Saturday, December 13, 2008

Pump Pressure

Question of the day: What 2 things determine pressure at the pump outlet?

11 comments:

Anonymous said...

flow and resistance

HUVVA

JDK said...

Almost: 1. Flow in of itself does not cause a build up of pressure at the pump; it is the Downstream Resistance to this Flow; this is the part of the pressure that causes heat in the system. 2.the other component is the pressure required to do the work such as in a hydraulic cylinder or a hydraulic motor; for the most part this pressure does not cause heat. Obviously any leakage in either work device would contribute to heat (resistance to flow) and not work.

pnachtwey said...

The pump curve and the system curve determine the operating pressure and flow. The operating point is where the two curves intersect.

JDK said...

I agree with your comment and appreciate it but I believe that is a step ahead of the question.

Ex. #1: The pump puts out flow (1 GPM). If it is routed directly to tank through a 3" tube - no work, negligible resistance to flow = negligible pressure at the pump outlet.

Ex. #2: The pump puts out flow (10 GPM). If it is routed directly to the inlet of a motor, through the motor, and back to tank through a 1/2" tube.

Pressure at the pump = Pressure to do work plus Pressure to get flow to and from the motor (or from the motor to tank).

Again, I fully agree with your comment but believe it's useful to understand this principle so it can be used to help determine the "system curve" that you discuss.

pnachtwey said...

So the pump curve is determined by the pump and should be provided by the manufacturer.

The system curve is the "resistance to flow" as a function of flow. So how does hydraulic harry compute the system curve? This may be more math than what hydraulic harry can do. Can the VCCM equation be used to generate a system curve?

This topic would make a good H&P article since it applies to bang-bang systems without accumulators.

JDK said...

I think there is a lot going on here so I will keep my comments directed at the question.

I am familiar with 3 types of pumps: fixed volume, variable volume pressure compensated, and variable volume via electrical, pressure, or manual control.

All 3 types of pumps supply oil to the system usually called flow. If each pump outlet was connected directly to the reservoir with a relatively big line vs its flow rate, there would be no resistance to flow(actually negligible would be more correct) and there would be no work being done. The pressure at the pump outlet would be approx. 0 PSI in all 3 cases.

If we add a flow control in each line we can change the resistance to flow and the pressure at the pump outlet. In addition to the resistance if we add work in the loop, for example a motor turning a fan, we can affect pressure at the pump outlet by changing the drag on the fan.

The type of pump used may change its flow output based on the pressure it sees and even limit max. pressure but the pump does not determine the pressure.

Having said all that its probably correct to say that in the case of a pressure compensated pump the pump limits the max. pressure.

However the pressure at the pump outlet is still determined by the resistance to its flow and the work that the flow is causing.

pnachtwey said...

That assumes the pump has necessary power. It is a one sided argument. If I increase the resistance to flow to infinity do I get infinite pressure? No! Pumps don't have infinite power.

The pump has equal say in determining the pressure. If the pump is turned off there is no pressure. If the pump power is reduced so will the pressure.

That is why this topic really must be discussed in terms of the pump curve and the system curve and where the two intersect.

JDK said...

I appreciate your comments.

"The pump has equal say in determining the pressure. If the pump is turned off there is no pressure."

I don't understand how this can be correct.

Fixed Displacement Pump puts out so much flow per revolution. An external relief is used to limit max pressure to protect the pump and system. The pump puts out flow/rev and has no say in the pressure.

Pressure Comp Pump puts out so much flow per revolution until outlet pressure approches the compensator setting. The compensator limits max pressure. Below compensator pressure the pump has no say in the outlet pressure.

If the pump is turned off, there is no pressure because there is no flow.

pnachtwey said...

http://www.gouldspumps.com/cpf_0011.html

I am not making this stuff up.

Your fixed displacement pump example doesn't invalidate the pump and system curve. What you have done is changed the system curve. Power the fixed displacement pump with a variable frequency drive. You can then add as much power to the system as you want to the limit of the VFD. As speed is increased the operating point will move up the system curve and the pressure will increase until the pressure opens the relief valve. Here the normal system curve changes and become a nearly horizontal line rising slight to the right.

If the fixed displacement pump changes from stopped to some fixed RPM and the operating point may skip the part were the pressure increases and just go to the point where the oil is relieved. That doesn't mean the pump curve and the system curve doesn't exist. You just skipped over it.

The pressure compensated pump is similar to a fixed displacement pump except the displacement is not fixed. The pump curve at any displacement is like that of a fixed displacement pump where the curve is almost a horizontal line. In the compensator or proportional band the displacement is variable. This moves the pump curve up and down. What makes this trickier is that the system curve is often changing which is why a pressure compensate pump is used. The pressure compensate pump adapts.

Why the resistance to what is obvious? If the pump doesn't add energy to the system there will be no energy to dissipate on the form of work or heat. The pump has equal say in what the pressure will be. In some cases the pump is driven by a motor that runs at only one RPM so there are only two choices. Off, no pressure, or at the flow and pressure determined by the pump and system curves at the operating RPM. The trick to finding steady state operation is to calculate the pump curve and system curves and determine where the two intersect.

My first hydraulic system used vane pumps and ran at three speeds, Off, low speed and high speed. The system curve never changed. The pump speeds controlled everything.

JDK said...

I read your last comment a couple of times; again, thanks.

I agree with most of it.

However,in the 3rd paragragh you say "This moves the pump curve up and down". I disagree. At a given pump RPM and compensator setting, the pump curve is fixed.

"What makes this trickier is that the system curve is often changing which is why a pressure compensate pump is used. The pressure compensate pump adapts." I agree 100%

"The trick to finding steady state operation is to calculate the pump curve and system curves and determine where the two intersect." The pump curve doesn't need to be calculated; it is in the manufacturer's data. You described the pressure compensated part of the curve perfectly. In addition, the horizontal part of the curve moves up and down with pump RPM. The vertical part of the curve moves left and right with the compensator setting. At a fixed RPM and fixed compensator setting, the pump curve is fixed.

Every comment about the system curve I agree with.

You mention my "resistance to the obvious"; there is none. I am trying to learn from you as well.

pnachtwey said...

Oops, I screwed up. I got my directions mess up.
A fixed displacement pump had an almost vertical pump curve, not horizontal. The relief valve response is part of the system curve.
Hopefully the relief valve is activated before the pump power is exceeded.

I screwed up the compensator pump section too. The compensator pump curve slopes down to the right depending on the rate flow and the pressure drop required to get that flow.