Pumps, flow and restrictions.....

[David Devuono]

Good to know who the experts are. Not sure to start a new thread or hyjack this one? I am making a tracked vehicle. I am not interested in using a dual variable pump (Zero turn pump) because I don't want to build a separate system for the rest of tractor hydraulics so I am going with a constant output pump. Splitting the flow equally to two motor spools. Ahead of one of the spools I will either have a priority flow control or a diverter valve to a 3rd 3 spool cylinder valve. The motor spools are 180 degree rotating spools and not push\pull spools. As such the work ports, I believe are blocked when in the (non-detent) center position. There will be an external pressure relief across the work ports in each motor circuit. I am planning to start out with a main system relief at about 2000 psi. My question is, what should i set the motor circuit relief vales at? Should they also be 2000 psi or can I go to 2450 psi (max constant motor pressure?) The main purpose of these valves is to protect the system from any shock loads the motors might put on the system from too sudden stops, steep down hill etc.

thanks

 

[wdchyd]

AKKAMAN,

Here's the video I promised, sorry for the delay.....:ashamed:
.Kubota Flow/Load Meter Test - Tractor Videos - Kubota, John Deere, New Holland and More....

Hope this keeps the discussion in the right direction of "Pumps, flow & restrictions"......(Man, I Hate restrictions!!!)

 

[IceStationZebra]

Pumps, flow and restrictions.....

Can some one now tell me how a smaller fitting, QC, bent hose, smaller valve etc (any kind of restriction), on the PRESSURE SIDE of the pump, can decrease flow?????:laughing::thumbsup:

AKKAMAAN, do I need to get Peter Nachtwey over here?

ISZ

 

[AKKAMAAN]

AKKAMAAN, do I need to get Peter Nachtwey over here?

ISZ

PN is one of the most brilliant minds in motion control..
But I think he prefere to put most of his effots into the other end of the motion control science....
Dont even think I could earn a seat in his class room....LOL....

 

[AKKAMAAN]

AKKAMAN,

Here's the video I promised, sorry for the delay.....:ashamed:
.Kubota Flow/Load Meter Test - Tractor Videos - Kubota, John Deere, New Holland and More....

Hope this keeps the discussion in the right direction of "Pumps, flow & restrictions"......(Man, I Hate restrictions!!!)

Thank you wdchyd! Great video shot, showing exact what we wanted to see....no reduced flow when we restrict with the "valve"....only when PRV start by passing we get reduced flow through restriction....

 

[wdchyd]

Thank you wdchyd! Great video shot, showing exact what we wanted to see....no reduced flow when we restrict with the "valve"....only when PRV start by passing we get reduced flow through restriction....

If I plumbed directly downstream of the pump you would have seen the effect of direct flow from the pump....GPM's would have been right up there with no PRV reduction....

Glad you liked it....A video first for me....

 

[AKKAMAAN]

If I plumbed directly downstream of the pump you would have seen the effect of direct flow from the pump....GPM's would have been right up there with no PRV reduction....

Glad you liked it....A video first for me....

Does not matter where you make the test, it is still the pump flow.....

You do not want to do this test without a PRV.........believe me!! on a Webster flow meter, there is a brass washer that breaks when you get to 6000psi, VERY DANGEROUS...that valve is not "proportional" and will build 2000psi just by a few degrees, when valve is 75 % closed....:shocked:

 

[wdchyd]

Does not matter where you make the test, it is still the pump flow.....

You do not want to do this test without a PRV.........believe me!! on a Webster flow meter, there is a brass washer that breaks when you get to 6000psi, VERY DANGEROUS...that valve is not "proportional" and will build 2000psi just by a few degrees, when valve is 75 % closed....:shocked:

You got that right!!!....we're using one tomorrow to test 80 Gpm, 2100 psi on a Dennison Vane Pump.....

 

[Jerry/MT]

Pumps, flow and restrictions.....

The pumps we use in hydrostatic systems are so called POSITIVE PUMPS, the are designed to force almost the entire flow FORWARD. A NEGATIVE PUMP are designed to allow flow going BACKWARD as well as forward. From now I only talk positive pumps.

