sampling time of fan coil unit with PID controller...


Thread Starter


I'm a student doing project on tuning parameters of PID controller of a fan coil unit, however
the output value of the PID controller oscillates between the high clamp and low clamp value that
result in poor performance. I don't know how to determine the sampling rate. Furthermore, there
is no bias input in the PID block so how can I obtain a good performance? Thank you!!

Heavner, Lou [FRS/AUS]


Have you done a simple bump test to try and identify the process dynamics? If not, it could be very enlightening.


Lou Heavner
Emerson Process Management

Dr. Igor Boiko

The problem you are referring to may have occurred due to undersampling or due to incorrect controller tuning. To determine the cause as
well as to assess the required sampling rate you would apply a conservative approach. Increase the sampling rate to the level that is 10-100
times of the highest possible frequency of the process variable signal you measure. If you don^Òt know it increase it up to the level your
controller allows. Tune the PID parameters. If the system works fine reduce the sampling rate to the level that doesn^Òt cause performance
degradation or keep it above that level.
Hope this would help.

Dr. Igor Boiko
Consulting in Control is available
(including modeling, simulations and control design)
[email protected]
Tel: 1-403-294-2745
The PID's sampling rate would ideally be an order of magnitude or more greater than the system response time, and ought to be of secondary importance. The system response, i.e., the time for a change in the controller's output to show up as a measurable input can be observed if you put the system in manual control, wait for
things to stabilize, then make a slight change (e.g., in valve position) and measure the time before you see a change in the loop's input. (Strictly speaking the time constant would be
measured at the point where the input is about 67% of its next stable state.)

Whatever the time constant is, unless you're trying to optimize the thing, double it (or more) and aim first for stable if sluggish control.

If the PID actually is using the I term the bias is taken care of. Otherwise, if it's P-only, the bias may be some fixed value (e.g., Barber Coleman analog controllers use/used 7.5 volts, midway in their inherent span of 6 to 9 volts).

Ken Irving <[email protected]>

Johan Bengtsson

My initial guess is that it is the tuning parameters, not the sampling time that is generating this behaviour (although it can be either or both)

Try to set the sample rate as high as possible, ie as low time as possible. The control don't take any real harm in a high sampling rate.

Have you done any tuning? If you have then how did you do? There are several tuning methods availiable producing very different results, like for example Ziegler-Nichols and lambda.

/Johan Bengtsson

Do you need education in the area of automation?
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]