PID Control of machine speed forward and reverse

[Peter Nachtwey]

Will a motion controller accept a 4-20ma position input?

Yes and the PLC can issue commands over the Ethernet/IP.
You should have provided all this information all the information above at the start.
Your system is a p!$$ poor design.
What kind of machine is it? I know about different sawmill machines.

Yes, you need a motion controller for hydraulic servo control but that won't help without some other changes.
You also need a true servo quality proportional valve. One that doesn't dump oil to the tank so there is no braking.
What supplies the hydraulic oil? Is it a constant oil source like a constant current source? Accumulator(s) may be needed.
You won't be able to make significant improvements with what you have now.

What you say about reading positions every 200ms but updating the speed faster doesn't make sense. In any case the updates are too slow.

I know about sawmills I have operated just about every type of machine they have installing new controls and making upgrades. I used to work for USNR in the 1980s. I have spent many hours in them. Delta Motion makes most of the hydraulic servo controllers used in sawmill in North America. You have an odd ball and poorly designed system.

I am a member of the International Fluid Power Hall of Fame. I know about hydraulic system design.

Now what are you willing to change?

 

[Robb B]

To elaborate on the drive system some more, the system is hydraulically driven. Think of it like a hydraulic setworks operating off a +/- 10V proportional valve with a 4-20ma input position signal. The hydraulic cylinder is 55 feet long and 4 inches diameter. To facilitate moving this much hydraulic fluid, the big spool when centered is open to the tank so the spool bypass allowing the 20,000 lb mass to continue coasting.

The position update time is currently 200 ms due to a very noisy 4-20ma signal which is being filtered with a 200ms running average. I'm working through how to clean up the noisy position signal at this very moment.

The velocity is being calculated every 50 ms.

How do you know it is noisy? Some variation of the milliamp signal is normal, but you should be seeing values to at least 1 decimal place being very steady. Is it possible that the position feedback is actually erratic? There could be movement in a coupling going to an encoder, maybe a cylinder sensor is tracking, there are many ways that "noise" could be error in measurement, rather than a randomly changing mA signal on a DC loop. These have been used for decades for good reason, very immune to "noise".

 

[Peter Nachtwey]

H

Hmm, such insignificant additions...
(What kept you from presenting this information in your first post?)

Agreed

I'd be curious to see:
hydraulic diagram (with explanation of how the central spool works)
control process curves (indicating PID controller settings)

yes,

At the moment, I continue to believe that control is carried out in torque mode, accordingly P-only (at least for initial experiments) is quite sufficient (this will allow us to ignore the possible problem of time scaling)

Hydraulic system work in velocity mode. The OP has a screwy valve that isn't linear. When the valve is sent a zero voltage the flow through the valve should be near zero but the OP stated above that oil is allowed to flow freely to the tanks so there is no opposing stopping force.

Is my assumption correct that the main difficulty is the noisy velocity "signal" (which is calculated by the discrete derivative of position)?

That isn't his main problem. The hydraulics are poorly designed. There are no cheap easy fixes when the hydraulic and mechanic design is a kluge.
The OPs system is basically a 20000 lb mass between two springs. The springs are the volumes of oil on either side of the piston A low natural frequency limits how fast one can accelerate. That is why it takes 4 seconds to ramp up or ramp down.
There isn't even any guarantee that the hydraulic system pressure stays relatively constant. There should be an accumulator to provide oil when accelerating because the HPU ( hydraulic power unit ) takes time to rev up.

I agree the with RClem that using a servo grade valve is a good start. I know that a Delta Motion RMC75 would help a lot. The PLC is sampling too slowly and updating positions and velocities too slowly. The PLC should be issuing commands to the motion controller over Ethernet/IP and the motion controller should be able move the actuator to the desired position within the limits of the feedback. I calculate the resolution to be about 0.023 inches per count assuming the 4-20 ma will provide about 26000 counts with a 16 bit ADC. The motion controller should be able to get with a fraction of an inch if the rest of the system is tight.

The OP said this was for a sawmill. I have spent a lot of time in sawmills automating and upgrading them with the Delta Motion RMC controller. I have operated just about every type of equipment in the sawmills. I have never seen a system as screwy as the OPs.

I find it hard to believe the hydraulic cylinder is 50 ft long. It probably pulls a set of pulleys with a cable wrapped around it that multiplies the distance the hydraulic actuators move. This would allow the system to move logs 50 ft but it magnifies the load seen that the hydraulic cylinder.

