Combined flow and pressure control of pump

Hi,

Just got a specification from a customer that said:
"The outlet flow to be controlled by:
- A combined flow rate/outlet pressure for all pumps (4pcs) (using the common PIT
and FIT)."

I've never heard of that I can use two PV to set a output. Is it possible? I will for sure talk with the customer what he wan't but always wan't to have good basic knowledge beforehand.

A follow up question is how to best control 4 pumps from one PV, but I leave that to another thread.

/Tim

Sounds like they want a low select. You can look it up on the ISA website.

SD_Scott said:

Sounds like they want a low select. You can look it up on the ISA website.

Click to expand...

Or search for "override control".

PID outs paraller, lower output is controlling?

Thanks! I'll google some ����

Lare said:

PID outs paraller, lower output is controlling?

Click to expand...

I would think that does not work with PID's, at least not without some additional fiddling. The I component of the "loosing" PID would keep winding up.

Firstly you would need two controllers i.e. one for pressure and one for flow just using a pump would not work, to try to control a pressure and flow with one pump would be difficult. Pressure and flow are related to each other, as pressure difference being the cause of the flow.
When a pressure difference between two points exists within a fluid and is not balanced out by the internal forces acting on the body, the fluid will start to move from the higher pressure point to the low pressure point to minimize the pressure difference. This continuous movement of the fluid is known as flow.
I often worked on in-line steam injector systems using two PID Controls, one for pressure & one for temperature so similar in nature, even these were difficult to tune. But when tuned correctly could achieve 101.0 DegC +- 0.5 C without difficulty (Before anybody gets smart yes this was liquid but it was under pressure so a temperature above 100 is easy).
Edit: It will depend on the flow restriction i.e. pipe size etc. I suggest you ask them do they mean that the max pressure should not be above x bar while maintaining a particular flowrate. maintaining a flowrate at a particular pressure will depend on each other so for example if there was a blockage the pressure will rise as the flowrate falls, if you try to reduce the pressure the flowrate will also fall. and vice versa.

Toine said:

I would think that does not work with PID's, at least not without some additional fiddling. The I component of the "loosing" PID would keep winding up.

Click to expand...

More info of wanted controlling is needed.


There can be at least 3 situations.
- Only pressure control for pumps
- Only flow control (Which needs basically reservoir tank on other side, so that pressure don't go high if there lower depand of flow than PID tryes to maintain.
- Pressure and flow control same time.
(This would be basically max pressure or flow rate, which one is reached sooner.
If flow rate is setted too big for pipes, then flow is never reached, and max pressure is maintained.
If other side have big changes on how much there can be pumped, then max flow is controlling first and when less fluid is needed, pressure starts to raise and pressure takes control.
Only pressure PID would pump to "empty" pipe too much fluid, and flow PID is needed to reduce this flow, if pressure won't rise above SP.)


Toine is right on this. "loosing" PId needs some manipulation or fiddling, so that PID change from one to other is fast.

There is several way to do that, but if you have PID out max variable, then put "loosing" PID output to little bit bigger via max-out than contollinmg PID output is.


That way both PID outputs are pretty close of each other. Then when situations changes, changing to other PID is faster than if another PID is having 100% at output, before it starts to drop.


You could also try to tune other PID faster without any manipulation to PID.

Of course PIDs can also be on series, put it is harder to tune.

Thanks for all the ideas and hint! I've read about low select and override. It turns out that the pid block for s7-1200 has an override feature to prevent windup of the inactive controller. Very good to know!

whit override it should be easier to manipulate PID.


This expalains high select and how it can done.

https://instrumentationtools.com/pid-controllers-with-output-high-select-logic/


Have done low select allmost similarly

Lare said:

whit override it should be easier to manipulate PID.


This expalains high select and how it can done.

https://instrumentationtools.com/pid-controllers-with-output-high-select-logic/


Have done low select allmost similarly

Click to expand...

It seems that the PID compact v2 in TIA portal suggest that I can use the built in override instead of a low select in this case. I've also read that link, very usefull in understanding what I wan't to achieve when selecting outputs

Before you get too deep into the weeds look at the process and find out from the customer what process performance parameters they want to control.

As a general rule, if you are controlling a variable speed pump you can control pressure or you can control flow, but you can't control both simultaneously. For every given total restriction from the piping there is a given system curve that defines the pressure vs. flow characteristics. If you change flow pressure always changes. If the restriction changes then you get a new system curve.

Thanks for the input. I'm well aware of the issue with trying to have two process values. This is only in early stage and I will get more info as we go but I want to be prepared for every eventuality and also get more educated as I go. Even if it most probably will end up with one PV and this was just a miss communication