharge flange unless the motor trips or expires. It might cavitate or make a lot of noise, but you will be getting flow....

CCfowler - I'll try and remember that quote - very useful.

Ssphogat - how about a response or some more data - say pump curve and system curve...

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way