Help on PID control

bb76 · Feb 25, 2022

If the system has been in for 16 years I'm assuming it has ran correctly at some point and it now isn't? If that is the case, what has changed?

Is this a new SP that is above the capacity of the system?
Your screen shots show some kind of auto/man/step set up. What is the state of those conditions when you are having the problem?
Does the prop valve fully travel?
Is the hydraulic system underfilled?
You found the engineer who installed it, what does he have to say about it?

Peter Nachtwey · Feb 25, 2022

Finally, it is a hydraulic press.
What has been written doesn't make sense at all.
You guys are too worried about the PID and not how a press works.

drbitboy · Feb 25, 2022

I agree with @Peter Nachtwey: as I said in my first response, the problem is not in the PID.

What is the pressure of the hydraulic fluid at the discharge of the pump when the press measurement is hovering at 68PSI?
What is the pressure of the hydraulic fluid at the press when the press measurement is hovering at 68PSI?
Is the valve open a little bit, or perhaps leaking a bit, when the press measurement is hovering at 68PSI?

Also, I think I may understand the process now; let me know if the following is correct.

There are three, maybe four, relevant elements in the system. In order from highest hydraulic fluid pressure (HFP) to lowest HFP, the are the following:

The discharge of the hydraulic pump has the highest HFP in the system
Next is the control valve, across which there is a drop in HFP
1. The magnitude of the drop in HFP here is a function of valve position, flow rate, and possibly downstream (down-pressure) elements
Next is the volume of hydraulic fluid above the press itself, that effects the pressing of the press onto the product
Finally a bleed(?) valve or orifice or other restriction, across which there is a large drop in HFP, and through which hydraulic fluid flows back to the pump inlet or pump intake reservoir.

There are obviously hydraulic lines in between those elements, across which there is some (I assume minimal) pressure drop.

So a simple, first-order, steady-state model of the system is



HFPpress = (HFPdischarge*Rbleed + HFPintake*Rcv) / (Rcv+Rbleed)

where
- HFPpress is the HFP above the press
- HFPdischarge is the HFP at the pump discharge
- HFPintake is the HFP at the pump intake
- Rbleed is the resistance of the bleed device, the reciprocal of its flow coefficient
- Rcv is the resistance of the control valve, the reciprocal of its flow coefficient.

Am I close? Am I missing anything significant?

Caveats

The actual system is obviously far more complex i.e. non-linear, and not always at steady-state. But if we are using a PID, then we are treating it as more or less linear in the region of operation, and the PID's job is to get the system to a target steady state.
Again I am ignoring pressure drops through the hydraulic lines, or at least lumping them into the Rbleed and Rcv model parameters.
This model ignores compressibility (half a percent per kPSI?), so there is an In-Out=Accumulation part of the model that I am ignoring, where Accumulation may be linear, or at least monotonic, with HFPpress, but that would only affect the dynamics because dAccumulation/dt is zero at steady state.
A fifth element would be the intake (reservoir) of the pump.
The last element, the bleed, may not be present, or maybe it is "leakage" to a sump
I understand fluid statics and dynamics fairly well but am willing to learn more, and I don't know a lot about the details of hydraulic systems.

Taylor Turner · Feb 25, 2022

40mpg said:
Hello all,

The Set Point is 71 psi

This loop has been used for over a decade. The problem is the PV never reaches the Set Point of 71, it hovers at 68-69 and cuts off.

Can you approach a set point higher than 71 or is that the limit of your process? Is your output saturating?

If open loop voltages are not getting you to or past your desired set point, your pump compensation needs to be adjusted. Seals leak, oil heats up. Lots of factors that allow upper limits one moment, then just not enough the next. x1.2 on the process.

40mpg · Feb 25, 2022

drbitboy said:
I agree with @Peter Nachtwey: as I said in my first response, the problem is not in the PID.

What is the pressure of the hydraulic fluid at the discharge of the pump when the press measurement is hovering at 68PSI?

What is the pressure of the hydraulic fluid at the press when the press measurement is hovering at 68PSI?

Is the valve open a little bit, or perhaps leaking a bit, when the press measurement is hovering at 68PSI?

Also, I think I may understand the process now; let me know if the following is correct.

There are three, maybe four, relevant elements in the system. In order from highest hydraulic fluid pressure (HFP) to lowest HFP, the are the following:

The discharge of the hydraulic pump has the highest HFP in the system

Next is the control valve, across which there is a drop in HFP

The magnitude of the drop in HFP here is a function of valve position, flow rate, and possibly downstream (down-pressure) elements

Next is the volume of hydraulic fluid above the press itself, that effects the pressing of the press onto the product

Finally a bleed(?) valve or orifice or other restriction, across which there is a large drop in HFP, and through which hydraulic fluid flows back to the pump inlet or pump intake reservoir.

