Honeywell PID Equation Query

A

Thread Starter

Andy Clegg

I'm doing a bit of work on an Honeywell Micro TDC-3000 system and wondered if anyone could confirm what the specific PID equation used. It is for the interacting form of PID, and it is Equation B where the Derivative term is acting on the PV rather than the error (SP-PV).

I have an idea that it is :

OP = [ K*(60*T1*s+1)/(60*T1*s) ]*error + [(60*T2*s)/(0.1*60*T2*s+1)]*PV

where:
K = proportional gain
T1 = integral time (mins/repeat)
T2 = derivative time (mins)

and that the 0.1*60*T2*s+1 bit in the derivative action is the filter implemented for the derivative action.

However, my reference book seems to indicate a different (and what seems to me incorrect) structure. I am also wondering if the proportional gain should also apply to the derivative part of the above equation as well.

Thanks in advance for any help.
Andy Clegg
[email protected]
Advanced Control Technology Club, Industrial Systems and Control Ltd.,
50 George Street, Glasgow, G1 1QE
Tel: (+44) 0141 553 1111
http://www.isc-ltd.com/actclub.html
Fax: (+44) 0141 553 1232
 
C
The following is the equation in our reference books for the Interactive PID Equation B. The nomenclature is different, but gain does work on both the integral and derivative time constants.

CVs = K * [((1+T1*s)/(T1*s))*((1+T2*s)/(1+a*T2*s))* PVPs - ((1+T1*s)/(T1*s))*SPPs]

Where:
CV = Output in %
a = Constant equal to 0.1 for interactive
K = Gain
PVP = Process variable in %
s = The Laplace Operator
SPP = Set point in percent
T1 = Integral time constant in minutes per repeat
T2 = Derivative time constant in minutes.

I hope this helps.
 
J

Johan Bengtsson

I don't really know about this implementation but the gain should normally affect all parts of the controller, including the derivative part.


/Johan Bengtsson

----------------------------------------
P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
M
Andy,

Derivative can be on either SP or ERROR depending upon the type of response you want to setpoint changes. Eurotherm controllers are selectible. You select PV-only if you want a more sluggish response to setpoint change and Derivative on ERROR for a more aggressive response to SP change.

Cheers,
Mitch
 
J

Johan Bengtsson

Just to try to make things less confusing:
(if I fail that - sorry)

These two things are completely unrelated to each other you can have the D part act on either control error or PV (process value). The difference is what will happen when the setpoint changes.

I do, however, think the original question was more about if the derivative action should be affected when you change the gain. I say you normally want that and good implementations
of a PID controller is done that way.


/Johan Bengtsson

----------------------------------------
P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
M
Just to clarify some of the comments on control action types:-

The Honeywell TDC3000 provides a number of different PID options (depending on the controller).
Equ A - all 3 terms (P,I & D) act on the error (PV-SP)
Equ B - P & I act on error (PV-SP). D acts on PV changes. This equ used to eliminate spikes on SP changes.
Equ C - I acts on error (PV-SP), P & D act on PV changes. Produces smoothest and slowest response to SP changes.
Equ D - Only Integral control.

The above taken from the Honeywell PM Control Functions & Algorithms manual.

Malc Boczek
Control Systems Specialist
Terra Industries
 
Top