Good PLC for Temp Control

I'm tasked with replacing a process's temperature controllers (which feature PID autotuning) with PLC control. Do any PLCs feature PID autotune or is that wishful thinking?

If autotune is not available with a PLC, then are any PLCs generally known to perform PID well, i.e. in a somewhat user-friendly way? More specifically, are any PLCs known to perform temp control well?

Those are my questions. I've probably exposed my ignorance in the mere asking, but here we are. Thanks in advance for any helpful responses.

Siemens S7-1200 has temperature PID feature with autotune function.

I haven't had much luck on autotune features within PLCs, but honestly haven't put much effort into using those either. I have used pidbot and Ignition. Both of these aren't free, but they both honor an unlimited 2-hour demo that can be reset, so essentially they can be used for free.

Pidbot, https://www.jlbcontrols.com/pidbot, is similar to Control Station LOOP-PRO software, https://controlstation.com/pid-loop-tuning/, which is not free and I don't believe has a demo at all.

I've had great luck using pidbot, I think it's worth a look at minimum.

Thank you both for your answers! Looking into Siemens S7-1200 and pidbot now. And in a side note, since I didn't think auto-tune was going to be a possibility, after I posted my question I started digging into PID tuning a bit. Seems fun. I suppose I could get a Siemens S7-1200 and try to tune it myself. If it goes haywire and I can't figure it out, I can go with autotune (I assume you have that option). Or same thing if I go with another PLC. Try it. And use pidbot for a plan B.

Again, thank you for your responses!

eaglewood said:

... since I didn't think auto-tune was going to be a possibility ...

Click to expand...

It's possible. AB CpLX/CLX has autotune using the PIDE instruction. I think Productivity from AutomationDirect has a tuner available. Those are just two. I'm sure other platforms offer it. But, a lot of this depends on your budget and/or how much you are willing to spend.

Honeywell HC900 has one of the best PIDs. Function Block only programming.
However.. I use a temp controller (Azbil SDC36) and send setpoint and temp from PLC to it. I also bring %power from temp controller back through plc for monitoring.

I love the Azbil controller, fantastic PID (supports 8 PIDs) and quick and precise fast Autotune. Easy to program also.
Now I do not have to mess with understanding the nuances of the PIDs found in all the different PLCs I use.

I wouldn't place much trust in PLC auto tuning unless it is the simplest of systems. Auto Tuning should be on the PC, not the PLC.

I've only been programming plcs for over 25 years. And I would go w/ a dedicated temp control unit. maybe something from red lion or something. Just my opinion. The D in PID is extremely difficult to tune. But then again, maybe I am just bad at my jobl

And Peter is the Goat on rolling his own PID algorithms. So what he said.

ganutenator said:

And Peter is the Goat on rolling his own PID algorithms. So what he said.

Click to expand...

But I agree with you about getting dedicated controllers. We use a dedicated Watlow temperature controller for our thermal chamber. It seems to be very stable even though we cycle the set point between extreme temperatures to test circuit cards.


No PLC or PC required for normal operation.

Although I agree with most others here especially Peter, you have not told us what the process is, for example a simple heat exchanger heating water or product in-line (or for that matter a direct steam injection) or pump speed control then a simple PID is easy to tune, however, if the process is like an oven where the load (I mean product) varies i.e. product/no product or gaps then keep dedicated controllers. perhaps explain the process in more detail & the expected accuracy.

parky said:

you have not told us what the process is

Click to expand...

Yes, so I will "hedge" my suggestion of a stand alone controller.
Stand alone controllers are great for applications where the process is static like in our thermal chamber where the load doesn't change.

, for example a simple heat exchanger heating water or product in-line (or for that matter a direct steam injection) or pump speed control then a simple PID is easy to tune

Click to expand...

Heat exchangers can be tricky if the flows change. Also, heat exchangers are non-linear. You can tune the system for the DLO or design level of operation which is to tune around just one set point. Otherwise the non-linearity of a heat exchanger will come into play. The www.controlguru.com site talks about this.

Stand alone temperature controllers will not know about changes in flow and non-linearity.

, however, if the process is like an oven where the load (I mean product) varies i.e. product/no product or gaps then keep dedicated controllers. perhaps explain the process in more detail & the expected accuracy.

Click to expand...

This is a good example where a more knowledge of the system is required. There are two ways to approach these problems and the right way is probably a combination of the two.
1. Use a feed forward term. Predict how much more heat is required as the rate of product increases.
2. Increase the integrator gain as a function of the flow rate. This has been discussed recently on other threads. Dead times may need to be adjusted by the flow rate too.


My answer about the stand alone controller may be right sometimes.

Parky has pointed out we don't know the process and I have pointed out why it is important to know the process.
Can you see why auto tuning doesn't always work?

ganutenator said:

I've only been programming plcs for over 25 years. And I would go w/ a dedicated temp control unit. maybe something from red lion or something. Just my opinion. The D in PID is extremely difficult to tune. But then again, maybe I am just bad at my jobl

Click to expand...

+1.
This is sound advice, and resolves issues like "what happens when I download a new program", or "how do I secure the controller". Dedicated controllers have these as standard features which are difficult or impossible to implement.

Maslow said: "When your only tool is a hammer, every problem starts to look like a nail."

We use a PID control for temperature control within the Beckhoff PLC platform. Although it does not have auto-tuning (as far as I'm aware), it's pretty easy to setup. We use it for setting and controlling the temperature of oil fluid in a reservoir tank. There are various function blocks for different types of controllers in the toolbox, depending on what your application needs are.

TF4100 Controller Toolbox


There is also the Temperature Controller (TF4110). Although I have not used this, it is obviously made specific for temperature control applications.

Peter, done probably over a hundred & fifty (most were different systems although many were of the same type i.e. heat exchangers).
Heat Exchanger, in-line direct injection, shell & tube, pressure, flow, nearly all of those hundred plus systems worked well with a simple PI only controller, one system the spec was 101.0 degrees C +/- 0.5, using a Rossi Cotelli direct injection core did require a little more control including "D" & a reasonably long time to get it right. Never had a problem with simple heating in-process so I do disagree a little. Some of the ovens I did were more complex & a dedicated controller was the norm & glad it was done that way.