 

[drbitboy]

The OP said this was for a sawmill.

Not that it matters, but are you sure of this? Please identify the post where you saw this, or perhaps you received a PM?

Also, it is still not clear to me whether the ±10V signal controls acceleration or velocity: it seems like it would be velocity, but when the signal is at 0V the machine (carriage?) coasts, which means acceleration is 0 at that point, and if the signal was controlling velocity that velocity would have been very small, i.e. near 0, just before the signal reached 0. I am not a member of the International Fluid Hall of Fame, I am very confused, but I do have my popcorn.

 

[Peter Nachtwey]

Not that it matters, but are you sure of this? Please identify the post where you saw this, or perhaps you received a PM?

Also, it is still not clear to me whether the ±10V signal controls acceleration or velocity: it seems like it would be velocity, but when the signal is at 0V the machine (carriage?) coasts, which means acceleration is 0 at that point, and if the signal was controlling velocity that velocity would have been very small, i.e. near 0, just before the signal reached 0. I am not a member of the International Fluid Hall of Fame, I am very confused, but I do have my popcorn.

The OP didn't say sawmill but he did say setworks which is what they call hydraulic positioning systems in sawmill. I don't know of any other mill that uses the term setworks.

dritboy, the OP's transfer function is wrong if he is truly moving using a hydraulic cylinder. There the open loop transfer function should have a natural frequency and a damping factor in it. It does look like control voltage is controlling acceleration but in a hydraulic system the control signal controls the spool position that controls flow or speed. The OP is masking the second order effects by accelerating so slow and having an open center valve when it is at 0% so there is no braking effect.

I do know about hydraulic motion control. This is a p!$$ poor design. The natural frequency of the system should be calculated. I have the formulas on the Power and Motion website.

 

[drbitboy]

The OP didn't say sawmill but he did say setworks which is what they call hydraulic positioning systems in sawmill. I don't know of any other mill that uses the term setworks.

dritboy, the OP's transfer function is wrong if he is truly moving using a hydraulic cylinder. There the open loop transfer function should have a natural frequency and a damping factor in it. It does look like control voltage is controlling acceleration but in a hydraulic system the control signal controls the spool position that controls flow or speed. The OP is masking the second order effects by accelerating so slow and having an open center valve when it is at 0% so there is no braking effect.

I do know about hydraulic motion control. This is a p!$$ poor design. The natural frequency of the system should be calculated. I have the formulas on the Power and Motion website.

@Peter Nachtwey: Thanks for the clarifications!

 

[Jasondelane]

Sorry guys. I suck at hydraulics. I freely admit it. It isn't really a setworks in the true sense of the term.

@Peter Nachtwey are you familiar with Tyrone-Berry Drives?

 

[Peter Nachtwey]

Sorry guys. I suck at hydraulics. I freely admit it. It isn't really a setworks in the true sense of the term.

Few people are really understand hydraulic servo control.
Your crime is not being up front with what you are trying to control.
Knowing the machine type and manufacturer would have helped a lot.
So would a link to the type of machine. The OEMS usually have pictures and videos of their machines.

Are you sure the cylinder is 55 ft long? I have never seen a cylinder that long. For a 4 inch diameter cylinder, the cylinder rod would be about 2.75 inches in diameter which is too thin for such a long rod. It could be bent too easily.
When designing the hydraulic systems there is a limit to how much load a cylinder rod can be as a function of the load and diameter.

@Peter Nachtwey are you familiar with Tyrone-Berry Drives?

No, but I know our controller has probably controlled some of them.
There are many types/models of Tyone-Berry drives.
There are many ways of moving logs through saws.

Some systems are very old and not well designed. Most of the end feed system I have controller used hydraulic or electric motors to move logs on a feed chain.

Just so everyone knows, Delta Motion makes most of the hydraulic servo control for sawmill system. USNR ( USA ) and the Bid Group ( Canada ) are the two major OEM but many of the smaller OEMs use our controllers too. When walking around a sawmill show about 4/5th of those companies making hydraulic servo systems use Delta Motion RMCs or Rockwell motion modules. The Delta Motions sales are about 30% to the sawmill industry. What we didn't sell, Rockwell did selling the HYD02 and M02AS which we made for Rockwell but are now being discontinued. I have spent a lot of time in sawmills and have operated about every type of machine that is used in them.