Axial Pump Cavitation at Low Shaft Speed 1

[tc7]

Hello all-
I am posting this pump question here because it seems the "PUMP ENGINEERING" forum mostly concerns itself with centrifugal pumps.

What could cause pump cavitation at low shaft speed only?

Explanation - We have a booster pump (fixed displacement, axial piston type) which supplies the inlet of an HP pump (pressure compensated variable displacement axial piston type). Both pumps are driven off the same shaft by a diesel engine (and in fact, both pumps are integrated within a single housing!). The boost pump circuit filters and circulates through a cooler to temper the fluid before it reaches the HP pump inlet. The booster system pressure is set by a relief valve tuned to 50 psi at shaft speed of 2500 RPM. During operation, shaft RPM is constant at 2500 RPM. During idle the shaft speed is reduced to about 1000 RPM. At this low speed condition, cavitation like noise is heard and it is believed to occur in the booster pump, but at high speed, all is well.

I had a couple of ideas and questions, but little ability to go through trial and error solutions:
1 How will the booster pump relief valve behave if instead of setting it for 50 psi at 2500 RPM, we set it to 50 psi at 1000 RPM. Could this make any difference?
2. If case leakage was excessive, could the booster pump be starving the HP pump and maybe the HP pump is really the one that is cavitating? Is case leakage ever likely to be higher at low shaft speed than at higher shaft speed?
3. How would clogged case drain filters or check valves affect cavitation behavior i have described?

What could cause cavitation at low speed only?

Thanks for any ideas or experiences you may have.
Tom

 

[Gkranz]

Reply:

I had a couple of ideas and questions, but little ability to go through trial and error solutions:

1 How will the booster pump relief valve behave if instead of setting it for 50 psi at 2500 RPM, we set it to 50 psi at 1000 RPM. Could this make any difference?
Answer: First part, no difference in relief valve.
Second part, no it will not.

2. If case leakage was excessive, could the booster pump be starving the HP pump and maybe the HP pump is really the one that is cavitating? Is case leakage ever likely to be higher at low shaft speed than at higher shaft speed?

Answer: First part, yes.
Second part, No case leakage will be higher at faster speeds and higher pressures.

3. How would clogged case drain filters or check valves affect cavitation behavior i have described?

Answer: Case drain filters will not produce cavitation, nor will check valves if they are in the case drain circuit.

What could cause cavitation at low speed only?

Answer: The charge pump is under sized and not providing the flow required at lower speeds. Increase your charge pump size to provide more fluid flow. Pressure is not your problem, flow is your problem. Also, charge pump may be worn and not providing proper flow. It takes pressure to collapse air, not enough (higher) pressure in charge pump circuit to do this, most likley it is the main pump.
Best Regards,
Gkranz


Westerndynamics.com

 

[tc7]

Gk-
Thankyou very much for a speedy and thorough reply. I should have mentioned that we have several systems using the pump arrangement I have described, all used in the same way and this is our first report of the type of problem. So I am sure the charge pump is sized reasonably correctly.

Your response helps narrow the problem down to only one possible explanation, and that must be excessive wear in the charge pump thereby starving the HP pump. BUT why does operation smooth out at the higher shaft speeds when as you say, the leakage will increase?

Also do I understand from your last sentence that the operating pressure of the charge pump is so low (50 psi) that cavitation is not likely to happen here and that any cavitation noise must be coming from the HP pump?

VR
tc

 

[Gkranz]

tc7 -

The leakage will increase with higher pressure, not higher flows. The charge pressure stays the same regardless of speed, only the flow increases with higher speeds. At higher speeds the charge pump is providing enough oil flow to the main pump.

At 50-psi cavitation it is possible but unlikley and the nosie would be very low and not heard clearly. Where as in the main pump pressures are higher and noise is much louder.

You need to jump on this ASAP, before you get very much damage to the main pump pressure plates, it will not take long, few hours in some cases.

