0% NPSH 4

[vesselguy]

Instead of tacking onto the discussion on NPSH below, I thought I start a new thread.

I know of the pump manufacturer's NPSHR test at 3% head loss. But is there a 0% NPSHR test? Of course anything is possible but is doing a 0% NPSHR test a norm or uncommon? Is is exactly as the name indicates? i.e, when the head starts to drop a little when you starve the inlet head then that's the 0% NPSHR.

 

[BigInch]

The difficulty in measuring 0 head loss would tend to make that impractical.

**********************
"The problem isn't working out the equation,
its finding the answer to the real question." BigInch

http://virtualpipeline.spaces.live.com/

 

[vesselguy]

Yes indeed.

 

[ccfowler]

Almost always, practical physical and economic constraints result in pumps operating with less than perfectly ideal suction head conditions.

NPSHr increases in a non-linear manner in proportion to the difference between the actual flow rate and the BEP flow rate. The NPSHa increases for flow rates below the BEP flow rate thereby inherently at least mitigating some of the NPSHr burden from off-BEP operation. The potentially more serious situation usually involves operation at greater than BEP flows where NPSHa drops as NPSHr increases. Since the pump's power requirement usually increases quite rapidly as the flow rate exceeds the BEP flow rate, pump's don't usually operate very long is this mode anyway.

Sometimes, a pump's NPSHr is based on 1% rather than 3%, so it is usually a good idea to be sure what standard is used for a particular pump's ratings. Also, it is common for a pump's ratings to be based on operation at a constant shaft speed (often but not necessarily a synchronous speed such as 1800 rpm or 3600 rpm), Sometimes, smaller, direct-connected pumps may be rated based on operation as powered by their standard induction motor so that some modest variation in shaft speed may be included in the pump's rated performance curves. Usually, these small variations in the rated performance are not troublesome, but I have encountered situations where problems of apparent performance deficiencies have resulted from a misunderstanding of the manufacturer's rating system.

 

[Artisi]

How could you test NPSH at 0% - there is no way of deciding when cavitation commences if there is no drop in performance.

 

[ccfowler]

Artisi,

I agree that testing for the magic 0% point is a serious challenge. I have never been involved in such an effort (nor am I likely to ever be so involved), but I suspect that the most likely means may involve some type of sonic or ultrasonic instrumentation--probably very fancy and expensive.

 

[JJPellin]

There have been a few technical papers written on the subject. I don't have access to my office right now, so I can't look them up. Do a search for "Insipient Cavitation" and you will get a few ideas. The papers I recall reading used a theoretical basis to determine the point of insipient cavitation based on available properties of the pump. They tested the theory using a clear pump and a strobe light to photograph the formation of cavitation bubbles in the impeller eye. I didn't see any practical application of this method. As I recall, the point of insipient cavitation generally requires such an extreme NPSH margin that it would be impossible to design for this in most real-world applications. In fact, I would go further to say that attempting to design for zero cavitation would likely result in a pump with outrageously high Nss or some other determent that would make it undesirable. Of the 2000 or so pumps in our refinery, I can think of a handful that don’t run quite will using a reasonable margin above 3% NPSHr. For critical services or water pumps with very high suction energy, I might require a 1% NPSHr test. It provides an additional piece of information that could be useful in making better decisions.

Johnny Pellin

 

[dabluffrat]

Hydraulic Institute defines the standard for determining NPSHr at a point in which the head curve deviates from a straight line, then 3% over this deviation is the "head drop point" (ANSI/HI 2.6-2000 section 2.6.6.3).

Higher temperature fluids will only have a 1% head drop applied.

0% in my interpretation would be the point of deviation from a straight line, no 3% head drop applied.

Since this is all dependant on the Test Engineer's definition of "deviation from a straight line," it's all subjective.

Did you know that 76.4% of all statistics are made up...

 

[BigInch]

This seems to be a bit of a red herring anyway, since I see little value ever having a 0% NPSHr standard, if the general pump user's typical problem is still not having enough head to prevent cavitation, even though he did provide a NPSHa > a NPSHr derived from a 3% point. For those that do know their NPSHr is an exact indicator of their cavitation point, they could always adjust the 3% points to suit their liking anyway.

**********************
"The problem isn't working out the equation,
its finding the answer to the real question." BigInch

http://virtualpipeline.spaces.live.com/

 

[1gibson]

Before purchase, ask the pump vendor for a set of typical NPSH results for the same size pump, and review them. Not a proposed NPSHr curve, but some actual test results with plots of the data.

You will see the 3% drop off point, and you can track the curve back to where you still like the shape (what you would consider a 0% value.) Measure the difference, and consider it as additional margin that you then specify.

If you want a 0% NPSH value, there is no test for it. Specify a 3% or 1% test and apply some margin to the results. You can get the information that you want, without specifying an impossible test for the pump vendor.



I've been reading for awhile, but had to join up just to post this!

 

[dabluffrat]

for reference... scroll down to 'Actual NPSH Test of a Centrifugal Pump'

http://www.pump-zone.com/pumps/centrifugal-pumps/performance-curves-and-npsh-tests.html

Did you know that 76.4% of all statistics are made up...