Impeller blade length affect of flow 1

[klyons]

I am not that knowledgable about pumps so if anything I say isn't correct feel free to correct me.

I have a centrifugal pump that is getting tiny holes in the impeller. I know this is due to cavitation. The pump is pumping much more than we need (its giving over 1100gpm when 950gpm is all that is required) so the dscharge is pinched. I believe this is causing the fluid to back up and stay in the pump becuase it can't escape and is causing the cavitation.

My theory was to put a smaller engine in the pump so the RPM and therefore the gpm produced would decrease.

However, another idea was proposed which makes sense to me but I don't know if it would really work or reduce the flow the amount we need. The idea was to machine the impellor so that the blade length was shorter and therefore there wouldn't be as much contact area.

As I said I think this makes sense but I don't know if it would reduce it enough. Also if this is a plausible solution how would one calculate what length is actually needed so we would know how much to reduce the length by?

 

[MortenA]

It dosnt make sence!

Cavitation occurs when the inle pressure is lower than the vapour pressure of the liquid that you pump.

As the liquid (with vapour bubbles) move from the center of the impeaalr towards the edge (and discharge) pressure increases - and when the pressure is abouve the vapour pressure then the bubbles implode - a quite violent affair that can cause excessive errosion.

So the bottom line is that cavitation is dependant on inlet pressure. In most cases (all that i can think of actually) less flow would mean higher pressure. Low pressure could however also occur if e.g. the feed is from a tank where the level varies - that is at low levek in feed tank then inlet pressure will be lower.

Best regards

Morten

 

[BigInch]

Trimming the impeller will reduce head.
It may be a case of too much flow is reducing the NPSHa below NPSHr, in which case pinching back the flow is exactly what you want to do. Try a discharge orifice plate designed for a fixed flowrate of 950, or a valve if you need to vary flowrate, or running at 90% rpm.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

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

 

[JoeySoap]

I'd put the pump on a test bed driven by an inverter drive and try to match the pump characteristic to your needs at a lower speed. This is more efficient than throttling and lower speed will greatly help with NPSHreq.

 

[Artisi]

Until NPSHa/r at the flow rate of 950 is know it's all guess work. If NPSHr exceeds NPSHa at this flow rate then cavitation is very likely and could account for the impeller damage.

More data required and a photo of the damage would also help.

 

[MartinSr00]

Altering the pump impeller geometry too much is a bad idea. If you're wearing holes in the impeller from cavitation in a short time (less than a few years), your NPSH margins are way below what they need to be. Trimming the impeller OD will likely not fix this.

Do a little homework.

Get a copy of the H-Q curve and NPSH curve. If you can't get them, run a test. Even if it's a crude test. Get some flow indication in the line and use a discharge and suction pressure gauge. Get the hydraulic institute standards for guidance on running a test.

Once you do this, analyze your system. You have to get your NPSH above the NPSH required. Remember that NPSH generally behaves like TDH with regards to the pump affinity laws. Slow the pump down, and NPSHR goes down (given the same affinity "pull point"). But TDH goes down too. Maybe that won't work with your application.

In your calculations, remember what NPSH is: It's the absolute head, including velocity head, less the vapor pressure head of the fluid. You'll want a margin of 20-30% or better NPSH over NPSHR.

I did a lot of cavitation test work in the 1980's and found that cavitation is present on the vanes at NPSH's 2-3 times NPSHR. You'd go broke if you insisted your pumps have no cavitation. Almost all do. The point is, you don't want the cavitation so advanced that the vapor bubbles in the impeller passages are super numerous and imploding on flow passage surfaces.

In the pump industry we used to say, "Every pump is a good pump". But to perform well, it must be given a good home. In the end, you may need a different pump, a modified system, or a pump moved to a location (perhaps at lower elevation) where NPSHR will be lower than NPSH.

It's much cheaper to solve your problem with a calculator or spreadheet and a bit of test data. Experimenting with different hardware can get expensive.

 

[Pumpone]

Hi,
The more common problem of cavitation is because poor succion pressure.
But there are others cavitation problems because of high velocity, eddy in recirculation flows, eddy in over capacity and thermal cavitation.
You have a lot of system characteristics that you can change to improve NPSH margen:
-Check operation point and correlation with BEP.
-Check is vapour pressure of fluid is high.
-Check minimun thermal and continous flow values.
-Check Suction Energy level for best NPSH margin.
-Check suction pipe and fittings to maximize NPSHa.
-Check Lower RPM with larger impeller or inducer.

 

[Artisi]

Pumpone -- this and the other threads you have responded to have been inactive now for weeks - it is unlikely the OP's are following this information now.

 

[lonovane]

Cavitation damage is most probably due to over capacity pumping where NPSHr exceeds available NPSH. Pinching the valve wastes energy. Much better solution, as you suggest, is to trim the impeller. To do this properly, you need the existing pump curve, impeller diameter and static/ friction head at your desired flow rate. You can decide on the new impeller diameter by applying affinity laws. One can help, if the above data is available.

lonovane