Before you get crazy and pull the main pump, replace with a new element the inlet filter to the charge pump, or remove it and re-check for noise. If no change, then it is down to the charge pump. Something caused the charge pump to wear, you need to find out why, the inlet filter to the charge pump is most likley the cause. Charge pumps just do not wake up and say "I am going to wear more than the others", something caused it to wear more.

Old saying, "before you replace the pump, find out what killed it, otherwise whatever got the old pump will kill the new pump too". You may find several systems may have the same problems, just have not surfaced yet. Normal life of hydraulic pumps, 8-10 years, any shorter something killed it. Good luck and troubleshooting.

Best Regards,
Gkranz

Westerndynamics.com

 

[budt]

tcz;

You originally asked if it would make a difference if the booster pump was set at 50 PSI at 1,000 RPM instead of 2,500 RPM.

What pressure is it showing when the engine slows to 1,000 RPM and the noise occurs?

If it drops below 50 PSI can it be raised to 50 PSI with the adjustment?

If it can be raised to 50 PSI I would leave it there and see if the noise stops.

If pressure can't be raised to 50 PSI at 1,000 RPM I would suspect the booster pump is internally bypassing enough oil and starving the HP pump inlet.


Bud Trinkel CFPE
HYDRA-PNEU CONSULTING, INC.
fluidpower1 @ hotmail.com

http://www.fluidpower1.us

 

[Gkranz]

Bud,

What kind of advice is this, using a pressure setting to determine the condition of a charge pump?

And for the record, we call these pumps a charge pump, not a booster pump.

And now you know why Peter says hydraulic guys are dumb, no wonder, we make up our own names, use relief valves and pressure gauges to check pumps.

Its like shooting from the hip, and hitting yourself in the foot.


Westerndynamics.com

 

[tc7]

Gkranz-

"Boost" pump is the term I originally used to reference that particular pump and Bud was just replying in kind. As a matter of fact, the manufacturer (Denison) also uses the term "boost" pump in the abbreviard literature that I have. I particularly favor the term "charge" pump, but what's in a name??

Anyway, we have since found the pressure does hold steady within 5 psi. at the lower speed, so Gkranz, you were absolutely right that it would and should make no difference.

Later I'm going to get the operators to try turning the HP pump volume control all the way in and see if the low RPM cavitation sound goes away. This should give us a good clue about something. I think!

I wish I had some pump data on what the volume of the charge pump should be. These pumps are not that old but Denison (!!!!now Parker!!!!!) cannot provide me with any specifics on this model.

Hey fella's thanks alot, I appreciate your time and ideas.
tc

 

[budt]

tcz;

Is the 5 PSI difference 5 PSI lower or 5 PSI higher at 1,000 RPM?

If Lower, can you raise it to 50 PSI while at 1,000 RPM?

If pressure can be increasd it means there is excess flow and some has to go across the relief valve at 50 PSI or higher.

If pressure cannot be increased it means all "BOOST PUMP" flow is going to another path at the decreased pressure. This also means ther is not ample flow to satisfy the path that is lower than 50 PSI at 1,000 RPM.

BTW, Thanks for taking up for me about Boost Pump terminology. Gary likes to do little digs to show his superior knowledge.

Even I have heard of "Charge Pumps" and use that terminology on a regular basis when conversing with others in the Fluid Power field. Plus, as you noted even the manufacturers use different terms for the same item in their literature.

Regards:


Bud Trinkel CFPE
HYDRA-PNEU CONSULTING, INC.
fluidpower1 @ hotmail.com

http://www.fluidpower1.us

 

[Gkranz]

Bud,

It has nothing to do about little digs, Bud. We have names for our components and terminology we use in hydraulics. The terminology and component names relate to the function of the components. We have marketing people at the factory, and people like you that call the same component different names, it is no wonder we have a communication problem with customers.

Examples: Motion control valve, over-center valve, really it is a counter balance valve.
Booster pump, charge pump, correct terminology is charge pump.
Dump valve, pressure limiter valve, really it is a relief valve.

