Pump Cavitation with High NPSH

Put the pump at the tank end of the run. Or, at least, a booster pump.

I don't think you've got any NPSH at the pump after over a 1/2 mile run of pipe.

Line size and Pipe flow resistance need to be considered. That's a LOT of pipe to have on the inlet side of a pump, bad initial design. Put the pump close to the tank and run the long line off the outlet of the pump... NOT the inlet.

That`s an epic length of suction line.

I`d also agree that any pump would be better positioned at the tank in an atmospheric system. The negative pressure created in your system will also find any microscopic holes in the suction pipe which will result in air being dragged into the line and contribute to cavitation.

Is low tank level causing a vortex maybe?

To be honest, I`d re-design for larger pumps connected at the tank end. Ensure full tanks at all times and conform to the design pipe dimensions required both suction and discharge sides as already stated.

I haven't looked into vortices at the tank but the negative pressure is close to the pump (not at the tank).

It seems like everyone thinks I should relocate the pumps closer to the tanks which is a very good suggestion for cavitation. Unfortunately, are located here for close proximity to the control room and to minimize the length of wire needed for the pump motor.

I think we could fix the problem by raising the LL level of the tank a few feet.

I was looking for some suggestions on why it's cavitating now. My limited knowledge on the subject tells me that if I'm above vapor pressure, I shouldn't cause vapor bubbles to enter the pump. Of course, I've always located pumps like this within 100 feet of the tank.

If you cannot move the pump, then increase the line size extending from tank to pump, and within the last ten feet to the pump, install a "bump tank", the purpose of the bump tank is to hold a reserve volume of oil, with an even larger suction pipe leading from the bump tank to the pump. Be sure and install baffles in the bump tank to prevent vortexing.

As a final last ditch effort, one could perhaps lower the bump tank, and the pump to achieve higher NPSH under flow conditions to the pump.

If your liquid is at ~10 psia during even part of the operating cycle, you might be drawing air into the system.

Air bubbles cause problems and can mimic true cavitation.

Describe your pump type...

Also, are you considering the true vapor pressure at an accurate temperature in your NPSHa calculation ?

As said above, pump beside the tank. Such a suction is against basics of pump installation.

Regarding the NSPH, it is a static value.

Instead, the pressure drop along the line with the speed of the fluid, a dynamic value.

Check value of the Lower NPSH and the pressure drop in the line.

Results smaller than the NSPH required by your pumps and fluid?...

Are there chances for air pockets in the line?

Maybe a close examination of the actual design and a change in the position of the pumps, can payback handsomely, avoiding the necessity of new pumps.

If the increase of flow required is such that change of pumps is unavoidable, it could be good to estimate the power saving with different configurations.

Let us know.

Best regards,

Abel

If you are cavitating the pump, then there is obviously some parts of the impeller that are under vacuum. I agree totally with other comments: Move the new (larger) pump very near the tank, and make sure the supply piping is several pipe sizes larger than the discharge piping you require. You failed to account for the viscosity of the oil reducing the pressure at pump suction during peak flow.

At low pressure, lot of diffrential pressure loss will be there through the suction pipe loosing the NPSH available and it is obivious thart pump is cavitating.

3000 ft of suction pipe line for oil is too long. I think you have to put the pump near the tank to pump the oil at the required destination and suitable discharge piping is needed in place of suction line. Existing suction line can be checked for suitability to discharge conditions.

I suggest you obtain a copy of Cameron Hydraulic DATA and review chapter 1 in detail.

This is an exceptional collection of hydraulic DATA and is a priceless reference for/when designing system and/or troubleshooting flow characteristic based issues in pumping systems.

Alas, you will find relocating the pump closer to the tank is the most cost effective and viable solution.

The reason for the cavitation is:

Excessive, additonal, negative (suction) pressure is being expressed on the fluid by the larger pump impeller which is far greater than than the fluid's ability to convey through the flow restrictions (dynamics) present in the piping and out of the pump body.

The new larger pump is being operated outside of it's suction design capacity.

In short: There is too much flow restriction/resistance in your intake piping for the larger pump impeller to overcome in order be able to lift/move the quantity of fluid it is designed for conveying so the water is literally being "pulled apart" causing cavitation.

Thanks!

Actually I was going to buy that book this weekend but forgot to make the trip to the bookstore.

I think we might be able to get rid of the cavitation through operations controls like lowering the minimum tank levels. The great thing is I can work with the operations guys to keep it at this level with the current pump and see if we get any cavitation.

I appreciate everyone's input.

You can stop the cavitation by throttling the pump output however you would be defeating the object of using a larger pump unless the loss of pump output is acceptable. (Check for cavition present in the pump by using a ultrasonic tester.)

Use a manual controlled valve on the pump output to throtle the pump discharge until cavitation is detected and measure the flow.

If the losses in pump flow output when throttled are acceptable, you could apply/install an automated control valve on the pump discharge with a control PID algorithim based upon min-max flow values to meet the dynamics of your operating system curve.

Check the formulae in the book and plug your numbers in for your application.

The results will reveal whether increasing the tank level will address your issues as well as tell you how much to raise and/or clamp the tank level.

You may want to check whether there are any sudden turns just prior to the pump inlet. It could be a case where you need "straightening vanes" near the pump suction.

The suction line piping configuration is very imporant in avoidig cavitation. The suction line into the pump should be a staright piece preferably 10 and not less than 5 diameters in length and that last elbow should be a long radius type. If there is a reducer at the pump suction it should be an eccentric type to avoid trapping air in the suction line. If space dictates a shorter section of suction piping install a suction diffuser.

If your pump supplier is worthy of your confidence and business he should be able to advise you in this application.

Good luck.

Lou Bindner

Dear cingold,

You have to provide the following details to arrive at a solution.

1. Pump Suction entry size and valve if any and any problem with valve

2. Line size.

3. Exit size at the tank.

4. Viscosity of fluid if the fluid is other than water.

Once you post this we shall be able to identify.

DHAYANANDHAN.S

Recommendations:

1. Move the pump to the tank end of the pipe.
2. Check that the tank is adequately vented and cannot be operated with the vent closed.
3. Make sure there is a tank-low-level facility to shut the pump off at low level.
4. Carry out a hydraulic test on the pipeline, and record the outcome, aiming to fix any leaks that become apparent as a result of this test.
5. Put a vortex breaker in the tank outlet connection.
6. Then return the equipment to service.