There are pumps with FIXED displacement and VARIABLE displacement. A variable displacement pump can, internally, change its displacement per revolution. That means flow can be changed with out changing the pump shaft RPM. Of course it still takes RPM's to turn the pump around..
From now I only talk fixed displacement pumps.

A fixed displacement pump needs a protection from extreme high pressure. WE use a Pressure Relief Valve, PRV, to BY-PASS flow when pressure exceed the settings on the PRV. Pump flow will still stay intact, the same. Oil just take a different path back to reservoir.

Can some one now tell me how a smaller fitting, QC, bent hose, smaller valve etc (any kind of restriction), on the PRESSURE SIDE of the pump, can decrease flow?????:laughing::thumbsup:

Anything that increases the pressure loss downstream of the pump will cause the pressure of the pump to rise. Consider a water pump with a gate valve attached to the outlet. If the valve is completely closed, the pump is deadheaded and achieves its maximum presure. (Of course that's a useless situation since with no flow, why have a pump?) As you open the gate valve the pressure drops and the flow increases. Conversely as you close the valve the flow will decrease and the pressure will rise. Downstream pressure losses act like that gate valve; they reduce the effective area that the pump discharges into. A bend, an orifice, a fitting, the friction of the hose or tube, etc all cause pressure losses, i.e. they are like closing down the gate valve and they force the pump to operate at a higher pressure and less delivered flow.

Even a so called constant displacement pump will show some flow change with downstream restriction, due to internal leakage as the pressure rises. Is it zero? In general No. So the discussion centers on where is the pump designed to operate. If we operate it at a fixed speed and at some range away from zero flow, where the pressure-flow characteristc is absolutely vertical ( i.e. dW/dp=0] n=constant) than we can say that downstream restrictions will not affect the flow rate.

By the way I have seen high presure ratio fans that at high speed exhibit vertical charactersitics similar to constant displacement pumps so this phenomenon is not limited to gear pumps etc. These fans are, however, unstable above certain pressure ratios due to certain aerodynamic considerations.

thanks for the thought provoking discussion.

 

[AKKAMAAN]

.
......Consider a water pump with a gate valve attached to the outlet. If the valve is completely closed, the pump is deadheaded and achieves its maximum presure.......

I think it confuses more than help, to refer to the waterpump system, which actually acts more like a constant pressure system....this is actually the reason why people misunderstand a constant flow system.....

Even a so called constant displacement pump will show some flow change with downstream restriction, due to internal leakage as the pressure rises. Is it zero? In general No. So the discussion centers on where is the pump designed to operate. If we operate it at a fixed speed and at some range away from zero flow, where the pressure-flow characteristc is absolutely vertical ( i.e. dW/dp=0] n=constant) than we can say that downstream restrictions will not affect the flow rate.

If you read the entire thread, you can see that we already have discussed the impact of internal leakage. My point is to proof, that a constant flow can not be reduced without out diversion. Internal leakage is a form of diversion too. wdchyd's video clearly shows that flow is constant through restrictor, until PRV opens.

A constant pressure system will react different on a flow restrictor. No flow diversion needed, but indeed "pressure diversion"....but thats another thread......more to come about that.....

thanks for the thought provoking discussion.

You are welcome!

 

[mmurphy]

Jerry, a centrifical water pump will only lift water to a "Head" (height) based on 2 things. Diameter of impellar, and RPM. (Your arm throwing something straight up in the air to apoint it stops and comes back down at you.)
Most big flowing impellar type pumps are started "Dead Headed" against a valve. That does a couple of things also. Reduces shock on system from a major change in flow, and takes less amperage to start motor.
Some now might be thinking, How can dead heading a pump (centrifical) make it less amps to start? The easiest way for me to tell you is if pump is pumping 100 galons/second (small pump) 8.34 lbs/gal would= 834 lbs/sec. Deadhead a pump will just turn the water in the impellar causing no lift. Do that with a positive displacement pump, things will fly apart every where.
I think where some of this is confusing some people at is, they put a restriction in and they see it slow a cylinder down. Now they say AKKAMAAN has got to have this wrong.
Most operations with a cylinder is feathered in or does not see the full pressure build up on both sides of orifice. (untill end of travel) So there for it is hard to understand the whole concept of the thread if they are thinking of how they are running a hydraulic cylinder with orifice in place.