Here is the game, XYZ Company has a motion control valve, ZYX Company has a counter balance valve, XYZ companies valve is much better because it controls motion, thus the name motion control must be better because the factory say’s it is.

A dump valve works much better than a relief valve, because it dumps the pressure and not just relieves it. By the way, this is the sign you see when you pull your camper up to a site to dump tour toilet.

It is hard enough to communicate with customers without hearing that my “thing won’t shift”, or “my dump valve” is making noise. Or how about “that shinny round thing will not go in and out fast enough”. It is like some people call a reservoir a tank, two entirely different things.

Booster pump implies that a pump is raising the pressure from one level to another level, boosting. A charge pump implies that it is used to charge or keep full with flow and pressure a closed loop circuit.
Just because some idiot at Denison does not know the difference, does not mean we do not.

So it is not about hurting Buds feelings or throwing digs, it’s about a CFPE helping to dumb down our industry and making communication that much harder. I have no apology in trying to keep the standards higher, not lower.

Best Regards,
Gkranz


Westerndynamics.com

 

[kcj]

bud's Q, does the 50 psi drop slightly at low idle? or does it rise?

Also, does the cavitation get better or worse as the main pump is reduced in pressure?

fixed pump flow is proportional to speed, but leakage across the gears, sideplates, rotor or whatever is more proportional to pressure, not as effected by speed.

At all speeds, (output - leakage) is available for charge/boost flow. If the pump is just marginally adequate, through design or wear, the leakage at low idle might be bringing the output just below what it needs and cavitation starts.

fictitious example: pump theoretical is 10 gpm at 1000 rpm, 25 gpm at 2500 rpm, but 1 gpm leakage at 50 psi.
at idle 10-1 = 9 gpm out, 9/10= 90% efficient.
at 2500, 25-1 =24/25 = 96% effficient.
at 100 rpm, 1 - 1 = 0 out.

the numbers don't matter, and efficiency by itself isn't the point. I'm just trying to show that efficiency as it wears is not a linear thing, it can have a very much prounounced effect at low speeds. At low enough speed, all output goes to leakage and nothing left to charge the main pump.

i think you a charge/boost pump problem.

kcj

 

[tc7]

I meant to answer Bud's question on Friday before I left, but thanks for the reminder KCJ for the same question:
The psi drops off slightly at the low speed. The LP gauge on the system is graduated in increments of 5, so a slight drop will only be reported as 5psi even if it is less (unfortunately I am working through 2nd hand information since the equipment is in a far away remote location, we don't have direct phone contact). Adjustments have been made to increase the psi slightly at the low RPM. This apparently made no difference in the observation of cavitation noise. My last instruction was to turn the volume control on the HP pump in (reduce) and tell me what they observe, but they have not reported up as yet.

Bud - what is being implied by your statement, " If pressure can be increasd it means there is excess flow and some has to go across the relief valve at 50 PSI or higher. " ? I don't follow the suggestion about "excess flow" . Isn't the charge pumps output fixed at 1000 RPM? or are you saying that due to the higher restriction (the additional ~5 psi adjusted on the relief valve), more flow is likely to get squeezed through the leak paths? An explanation would be helpful to me. Thanks.

 

[budt]

it is as kcj explained. A pump has a known cubic inch/revolution displacement due to the swept volume of the gear teeth, vanes or pistons. Taking the CIR X the RPM gives the maximum flow possible at 100% efficiency.

All pumps have clearances so they do not pump 100% of the swept volume and bypass doesn't change as RPM changes. That is why % efficiency is at a rated RPM and decreases as RPM's decrease.

What I was trying to get across is that the charge/boost pump will only be putting out excess flow if the relief valve is bypassing some oil at set pressure. If it cannot be set to a higher pressure at the lower RPM it means there is no excess to makeup for the HP pump losses.

As kcj mentioned the charge/boost pump flow may have been marginal when new and with wear has dropped below the required amount at reduced RPM.



Bud Trinkel CFPE
HYDRA-PNEU CONSULTING, INC.
fluidpower1 @ hotmail.com

http://www.fluidpower1.us