PID tuning software from Rockwell

Hi,

I'd like to ask does anyone come across the PID tuning software from Rockwell - RSTUNE (if not mistaken).

What's the difference using RSTune comparing to the manual tuning in ladder logic?

Normally in ladder logic we'll be entering P, I & D values (try & error), so does the RSTune have auto tune features?

Customer told us that EuroTerm has auto tune features, but not AB/Rockwell product, anyone can clarify?

Thx.

Have a look here. Every thing you wanted to know and more. I have not used it myself, co. won't buy it.

http://www.software.rockwell.com/rstune/

RSTune is a badged version of a package called Expertune tailored for use with Rockwell PLC's. We use Expertunes ASCII version which allows us to use Scada trend data or even manually entered data for analysis. The package is expensive but worth its weight in gold if you regularly need to tune/optimise PID loops. Try www.expertune.com for more information.

Thx for the info, will chk it out

just so you'll know ...

in practice, it works something like this ...

(1) you set up RSTune by telling it the address of your PID’s control block, etc. ... RSTune will communicate with your PLC through RSLinx and begin to graph the response of the process ...

(2) you get the system running along and let it settle down at a steady state ... you can do this in manual mode ... or in automatic mode using your present PID tuning values ... you must make sure that no load changes, or other system upsets, etc. occur during the autotune procedure ...

(3) you tell RSTune to start the test ... it will ask permission to increase the setpoint by about 7% ... (if you’re in manual mode, it will ask permission to increase the controller’s output by about 7%) ...

(4) you watch the process on the graph and wait until it eventually settles down again at a new steady state value ...

(5) you click a button to tell RSTune that the new steady state has been reached ...

(6) RSTune will automatically calculate the new PID values that it recommends and display them on the screen ... you can click a button to have these new values downloaded to the PID ...

there are more features than this, but this is a basic overview of what’s involved in the autotune procedure ... I’ve found that the system works quite well for the few processes that I’ve tried it on ... just be aware that if your system is erratic and unstable (even during manual operation) then you’re probably not going to get satisfactory results from RSTune ... the basic idea is that it bases its calculations on your system’s response while changing from one steady state to another steady state ... so if your system won’t provide a steady state response (for one reason or another) then don’t look for miracles from RSTune ... at least that’s the way I’ve seen it work ...

I get real worried when people tune loops by just "trial and error" - I am far from the expert that Ron B is , but like to think that I can achieve a reasonable result .
I have had a demo of expertune , impressive , not just the loop tuning capabilities , but the fault find stuff too - can detect stiction , over and undersized valves and much more - very nice but a little heavy on the price .

A little story that may amuse :-

I had to recommission a big dryer for a fairly prestigious customer in China - the original engineers weren't welcome back .
I read the "software engineer's" report - he had tuned the temperature control loops with P , no I and D only , his report stated that it was the PLC's fault that the loops appeared to freeze and wouldn't unwind - his solution ? a timer that popped all the loops to manual for a cycle every 30 seconds - he said he noted an improvement . The PLC was a phoenix . The programmer ? a senior engineer for Siemens integrated solutions , and guess what , he even had the gaul to suggest to me PID terms to try - he is also the only engineer I have met who tried to preheat a passive cooler .....

Fred, that story is.... priceless...

Yet another reason not to have PC programmer messing with process control...

Worse , How on earth is this charlie writing software if he doesn't understand what it does ?
I'm resident in Indonesia , and I received the start-up request , together with a copy of the guys report (there were even two twits on site) - I couldn't believe it ! he even said he was going to complain to phoenix - now agreed , phoenix takes a bit of getting used to , but when you do , it is usable . Again , the fact that this guy didn't have the brain to even look on the net ( possibly one of Ron B's superb articles on PID , that could teach an aardvark to tune a loop) , terrifies me .
When I finaly lost it and had a fanny wobble because he hadn't included the up to date programming software in the bundle , and I couldn't get coms up with the processor ( phoenix has a sort of EDS for the hardware like AB) , he then proceeded to tell me that as it was a steam heated air dryer I should use this setting and that setting , and he still forgot about the bloody I term!

As I mentioned in a post some time back , I have started contracting for one of the real big drive guys , and have three months works coming up - I wondered if I wasn't taking on too much ,
Then I think back............

I would use a tool like FasTune (see https://processsoftwaresolutions.com) - its a quick way to obtain tuning parameters quickly and visually verifying the results. It connects to your ControlLogix/CompactLogix plc via RSLinx Classic, and uses the results of a manual bump to obtain the best gains, and lets you modify the speed of the response. You can also use it to verify the results by putting your loop back in auto and visually see the results of your tuning.

I hope he has his PID tuned by now, more that 9 years later. The PIDE instruction has an auto-tuning feature built in.

Tuned! I'd expect it to sing and dance after 9 years or more.