31.1 NPSHr . . . too close for comfort? 1

[Subystud]

Pump application has me at 31.1' NPSHr for a 50hp, or $4500 more for a 60 hp variety with 18'.

Application is a hydronic 35% glycol loop recirculating out of an open tank. Level in the tank from pump centerline is atleast 2', adjustable up to 4'. Suction is ample at 4 ft/s up to pump connection.

Is this too close for comfort? At sea level, 33.9' plus 2' = 35.9' NPSH available. Should be good, right? Only concerns may be long term operation, wear, and if this could lead to cavitation issues.

Thanks

 

[LittleInch]

It's a bit tight if you include even something for exit losses from the tank and friction losses from the pipe, plus a small amount for vapour pressure.

Always remember that cavitiation limit / curve is higher than NPSH, commonly by 3-6 feet. See if the vendor can supply that curve or absolute head for your duty point

NPSH is measured at the point where the differential head reduces by 3% as this is measurable.

Your margin could easily come down to 3 feet or less.

Too close for me. Can you lower the pump even by 3 feet or so. Makes no difference to the pump differential, but might be cheaper than a different pump?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ashtree]

In theory if you have everything right you have a margin. And a margin is a margin. But do you have everything right? Are there any unknowns or uncertainties? How well do you trust the pump performance curves? How much variation in operating parameters is likely or possible?

I would agree with Little Inch that its a bit close for comfort and i would normally reject a design if it was that close. But if the consequences of failure are low and or its very difficult to do something else it would be worth proceeding with the current pump. Otherwise consider some alternative design or pump.
I have previously posted a reply to a different problem and included pictures of a pump that had a cavitation failure. I dont remember the numbers now but it had been designed with a margin about as slim as this one and had two major impeller failures in about 12000 hours of operation.

http://www.eng-tips.com/viewthread.cfm?qid=407008

The first failure occurred after about two years and 6500 hours of operation. There was close to 10 years of operation and about 12000hrs when the second failure occurred. The failures were unexpected and the exact cause unknown. However the design was very close to the NPSH limits.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[Artisi]

Too close for comfort and would be classed as a very poor engineering decision.

Can you elevate the tank by a few feet?

Can you post the curve?

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.)

 

[Subystud]

Pump skid mounts directly to the tank skid and both sit on the ground. It's not possible to adjust the height of the pump. I can easily get 3' above the centerline of the pump, so 3' liquid level above pump centerline is what we should assume. 33.9'+3' = 36.9' NPSH available.

See pump curves attached. Sounds like we're a bit too close for comfort. . .

 

[Subystud]

and the 50 hp pump curve.

 

[Artisi]

The smaller unit is very marginal re NPSHr/a and the larger unit has poor efficiency, I would be looking at alternate pumps / manufacturers to see what is available.
Worth running power costs over a year based on efficiency, you might get a surprise as to the savings, cheapest initial pump cost is not always the best selection.

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.)

 

[Subystud]

Artisi: here's my cost over one year considering 7,000 hrs/yr and $0.08/kWh

50hp (45.46 hp rated power)
$19,984/year

60hp (49.26 hp rated power)
$21,655/year

So the 60 hp variety cost $1670 more per year to operate versus the 50 hp. The 60 hp costs $4500 more than the 50 hp.

I don't understand what point you're trying to make regarding "cheapest initial pump cost not necessarily being the best selection". 50 hp variety has lower OPEX and CAPEX. Your comment makes it seem as though you thought the 60 hp pump would have a lower OPEX, thus it's able to make up the difference in CAPEX.

 

[Artisi]

NO, I said that the smaller pump is marginal on NPSH and the larger pump was not particularly efficient and to look at alternate pump selections / manufacturer to find a pump unit that meets NPSH and with a higher operating efficiency, I didn't say to compare the power costs of the two units you are discussing, plus it is very obvious the 60 hp unit has higher running costs as it is less efficient.



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]

Your 50hp pump is running twisce as fast which is probably why it has a much higher NPSHr. It probably also means that it's cavitation start level is also higher than the 60hp pump which runs slower.

As Artisi says, other pump suppliers exist and may well have a pump more suited to your particular needs.

I will say that 84% efficieny is very high - if you could get the 50hp to work then it would be best, but the margins are too tight given that the inlet head is effectively fixed by the sound of it.

Have you asked the pump vendor if he can put an inlet screw (inducer) on the pump inlet to lower the NPSH? Often offered as an option if you ask.

Like this.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

Is there a vortex breaker of sorts on the exit line from the tank? Strainer on pump suction ? You've got to be careful with strainer head loss - the numbers vary somewhat amongst vendor quotes. What is your suction head loss at max pumping flow / min level / max viscosity ? Given that this NPSHr is not performance tested/verified, what design margin would you allocate to the quoted NPSHr? Is there a capacity recycle line, and does it go back into the tank ? A sketch you can show us will help.

 

[Artisi]

Select new pump and save all heartache.

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.)