 

[Egon]

Diameter of impellar, and RPM.

Wouldn't this just affect the mass and velocity of fluid moved [discharged]?

 

[mmurphy]

Wouldn't this just affect the mass and velocity of fluid moved [discharged]?

They are interdependant of each other. And yes size of pump comes in to the picture also. I was thinking of a one pump size at the time of posting.

 

[Jerry/MT]

Jerry, a centrifical water pump will only lift water to a "Head" (height) based on 2 things. Diameter of impellar, and RPM. (Your arm throwing something straight up in the air to apoint it stops and comes back down at you.)
Most big flowing impellar type pumps are started "Dead Headed" against a valve. That does a couple of things also. Reduces shock on system from a major change in flow, and takes less amperage to start motor.
Some now might be thinking, How can dead heading a pump (centrifical) make it less amps to start? The easiest way for me to tell you is if pump is pumping 100 galons/second (small pump) 8.34 lbs/gal would= 834 lbs/sec. Deadhead a pump will just turn the water in the impellar causing no lift. Do that with a positive displacement pump, things will fly apart every where.
I think where some of this is confusing some people at is, they put a restriction in and they see it slow a cylinder down. Now they say AKKAMAAN has got to have this wrong.
Most operations with a cylinder is feathered in or does not see the full pressure build up on both sides of orifice. (untill end of travel) So there for it is hard to understand the whole concept of the thread if they are thinking of how they are running a hydraulic cylinder with orifice in place.

Thanks for your response. I guess I should have read the whole thread. I was responding to the first post.

Respectfully, I understand how a centrifugal pumps starts as I start them everyday on my irrigation system and I'm familiar with the performance aspects of centrifugal pumps.(When deadheaded between line changes these pumps produce about 65psi. Depending on the irrigation configuration they run up to 500+ gpm and even higher when filling lines.)

I take exception to your statement that only the centrifugal pump impeller diameter and the rpm (basically tip speed) determine the pump head. The other requirement is the discharge restriction. As an example in my irrigation system when fillling lines, the disharge pressure as measured in the discharge manifold is zero and the line flow is basically open channel since the lines are not running full. As the system fills, the sprinkler nozzles and drains set the pressure and it rises above zero psi. When the drains close, the nozzles are controlling the pump head ( absent a leak that occurs occaisonally) and it rises to it steady state value that's determined by the line losses and the aggregate nozzle area.

I have to admit that I don't have much practical experience with gear pumps or other positive displacements pumps so I guess I'm just showing my ignorance in that area. None the less I enjoyed the discussion and most importantly I learned something new.

 

[Jerry/MT]

I think it confuses more than help, to refer to the waterpump system, which actually acts more like a constant pressure system....this is actually the reason why people misunderstand a constant flow system.....

If you read the entire thread, you can see that we already have discussed the impact of internal leakage. My point is to proof, that a constant flow can not be reduced without out diversion. Internal leakage is a form of diversion too. wdchyd's video clearly shows that flow is constant through restrictor, until PRV opens.

A constant pressure system will react different on a flow restrictor. No flow diversion needed, but indeed "pressure diversion"....but thats another thread......more to come about that.....



You are welcome!

I guess I should have read the entire thread but I didn't.

I have to admit that I don't have much practical experience with gear pumps or other positive displacements pumps so I guess I'm just showing my ignorance in that area. None the less I enjoyed the discussion and most importantly, I learned something new.

 

[Egon]

I take exception to your statement that only the centrifugal pump impeller diameter and the rpm (basically tip speed) determine the pump head.

Don't forget the number of stages in determining head.