There are obviously hydraulic lines in between those elements, across which there is some (I assume minimal) pressure drop.

So a simple, first-order, steady-state model of the system is

HFPpress = (HFPdischarge*Rbleed + HFPintake*Rcv) / (Rcv+Rbleed)

where
- HFPpress is the HFP above the press
- HFPdischarge is the HFP at the pump discharge
- HFPintake is the HFP at the pump intake
- Rbleed is the resistance of the bleed device, the reciprocal of its flow coefficient
- Rcv is the resistance of the control valve, the reciprocal of its flow coefficient.

Am I close? Am I missing anything significant?

Caveats

The actual system is obviously far more complex i.e. non-linear, and not always at steady-state. But if we are using a PID, then we are treating it as more or less linear in the region of operation, and the PID's job is to get the system to a target steady state.

Again I am ignoring pressure drops through the hydraulic lines, or at least lumping them into the Rbleed and Rcv model parameters.

This model ignores compressibility (half a percent per kPSI?), so there is an In-Out=Accumulation part of the model that I am ignoring, where Accumulation may be linear, or at least monotonic, with HFPpress, but that would only affect the dynamics because dAccumulation/dt is zero at steady state.

A fifth element would be the intake (reservoir) of the pump.

The last element, the bleed, may not be present, or maybe it is "leakage" to a sump

I understand fluid statics and dynamics fairly well but am willing to learn more, and I don't know a lot about the details of hydraulic systems.

What is the pressure of the hydraulic fluid at the discharge of the pump when the press measurement is hovering at 68PSI? I have not look but planned to look (the pump line at the pump's outlet says max 3000 psi)

What is the pressure of the hydraulic fluid at the press when the press measurement is hovering at 68PSI? The pressure transducer located at the top of the press reads 1777 psi

Is the valve open a little bit, or perhaps leaking a bit, when the press measurement is hovering at 68PSI? The valve, as indicated by the PLC (CV 100% and 0 Volts at the analog output) is closed. The physical valve does not have an external indicator. I plan to put a multimeter at the valve to see the actual voltage coming in

You are spot on with the elements in this press system! I am in agreement that the issue lies within the mechanical system and not the PLC system. The situation is, with every question I ask of the operator crew, they always answer 'its always been broken' when it comes to the 71 tons SP. Another thing is, the people that knows about the exact specification of this Savage press has long retired. So questions like what psi is required for the press to achieve 71 tons gets blank looks everywhere, and there is no documentation.

As it stands, I have a plan to inspect the mechanical system of the press, by verifying with the maker of the press system (Savage). I plan on checking pressure at every element indicated above against the data from the manufacturer (Savage). I will plan a meeting with the original programmer as well, whenever I finish a request for fund for his time. I also will put extreme effort in convincing the customer that the issue lies not in the PLC and maybe it is in the mechanical system of the press.

So can you guys help me when I get asked the dreaded question "Well why does the PID command the valve to close by putting 0 volts at the analog out when the PV hovers at 68 tons?"....I have a hard time answering that as the customer is adamant that if the valve can open just a little bit "longer". I am thinking of doing a test where i write a "-1 volts" to the analog out and force the valve open to see.....

timryder · Feb 25, 2022

This is probably a Bi-directional servo value but only being used for one direction. That kind of Valve has a -10 to +10 input range where 0 is stationary. 0 to +10v is the Advance direction of the servo valve and 0 to -10v is the Return direction of the servo valve.

My guess was it was piped on the Return side (B Port) of this valve and not the Advance (A Port)

Peter Nachtwey · Feb 25, 2022

40mpg said:
What is the pressure of the hydraulic fluid at the discharge of the pump when the press measurement is hovering at 68PSI? I have not look but planned to look (the pump line at the pump's outlet says max 3000 psi)

3000 PSI max system pressure makes a lot more sense than 78 psi.

What is the pressure of the hydraulic fluid at the press when the press measurement is hovering at 68PSI? The pressure transducer located at the top of the press reads 1777 psi

That makes a lot more sense. Is that the output from the pump?

Why didn't you provide this information up front. You have been wasting our time.

Is the valve open a little bit, or perhaps leaking a bit, when the press measurement is hovering at 68PSI? The valve, as indicated by the PLC (CV 100% and 0 Volts at the analog output) is closed.

This doesn't make sense.

