Micrologix 1500 PID loop

Hi Folks,
I have never used PID loop, and must setup one now. Could somebody give me an example for one when the read value is higher than the set point, then the output must be lowered? The help files in RSLogix 500 are not clear to me. Thanks in advance.

Off the top of my head, set E=SP-PV. This will make the error negative when the PV (process variable) is higher than SP (the set point). The PID output will reduce as the error gets more negative.
You need to get the Micrologix 1500 instruction set reference from the AB web site, it has a lot more information about PID control and also a very good method of tuning your PID loop. Publication 1762-RM001E-EN-P

Here is the bit from the manual that explains the control mode CM.



Change the bit during programming by double-clicking the PID block to access the PID settings.

Thanks JohnW!, I'll try that.
Still need to get familiar with the meaning of each parameter in the PID file.
If someone can attach an example of a pH control PID loop it'll be great.
Thanks again!

How are you dosing for pH control?

I've an LMI analog dosage pump.
pH must be over 7.5 to be acceptable, so chemical is added in a recirculation line pumped from and to the same tank.
I tried your suggestion and seems to work, but still have trouble because the output increases gradually from 0 to 100% and stays in there.
The loop is timed every 1 second, gain is 1.0 tried set point 8.0, min value is 0, and max is 14.
Thanks again for the help!

pH control can be quite problematic with PID. I got involved in a problem at a sewage treatment plant (in a factory producing cooking oil)about 15 years ago where the Step5 PLC people had spent 3 weeks trying unsuccessfully to get it to work (I'd never even seen a PLC at the time, never mind knowing my way around Step5!).

In this case it was a screw feed system adding chalk (also heading for a setpoint of about 7.5, if I remember correctly). The problem was that once you get a pH below about 6, because the pH value initially changes only very slowly as you add chalk, the PID tends to ramp up to maximum. However, once you get near to 7, the pH value suddenly starts to change very rapidly. A normal PID controller can't cope with this and so it tends to overshoot badly. (With the result that the water going to the municipal water treatment plant was out of spec, resulting in a very big bill for the company concerned!)

I haven't been involved with loop tuning for the last ten years or so, so I don't know whether modern controllers may have means to cope with this. I solved the problem at the time by switching off the PID controller (in the software) as the pH approached 7 (i.e. stopping the feed screw dead) and then re-initialised and restarted the PID control. Now that the offset was relatively small the PID loop would regulate quite happily until the next dose of heavily acidic sludge arrived.

Guille,

I need more information. Did you follow the loop tuning procedure in the manual? I would say your loop update time is far too fast for pH control. Also as you probably know, pH is exponential so you may need to change your PID parameters for different pH levels. Also it sounds like you have no integral term. Post a screen shot of your PID settings so I can have a look. Do you know how to set up a trend graph? If you do then use that for your tuning.

I think RMA hit your problem on the head. pH is a logarithmic scale. Using PID is a very difficult way to attempt correction. You are probably going to be better off using your own control algorithm, with the amount of chemical feed based on tabulated values. You have a lot of dead time in your system as well, as a result of mixing and sample line recirculation. Include some time delay between changes in the feed rate and the next increase in chemical feed.

Chnage the topic for controls Ph.

Exact, it is a logarítmic. But it is not identical for all kinds of products, with small dependent variations of the product. The kind of product, classifies the microorganisms reagentes (sewer, batter, gas) that determines the rank of intensity of the reaction, defined a smoother or abrupt curve. Also the definite product the position of the "knee" with value of Ph more bass. Therefore, there is existence of sensors of Ph appropriate for each class of product, obtaining the compatibilization of the curve and a sign "married with the curve of reaction".

To these facts the control must "recognize" the form of the curve, way of control:

PID passive - Maintain the block PID in RUN, but do not transfer for the output, up to reaction stabilize.
PID adaptative- Some of the parameters to be adapted the situation, in case of, in agreement cited the Time of integration, should alter
Controlers modern, basically possessed algorithm for recognize "the knee" and adapt, beyond recognize incompatibility between sensor and sytem.
Each component reagent, possessed influence in the curve, the intensity of dosage (saturation) can modify the form of the curve.Ph requires "control online" or prepared specifically, as Jenkings affirmed data in table.

Contend based in documentations, without practice, but promptly!
Obs:My HD with defect.I have thesis in english in the subject. (Two post, PC bug publish to send long text)