Honeywell PD Control

[controlnovice]

We have a new (to us) exothermic reactor process in which I've had success in using PD control (P on error, D on PV)using Emerson's DeltaV. I did try PID at first, but did not get the response we were looking for. The PD worked great.

Now we want to introduce this product in another plant that has Honeywell's Plantscape control system. From their documentation, they do not offer an option for PD control (they have 5 options only, PD is not one of them).

Does anyone know if the 'I' component can be 'dialed out' in a PID block using Plantscape? Does making the 'I' tuning parameter 0 (zero) do the trick? Does anyone have other ideas for a likeness of PD control using PID?

PD control is not that UNcommon. I'm surprised, and frustrated, it is not standard in Honeywell's control system.

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[Ashereng]

Setting it to zero should work, if the integration term is in units of cycles/time.

If it is time/cycle, you need to check if zero will work.

Honeywell also has a "badval" string (not "blank" but a separate "badval" string). That may work.

I have not set up a PD controller - usually, it's P or PI only. A PD controller sounds like you are trying to catch up to a change in PV with a bit of dead time. Is that right?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[controlnovice]

We tried setting it to zero in another control system, and it didn't work (it was time/cycle).

I've contacted Honeywell, and they couldn't answer. They will bring it further up the line....

I have not set up a PD controller - usually, it's P or PI only. A PD controller sounds like you are trying to catch up to a change in PV with a bit of dead time. Is that right?

I think I'd describe it as predicting where the PV will be, and moving the valve on this predicted PV.

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[Ashereng]

Let me know what HWL comes back with. I'd be interested in knowing also.

Thanks,

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[controlnovice]

Ashereng,

Honeywell did respond and the result is setting the Integral term to 0 to obtain PD control.

We also have to set the bias of the controller to 0. In our case, we want the valve to be closed if the error is 0.

PD control is often used on exothermic reacions.

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[Ashereng]

controlnovice,

Well, I am glad my response is correct. I was going on memory, and it isn't what it used to be.

With regards to PD control on exothermic reactions, why not the integral term? What is it about and exothermic reaction that precludes using the integral? Like I said, I don't usually use PD, more typically, just P, and PI.


"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[controlnovice]

Well, as my username suggests, I'm not the expert....

I've read/seen/used PD control on Exothermic batch reactor mainly using cascade controls with the cooling water or tempered water system. I should not have stated that it is often used for exothermic reactors solely based on my knowledge! I do know of other exothermic reactors than use PID.

Basically, the 'I' term can slow down the response. Something we can't afford on our exothermic process.





______________________________________________________________________________
This is normally the space where people post something insightful.

 

[Ashereng]

Yes, the Integral term adds lag time to the system.

I guess in your exothermic process, you can live with the offset.


A novice is a relative term. I come across a whole bunch of business cards that say "Senior XXXXX". You talk to them a bit, and it seems like their 25 years of experience were not very fruitful.

I try to judge people's skill/knowledge/experience based on the quality of their work produced.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[GerryMoore]

The I in any PID is usually used to correct any offset found in the control system over the long term.

I don't know anything about your process, so I can't comment as to how this applies to your situation. I like using D. Its not required very often so its nice when you finally get a chance to play with it.

"A novice is a relative term. I come across a whole bunch of business cards that say "Senior XXXXX". You talk to them a bit, and it seems like their 25 years of experience were not very fruitful.
"

As my old boss used to say, "Do you have 25 years experience, or 1 year of experience 25 times?"

 

[controlnovice]

In the 'Old' days, pneumatically speaking, the D term was rarely used because the vendors would charge extra for it!

Now that we're going electronic, it's an easy addition and I believe more people will start using it on non-noisy PVs.

______________________________________________________________________________
This is normally the space where people post something insightful.