Open loop flow control scheme

[MickW]

Hello All

Could we get some expertise on flow control ?

-Using a PID loop in Productivity 2000 with an analog output,
How can we convert that to a relay, open or close to control the flow of product?
We don't have any feedback loop ,just using a pneumatic air cylinder and inching valves
Thanks

 

[I_Automation]

Scale your PID output to 0 to 100. Do math on the PID output to match the timer accumulator

Run a looping timer 100 units (100 seconds, 1.00 seconds, 10,0 seconds - whatever you choose. But Don't run if PID output is 0

Do a LIM 1 Timer.ACC PID_Adjusted to OTE the valve output

The output will come on the percentage of time the PID says it wants.


EDIT: You could do any timer preset if 1, 10 or 100 seconds isn't appropriate, (say a 5.00 second loop is preferred) just a different multiplier to get PID_Adjusted

 

[lfe]

Why do you use a PID if you don't take the Process Variable into account?

I would do it with a table with two columns, Flow - Position, since the relationship is surely not going to be linear. Depending on the desired flow rate, the valve position is calculated by interpolating from the table.

But using a pneumatic cylinder will not work well since the movement in relation to the time of the impulse is approximate, the air is compressible, and when many position corrections have been made it will surely go out of calibration.

The ideal would be to have position feedback, or at least, for example, use an electric linear positioner, the ones that work with a thread.

 

[drbitboy]

Hello All

Could we get some expertise on flow control ?

-Using a PID loop in Productivity 2000 with an analog output,
How can we convert that to a relay, open or close to control the flow of product?
We don't have any feedback loop ,just using a pneumatic air cylinder and inching valves
Thanks

As noted by @lfe, a PID cannot be used without a feedback loop, so the process as described does not make any sense unless "We don't have any feedback loop" means there is no feedback on the position of the pneumatic cylinder and or the device that controls the flow of product, but there is feedback on the flow of product.

If that is the case, and that is a big if, the control scheme is to have the PID compare the flow rate feedback to a setpoint, and use any difference (error) to somehow turn a relay on or off, and that relay somehow controls the inching valve, which controls the speed (rate of change in position) of the device (final control element) that controls the flow of product. So the system cannot specify a specific setting of the final control element, rather it can use the inching valve to direct that element to increase flow, decrease flow, or leave flow at its current rate.

Am I close? Or could we have a more complete description of the process?

 

[Ken Moore]

This has been discussed and dissected many times, do a search on "timed proportional control".

 

[MickW]

Thanks again for your Inputs!

the control scheme is to have the PID compare the flow rate feedback to a setpoint, and use any difference (error) to somehow turn a relay on or off, and that relay somehow controls the inching valve,

 

[lfe]

Thanks again for your Inputs!

the control scheme is to have the PID compare the flow rate feedback to a setpoint, and use any difference (error) to somehow turn a relay on or off, and that relay somehow controls the inching valve,

What you describe is a closed loop control!

 

[drbitboy]

the control scheme is to have the PID compare the flow rate feedback to a setpoint, and use any difference (error) to somehow turn a relay on or off, and that relay somehow controls the inching valve,

that's a lot of "somehows" that make it difficult to answer the query from Post #1.

• What does the relay do?
• Is there feedback from the position of whatever final flow-control element (e.g. slide valve) that the inching valve changes?
• Does the inching value control the rate of change of the final flow-control element?
  • If yes, is that rate fixed or variable?
• Can the inching valve either close or open the final flow-control element?
  • Are there multiple inching valves, one opening and one for closing?
  • Or is there one inching valve driven by one relay, and the flow-control element opens more when the relay is activated and closes move when the relay is deactivate?

 

[I_Automation]

Another thing, if it's basically the same program just running different outputs you could run your routine setting bits like Call_Output_A, then in an outputs section XIC that bit, EQU the HMI value needed and then energize that output. Plus you may find more than one HMI value turns on the same output for A

 

[robertmee]

that's a lot of "somehows" that make it difficult to answer the query from Post #1.

• What does the relay do?
• Is there feedback from the position of whatever final flow-control element (e.g. slide valve) that the inching valve changes?
• Does the inching value control the rate of change of the final flow-control element?
  • If yes, is that rate fixed or variable?
• Can the inching valve either close or open the final flow-control element?
  • Are there multiple inching valves, one opening and one for closing?
  • Or is there one inching valve driven by one relay, and the flow-control element opens more when the relay is activated and closes move when the relay is deactivate?

In otherwords.....

 

[drbitboy]

In otherwords.....

I don't understand how people expect answers to these "how long is a piece a string" queries. I mean, if someone is mucking about with a PLC at this level, I would expect them to have enough of a technical/analytical/arithmetic/mathematical sense to provide relevant and somewhere-near-complete information.

Is that an unreasonable expectation? Or have businesses and plants let go so much support staff that they end up saying "Hey you-who-happened-to-show-up-today: can you take a stab at getting this PLC-controlled line back into operation?"

 

[robertmee]

I don't understand how people expect answers to these "how long is a piece a string" queries. I mean, if someone is mucking about with a PLC at this level, I would expect them to have enough of a technical/analytical/arithmetic/mathematical sense to provide relevant and somewhere-near-complete information.

Is that an unreasonable expectation? Or have businesses and plants let go so much support staff that they end up saying "Hey you-who-happened-to-show-up-today: can you take a stab at getting this PLC-controlled line back into operation?"

Cant speak for all, but in my industry that's exactly what's happened. Most plants cant find qualified technicians and Engineers because its long hours, not great pay, call in responsibiites, dirty and dangerous environment, etc. I service about 30 different plants and several contract me to be on call remotely to handle mundane issues.

 

[I_Automation]

"Hey you-who-happened-to-show-up-today: can you take a stab at getting this PLC-controlled line back into operation?"

That's exactly how I got into the field.

A stamping plant with a PLC2 controlled fully automated stamping press.

I was taken up to the down press, told " that's a PLC" handed the 3 inch thick greenbar paper printout and told to figure out why it wasn't running and get it running. 2 full days later I learned what a PLC was and got it running.

 

[Tom Jenkins]

The attached may help.

 

[MickW]

thanks everyone