controlling presure with VFD

[Peter Nachtwey]

A lot will depend on how sensitive is the pressure rate of change with respect to flow. If the volume of fluid under compression is small then pressure can change too quickly for PLCs. dP/dt=β*Q(t)/V
dP/dt is the rate of pressure change
β is the bulk modulus of the fluid
Q(t) is the flow as a function of time
V is the volume fluid under compression
Another way of look at this is
ΔP=β*ΔV/V
If the volume under compression is 100 cubic inches and 1 cubic inch of fluid is added then the pressure will increase 2000 psi for oil and 3000 psi for water. This means the pump must be able to add fluid at a very controlled and small rates.

 

[dougpdf]

That might explain some of my problem. Even if I set a fixed speed of approx 100 rpm, the pressure starts low and gradually drifts up.
Most systems of this type are using an internal gear pump. I'm using a vane. I'm compressing aprox .03 cu in of hydraulic fluid and trying to maintain approx 500 psi +/- 10.
This is an injection molding press, so for most applications, it injects with 10 gpm or so at 1500 psi and I want it to drop down to 500 psi at a given setpoint to 500 psi utilizing the pump speed to control that pressure. I can do that, it just starts drops below 500 and gradually ramps beyond 500, and fluctuates up and down..all this "hold pressure" is happening over a 6 sec period. I want it to drop to 500 and stay at 500 and/or have some capability to have multiple setpoints during that 6 second time frame.
I still thought I would be able to tune it out with my plc.
My skills are somewhat limited.

 

[dougpdf]

sorry....that really doesn't read very well.
lol

 

[dougpdf]

and that's .03 gallons I'm compressing, not cu in...duh.
It's late

 

[DickDV]

Doug, look in the Instruction Manual for the list of standard macros. One of them should be a PID macro. Read the setup instructions for that macro and you should be ok. Incidently, the ACH series of drives are intended to be HVAC drives. The industrial drives were ACS___. The ACH and ACS400's and 500's were the same hardware and software but that is not the case with the ACH and ACS550's. Both the 400's and 500's are well past their useful lifespans and I would advise budgeting for a newer replacement right away. Average life of VFD capacitors in those days ranged around 11 years and those units are likely in the 20 year old range.

 

[DickDV]

jraef, I'm starting to understand how you are so familiar with the ABB product! If I ever get to Northern California again, I'd like to share a cup of coffee or something similar with you.

 

[dougpdf]

These drives were no doubt past their prime when I picked them up used of ebay 4 or 5 years back as an experiment. They've been running 24/5 since then. Ramping up and down continually every 40 seconds. Real work horses.
I did buy 30hp units for 20 hp motors, so that probably helps their lifespan.

I'm curious if a servo drive would work on a conventional motor? The Delta drives from Asia are pretty darn economical and very friendly for an inj molder

 

[Peter Nachtwey]

Whoa guys.
I see the pump can pump up to 10 GPM. If it can pump even more then 1% will be a bigger number.
10 gpm=38.5 cubic inches per second
0.03 gal = 6.93 cubic inches under compression.
If you can control the flow rate to with 1% that would be 0.385 cubic inches per second.
Using the formula above
dP/dt=β*Q(t)/V
200,000*0.385/6.93=11111.1111 psi/sec.
Even if the scan time is 0.001 seconds then the pressure can easily change 11.111 psi per second.
A hydraulic motion controller would need to have a scan time of 0.5 milliseconds.
Another issue is the pump ripple. The VFD may turn at a constant rate but the pump doesn't create flow smoothly as it turns. There are pressure pulses.

 

[milldrone]

Most systems of this type are using an internal gear pump. I'm using a vane.

Many vane pumps ( not all ) depend on the centrifugal forces on the vanes to get proper sealing. I'm suspecting that the low speeds are not achieving a good seal all of the time.

 

[dougpdf]

Shoot, it's so close using my lowly pid loop in the plc, I hoped a process/motion controller would narrow the gap to an acceptable level.
Sounds like I've got a number of things working against me, in particular pump ripple.
The modern systems are using a servo motor/drive and internal gear pump.
The holding pressure we're discussing the critical element in my particular process. Maybe a proportional valve is in order.

 

[dougpdf]

0.03 gal = 6.93 cubic inches under compression.

Peter, I gave you bad data. While holding the full vol of the cylinder is being pressurized. Not the fluid in the front.
We're pressurizing .5 gal, not .03 gal

 

[Edmhydraulics]

I will chime in here as well from a hydraulic standpoint, most vane pumps are designed to operate at a minimum speed of around 300 rpm to keep the vanes engaged on the ring. Plus pressure ripples might be causing you some issues that are compounding the low speed issue, there might be some dynamics inside the pump that are magnifying your problems. If I were you, I would move away from the vane pump and go with a good quality gear pump. If you can't do that then I would keep your speed as high as possible and perhaps use a relief off a branch circuit with a solenoid valve that only opens when you need low pressure, the relief would maintain pressure at the low set point and when higher pressure is requested the solenoid valve would close and pressure would be controlled by the pump speed and demand.

But doing something like this is not the right way and is only putting lipstick on a pig. Best solution is a gear pump and a motion controller by far.

 

[Mickey]

Why not run the pump at a constant speed and control the pressure with a recycle valve, under PID control. Just thinking out loud.

 

[Peter Nachtwey]

Why not run the pump at a constant speed and control the pressure with a recycle valve, under PID control. Just thinking out loud.

Wasteful. Everyone wants to say energy now.
Edmhydraulics has the right idea. Servo motors and gear pumps. Controlling pressure requires very little flow but a motor that can apply torque at low speeds.

 

[Edmhydraulics]

Thank you Peter, nice to get kudos from the master of hydraulic motion control. I think that servo driven fixed displacement pumps are the future of the industry from an energy conservation aspect.