PID AB Stdio5000

infer

Member
Join Date
Feb 2021
Location
UK
Posts
13
Hi All,


I have a bunch of actuators that are 4-20ma controlled by a PID function in Contrologix.

Some of the actuators are now replaced to a new type (SKB32) that are controlled directly with mains, one wire drives it to open and other one to close. We can't use OMRON controllers, so my question is what is the easiest way to convert the existing 4-20ma PID to work with the new actuators?


Cheers
 
Last edited:
What sort of actuators are they ? The name SKB32 suggests they're a Siemens linear electrohydraulic actuator. Do they drive a gate valve, or something else ?

It sounds like the older ones had a position that was proportional to the 4-20 mA signal, and the new ones have a "bang-bang" drive to open or close them.

What sort of feedback do the new actuators have ? Limit switches ? Analog position (the SKB32 has a potentiometer feedback option).

You are probably going to have to implement a position controller, whether you do it with analog and another PID, or with some other logic.

The worst case, in my view, is that you have to use the process feedback instead of a position feedback on the actuator itself.
 
Hi Ken,


Yes they are Siemens SDK32 acuators:


SKD32
3-position
control signal
The actuator is controlled by a 3-position signal either via terminals Y1 or Y2
and generates the desired stroke, which is transferred to the valve stem:
● Voltage on Y1: Piston extends Valve opens
● Voltage on n Y2: Piston retracts Valve closes
● No voltage on Y1 and Y2: Piston and valve stem remain in the
respective position
 
It's quite a challenge to take a closed-loop device and replace it with an open-loop one.

What feedback do you have from the actuator ? My quick reading of the product data sheets suggests it has a potentiometer and limit switch options.

Ideally, describe the process it is controlling and how you get feedback from that process.

Maybe you're going to have to drive the valve all the way in one direction, then open or close it by dead-reckoning to a target position, then use time * speed to adjust the distance periodically. Hysteresis will be a headache and so will power cycles.

In general, PID output is in percent. Is yours ? Does it go through any scaling into engineering units before it's fed to that 4-20 mA output module ?
 
I would think you may need to change from PID control to "Floating Point" control. The basic idea is that you have a setpoint, error, deadband, and control band. If the error is greater than the deadband, you would calculate an on-time for the output based on the control band.
 
I would think you may need to change from PID control to "Floating Point" control. The basic idea is that you have a setpoint, error, deadband, and control band. If the error is greater than the deadband, you would calculate an on-time for the output based on the control band.


This sounds like something I need. In plain language I have process valvue and need to cycle to outputs to drive the actuator to close/open to keep it at the setpoint. I have no bloody idea how a sample code could look like though.
 
Boundless Control

https://library.e.abb.com/public/a566f5a59fc1120ec1257b59004fd2bd/IM_CM_S-EN_L.pdf. Look on page 32 for an explanation if how it works.

I have done this before on HVAC applications for non critical environments. I have also used this method for flow control on VERY slow systems.

I personally think you are going to spend a lot of time making this work. I would change the valve actuators to 4-20 with feedback.
 
You can't reasonably solve this problem by keeping all the terminology abstract.

You either need to close a position loop locally on the actuator to make it more like the analog 4-20 position controller on the previous device, or you need to close a process loop on the actual process.

Again: what kind of feedback sensors or features do you have on the electrohydraulic actuator ?

Process loops are generally much slower than position loops, and we don't know about any of the details of the fluid, the process, the instruments, or the signals. The boundless control mechanism TWS cited is certainly one approach, but it may not be appropriate for your actual process.
 
Infer has not provided key information like what is really being controlled.
What does voltage on Y1 mean? I am assuming it is just normal AC power and not proportional so the motor moves or doesn't move. There is no rate control.
I have talked about controls like this with forum member Pandiani. They are used in power plants. Pandiani has lots of experience with what he called step controllers because the motor moves the valve in pulses or steps and they are made by Siemens. The valve position is not known. The valve is simply stepped open or closed until the PV=SP. This is a slow process. The rate pr duration of the steps is dependent on the error. Since the valve is bumped open or closed, it is essentially incrementing the steps or pulses. This means the "floating control" will not work because it would add yet another integrator which would be a disaster. Floating control is a slow integrator. The motor and valve already integrate the steps/positions. This means there would be TWO integrators and that would not be good.



DO NOT USE A PID. Keep it simple. Just generate a pulse with a length proportional to the error every few seconds. This is kind of like floating control but the "floating control" is already built into how the motor moves the valve. Like floating control, this is a slow way to control but it works and is stupid simple.
 
Infer has not provided key information like what is really being controlled.
What does voltage on Y1 mean? I am assuming it is just normal AC power and not proportional so the motor moves or doesn't move. There is no rate control.
I have talked about controls like this with forum member Pandiani. They are used in power plants. Pandiani has lots of experience with what he called step controllers because the motor moves the valve in pulses or steps and they are made by Siemens. The valve position is not known. The valve is simply stepped open or closed until the PV=SP. This is a slow process. The rate pr duration of the steps is dependent on the error. Since the valve is bumped open or closed, it is essentially incrementing the steps or pulses. This means the "floating control" will not work because it would add yet another integrator which would be a disaster. Floating control is a slow integrator. The motor and valve already integrate the steps/positions. This means there would be TWO integrators and that would not be good.



DO NOT USE A PID. Keep it simple. Just generate a pulse with a length proportional to the error every few seconds. This is kind of like floating control but the "floating control" is already built into how the motor moves the valve. Like floating control, this is a slow way to control but it works and is stupid simple.


I have something similar controlling a tracker on a production line. I never thought of scaling the pulse length to the error value though, maybe I can refine that a bit. Thanks!
 
I have something similar controlling a tracker on a production line. I never thought of scaling the pulse length to the error value though, maybe I can refine that a bit. Thanks!
Don't make the pulse lengths too long or try to get the PV to the SP in one long pulse. It should take many pulses with enough time in between to let the system react.

Another option is fixed length pulses and lower or higher frequencies or intervals. It would do the same thing but I don't think it is a easy to implement.
 
Alternative using actuator hardware slidewire feedback option and DIN rail mount converter in order to use a 4-20mA control signal

The SKB32 actuator has an option for a 1000 ohm slidewire feedback module.


1000-ohm-slidewire-is-optional.jpg




positiom-feedback-potentiometer.jpg




API 3200G is a DIN rail mount module that converts a 4-20mA input control signal to 3 position (open/close) output control when a potentiometer slidewire feedback signal is available for closed loop position control.


API-slidewire-feedback-wiring.jpg
 

Similar Topics

I have S7 1512C controler for controlling 48 PID temperature loop, the output is PWM. Please I need the best, most efficient way to write the...
Replies
5
Views
212
Hi all, I'm having trouble solving a problem I've been working on for several months, and thought you might like a stab at it. The machine runs...
Replies
22
Views
840
How can I connect PID Output to a valve. In ladder logic program is there any logic do I want to add between valve and PID? PV=SP What will be the...
Replies
7
Views
383
I'm setting up a PID for a 50hp RO pump in a water treatment plant. When I go to manual mode on hmi I set the swm bit, I also ons the pid output%...
Replies
5
Views
523
I am setting up control for Hypochlorite dosing. The easy part is the dosing calculation for flow pacing but I would also like to setup trimming...
Replies
8
Views
889
Back
Top Bottom