The physical valve does not have an external indicator. I plan to put a multimeter at the valve to see the actual voltage coming in

Well duh, it is about time.

You are spot on with the elements in this press system! I am in agreement that the issue lies within the mechanical system and not the PLC system. The situation is, with every question I ask of the operator crew, they always answer 'its always been broken' when it comes to the 71 tons SP.

71 tons makes a lot more sense than 71 psi.

Another thing is, the people that knows about the exact specification of this Savage press has long retired.

That is a problem so now you need to do some research.

So questions like what psi is required for the press to achieve 71 tons gets blank looks everywhere, and there is no documentation.

It doesn't surprise me. You need to know the pressures on each side of the piston and the areas on each side of the piston to compute the net force.

As it stands, I have a plan to inspect the mechanical system of the press, by verifying with the maker of the press system (Savage). I plan on checking pressure at every element indicated above against the data from the manufacturer (Savage). I will plan a meeting with the original programmer as well, whenever I finish a request for fund for his time. I also will put extreme effort in convincing the customer that the issue lies not in the PLC and maybe it is in the mechanical system of the press.

Finally.

So can you guys help me when I get asked the dreaded question "Well why does the PID command the valve to close by putting 0 volts at the analog out when the PV hovers at 68 tons?"....I have a hard time answering that as the customer is adamant that if the valve can open just a little bit "longer".

I think this part is a misunderstanding of how presses work.
This wouldn't surprise me at all.

I am thinking of doing a test where i write a "-1 volts" to the analog out and force the valve open to see.....

What is the valve's manufacturer and model number? Provide a link. I am not going to waste my time looking it up.

Trying to control something you don't understand is difficult.

drbitboy · Feb 25, 2022

40mpg said:
So can you guys help me when I get asked the dreaded question "Well why does the PID command the valve to close by putting 0 volts at the analog out when the PV hovers at 68 tons?"....I have a hard time answering that as the customer is adamant that if the valve can open just a little bit "longer". I am thinking of doing a test where i write a "-1 volts" to the analog out and force the valve open to see.....

The way to answer that query is to demonstrate that the Control Variable (engineering units i.e. in this case [-10:0]) from the PID is not 0 (100% output), i.e. it is not the PID that is holding the signal to the valve at 0V.

What would be minimally useful to answer that query would be to see a screenshot of the PID instruction, i.e. the block at the top of Page 684 here, when

you think the PID is controlling the pressure, and
the PLC is in RUN mode,
the press is "hovering around 68"

What would be most useful in addition to that screenshot would to see a screenshot representing every bit of the PID structure, including all elements .DATA[0] through .DATA[16], i.e. the data for the table starting at the bottom of page 686 here. Expanding the .CTL DINT into its component bits would also be helpful.

What would also be useful would be to hear about what is happening dynamically e.g. which values are bouncing around a single point, which are increasing or decreasing, etc.

Even better would be the entire program in PDF format, preferrably a couple of versions with live data (is that possible?) 10-20s seconds apart.

Finally, have you considered configuring -0.5 as the [Max. (100%)] engineering value for the PID output? That would ensure the PID does not close the valve, so if the valve is closed (0V measured signal from output card), then summat else is doing it.

I have been wondering for a while if the hydraulic pump is capable of regularly providing adequate HFP to meet the SP of 71.

But the first thing to find out is why the signal to the valve is 100% and the valve is closed; it is almost certainly not the PID doing it.

Peter Nachtwey · Feb 25, 2022

drbitboy said:
I understand fluid statics and dynamics fairly well but am willing to learn more, and I don't know a lot about the details of hydraulic systems.

Check this out
https://deltamotion.com/peter/Mathcad/Mathcad - Hydraulic Cylinder.pdf
Look at this picture.
https://deltamotion.com/peter/Pictures/RMC100/lawton.png
It is from about 1998. I was helping Lawton Press setup their press. Notice that when pressure is building at about 14 seconds the valve ( green line ) open up to increase pressure. Adding oil increases pressure. Subtracting oil at about 17.5 second reduces pressure. Notice that when the press is at the set point at about 16 seconds the output to the valve IS ZERO! That is because we don't want to add or subtract oil to maintain pressure. It makes sense to me that the OP's valve should be shut when at the set point or nearly shut when at the set point. What is wrong is that should be CV of near 0% NOT 100%.

The only case where the output will not be near 0% at the set point is if there is a leak someplace. You can hear any significant leaks when the pump has the pressure as high as 1700 psi. You can also feel for warm spots or use an infra red camera to find them.

OkiePC · Feb 25, 2022

Peter Nachtwey said:
Finally, it is a hydraulic press.
What has been written doesn't make sense at all.
You guys are too worried about the PID and not how a press works.

