Embedded Transmitter PID vs. PLC / Controller PID

[kdcui]

Hi -
I have an application where there is an option to use a PID loop embedded in the Transmitter to directly control (via 4-20mA output from the transmitter) a final element (such as a control valve).

Does anyone have any experience using something like this over having everything originating at the PLC / Controller? Just trying to weigh the pros and cons. My gut tells me this could be tough to configure and tune, but on the other hand there could be performance gains.

In this case, the Transmitter is a PROFINET device so I would have visibility into the "inner workings" of the Transmitter.

Thanks

 

[Robb B]

Yes, I use an omron temperature controller which has built in PID functions. There's also a micrologix in the same panel, but it only controls the startup/shutdown sequence. The damper modulation is done by the omron. It works really well, though it is about 20 years old so the newer ones probably have a nicer interface. I'm not sure about prices...

 

[danw]

Are you sure about a PID function block in Profibus?



Back in 2010 or thereabouts, the difference between Foundation Fieldbus (FF) and Profibus was that FF had PID function block functionality, Profibus did not.


I haven't kept up with Profibus so maybe PID has been added, but the 2009 edition of Catching the Process Fieldbus: an Introduction to Profibus does not mention PID function blocks. But that was over a decade ago.



My opinion is that control belongs in the PLC/DCS, not in a field transmitter, but I'm an old guy.



I just watched the Chemical Safety Boards' video animation of the 2018 Ethylene Release and Fire at Kuraray America in Pasadena, Texas.


3 of 4 reactors had a maximum design pressure of 1150 PSI, one had a maximum design pressure of only 740 PSI. 28 people injured in the release of excess pressure to a flare (massive fire, burning ethylene blowing out of the relief valve under pressure) because the operator didn't remember that critical difference.


Is the operator going to remember that the PID control is in whatever sensor instrument is there, when all the others are not?

 

[cardosocea]

Is it a single variable control that will not be impacted by any other variable EVER? The benefit of the PLC is flexibility. If in the future you want to put a feedforward on that device, how easy would it be? Then there's obsolescence to consider. Will the manufacturer always make that or similar device to replace?

Some instances like hydraulic controllers, for example, the control you get is way better so there's a definite gain to be had there. in other situations (like temperature) there definitely isn't. Swings and roundabouts really.

 

[kdcui]

Thanks all for the input.
I have some other constraints too that might make the "use PID in transmitter" bit more risky (from a project execution perspective) so I am sticking with the PLC.