Pump test stand and load simulation

[jhiguy]

Hello,

First, I have had one post deleted by a moderator so apologies in advance if I am unintentionally doing something incorrect with this post. I read all of the posting tips and couldn't find a reason why my last post would have been deleted, but again, I hope I am not breaking some rule by asking for assistance here.

We designed an electrically driven hydraulic pump (operating 500 psi shut off 650 psi, 0.8 gpm) and I want to build an automated test stand. We did build a test stand that was manual for some initial testing but now I want to run about twelve thousand cycles along with some other parameters. In the manual system the guy who set it up used a manual flow control valve and gauge to set circuit operating pressure to 500 psi by restricting flow to simulate pressures that we typically see in our circuit during actuation.

My first question would be is if it is okay to use a flow control valve to adjust pressure like this? My understanding is flow control is speed control, not pressure control?

I am going to set up a control and daq system, chiller/filter, pressure transducers, eddy current for rpm, amp draw, temp, relief valve and I am not sure on the best way to simulate the load in the circuit and if there is a better way than using flow control valve as in the example above?

I have looked for an electronically controlled proportional flow control valve to do the same thing as above and I haven't found something that has accurate control down to the 0-1gpm range. I spoke to Brand Hydraulics (U.S.A.) and their EFC proportional control valve starts at 1 gpm and will not work in this application.

Is my approach correct, or is there a more "standard" way that one would adjust and control variable pressure in the test circuit?

Any help is greatly appreciated. Again, my apologies if this is not the right place to ask my questions or they are somehow inappropriate.
Thank you,
jhiguy (obviously not a PE)

 

[hydtools]

The manual flow control valve may be simply a needle valve used to create resistance to flow and not to control flow. Is your pump a positive displacement pump? If so, a flow control must somehow bypass or dump excess flow. If you want to create pressure you must provide resistance to flow rather than trying to control flow.

Ted

 

[PNachtwey]

The application appears to be legitimate enough. I have no idea why a topic on this should be deleted. Student questions are not allowed so if you do a poor job of supplying information, like a student, the topic could be deleted.

Hydtools has a good question. What kind of pump are you testing?
I have questions too. What is bylining? Pressure or pump RMP, current, what?

What you want to do is too have the pump outflow into a tank of a size yet to be determined. The bigger the tank the slower the pressure will increase or decrease so the bigger the tank the less sensitive the system will be to changes in in-flow and out-flow.
You want to have a good quality valve on the outlet of the tank. I should have the ability to flow 1 gpm at the pressures you are testing at.

There is a key equation for your application.
dP/dp = β*(Qin(t)-Qout(t)/Volume
P is pressure so dP/dt is the rate of change of pressure.
β is the bulk modulus of oil or how compressible it is. 160,000 psi to 200,000 psi is a typical value.
Qin(t) is the in-flow from the pump.
Out(t) is the out-flow through the valve.
Volume is the volume of the tank.

Now you can see larger the volume of the tank the smaller dP/dt will be for a mismatch in flow between Qin and Qout.
To maintain a pressure in the tank or outflow of the pump the out-flow must be equal to the in-flow.
To reduce pressure the out-flow only needs to be more than the in-flow
To increase pressure the in-flow only needs to be more than the out-flow.
This should be obvious but there is a little more to it than "pressure is resistance to flow"

I have a solution to your DAQ needs.
Get a RMC75E AA2 and possibly more I/O and NI LabView. The RMC and LabView communicate over Ethernet. All the testing can be automated and recorded.
You should be able to draw nice plots for each test.
This setup will allow you to control the system and gather data at 1ms increments or faster.
Obviously I/we and our distributors have done this many times before.

Monitoring torque is tricky because the voltage and current is AC so deriving a value for torque should be done by a strain gage torque sensor not a current sensor. There is often too much noise on the voltage and current. Filtering causes a lag in the signal.

























Peter Nachtwey
Delta Computer Systems

http://www.deltamotion.com

http://forum.deltamotion.com/

 

[jhiguy]

Thanks hydtools,

That is correct, the manual flow control valve is the needle valve type and yes it is used to control/restrict flow. Not sure I understand the difference between controlling flow and restricting flow, are you not controlling the flow by restricting it? The manual valve is similar to this Link which works to increase pressure in the system by restricting flow. What I would like is an electrically controlled version that has enough sensitivity to work in the 0 to 0.8 GPM range.

Pump:
Positive displacement
4000 rpm
no load ~5.5 amps
with load varies between 25-30 amps
24VDC
displacement 0.0469 C.I.R.(0.812 GPM)
tank will be 5 gallon which is more than five times gpm and the normal rule of thumb of 3x, figuring the heavy cycling during testing.
No accumulator in system and non-pressurized reservoir/tank

Thanks PNachtwey,
I am using Raspberri Pi with MyOpenLab to build the control/DAQ system. Someday maybe we will purchase NI's Labview for company use but for this instance it's a bit of a stretch. The DC ripple detector is for RPM, I think it may work okay since I have a high end 70 amp rated DC power supply along with the pump motor being DC, sorry for not providing that information earlier.

From the discussion, which I really appreciate, it sounds like restricting the flow is acceptable (obviously to a point). Should I open another thread regarding my search for a type of needle? valve that can be controlled electronically and accurately in the 0-0.8 GPM range?

Thank you again, very much appreciated.
Garry

 

[hydtools]

I did not say restrict flow. I said create resistance to flow, flow rate remains constant as long as pump speed remains constant at the demanded pressure caused by the resistance to the flow.
Maybe consider an adjustable relief valve that will handle your flow rate.

Ted

 

[jhiguy]

hydtools

My mistake, thanks for your help.
Garry