It only took 45 posts! Glad you're here, Peter, to drive in the winning run.

40mpg · Feb 28, 2022

Solved

Hello,

I would like to update the situation. First, thank you all for jogging my brain and helped pumped the brain juice. Every idea that was mentioned by everyone here certainly helped me look at the problem from a broader perspective.

1. The assumption that I've had about the proportional valve (PV2) opening wider to increase pressure to drive tonnage up was incorrect.

2. Looking at the P&ID (thank god the manufacturer had a copy). I've identified that there is another proportional valve (PV1) in the system, and its job is to open small enough so that enough pressure is introduced for PV2 to work with. The piece of code that controls PV1 is just a standard compute expression.

3. In the system's current state, the pressure introduced by PV1 is inadequate for the press to reach required tonnage. Maybe when the pressure system was just installed, with everything new with no pressure leak, those values were good enough to hit every SP. Maybe the pump curve is not the way it was new (almost certainly) and pressure leaking (checked at every Test Point and yup, measured pressure is below calculated pressure)

So I tweaked the compute expression for PV1 and boom, enough pressure everywhere to hit the required SP consistently.

Its pretty that the equation for calculation pressure is present on the P&ID, i wonder if that is normal..

Thank you everyone!

drbitboy · Feb 28, 2022

Thanks for the feedback.

Is PV1 on the high-pressure (supply) or low-pressure (return) side of the press?

Peter Nachtwey · Mar 1, 2022

40mpg said:
2. Looking at the P&ID (thank god the manufacturer had a copy).

It would have been good to make this available at the start.

3. In the system's current state, the pressure introduced by PV1 is inadequate for the press to reach required tonnage. Maybe when the pressure system was just installed, with everything new with no pressure leak, those values were good enough to hit every SP. Maybe the pump curve is not the way it was new (almost certainly) and pressure leaking (checked at every Test Point and yup, measured pressure is below calculated pressure)

Well duh. Should have done this at the start too.

So I tweaked the compute expression for PV1 and boom, enough pressure everywhere to hit the required SP consistently.

It wasn't a PID problem after all.

Its pretty that the equation for calculation pressure is present on the P&ID, i wonder if that is normal..

I have been at this for 40 years. I have NEVER seen a hydraulic designer provide an equation. Just a circuit.

This press still seems like a kludge. Most press manufacturers don't really understand how a press works. They can eventually make the press work by hit or miss.

Taylor Turner · Mar 1, 2022

Peter Nachtwey said:
I have been at this for 40 years. I have NEVER seen a hydraulic designer provide an equation. Just a circuit.

This press still seems like a kludge. Most press manufacturers don't really understand how a press works. They can eventually make the press work by hit or miss.

Hopefully, Beckwood will change your opinion on that matter in the near future. It's a matter of documentation now.

I actually used a Parker valve that has some system identification documentation with it. It just takes an explosion proof design to get it....
https://www.parker.com/Literature/H...ctrohydraulics/Bul HY14-1450-M1 BD15-BD30.pdf

Beckwood also has an outstanding service department to troubleshoot and retro fit older hydraulic systems.
https://beckwoodpress.com/service-retrofits/service/

Peter Nachtwey · Mar 1, 2022

Taylor Turner said:
Hopefully, Beckwood will change your opinion on that matter in the near future. It's a matter of documentation now.

Actually Beckwood was a shinning light even 20-25 years ago. Beckwood was ahead of it's competition. I remember that Beckwood has a C press that used a one Mitsubishi motor to control both position and force. No valves. Years ago Beckwood pioneered press leveling where the load does not need to be even and the platen/slider still stayed level. This is important for 4 corner presses. Beckwood would demonstrate this at trade shows. They did it with a RMC100. Now with new RMCs it is easy.
https://deltamotion.com/peter/Videos/SyncPosForce.mp4

The problem I see now is that many of the people that I worked with that had all the old 'tribal knowledge' have gone. This has happened at many places. This will be a problem for Delta too since no one really wants to get into the control theory I have done.

Help on PID control

bb76

Member

Peter Nachtwey

Member

drbitboy

Lifetime Supporting Member

Taylor Turner

Member

40mpg

Member

timryder

Member

Peter Nachtwey

Member

drbitboy

Lifetime Supporting Member

Peter Nachtwey

Member

OkiePC

Lifetime Supporting Member

40mpg

Member

drbitboy

Lifetime Supporting Member

Peter Nachtwey

Member

Taylor Turner

Member

Peter Nachtwey

Member

Similar Topics