Host Engineering Forum

General Category => Do-more CPUs and Do-more Designer Software => Topic started by: Mike Nash on May 02, 2016, 06:38:01 PM

Title: Do-more PID Gain "Gotcha" in PID View
Post by: Mike Nash on May 02, 2016, 06:38:01 PM

I have a PID loop with a gain of 0.0001, because my PV input is around -40,000 to +40,000 since I am attempting web loop control via integrating speed feedback from an AC drive trying to follow a material feeder encoder. Looks promising over just a single photoeye start/stop in dry run testing.

The problem is that the PID View defaults to a min gain value of 0.000. I can enter 0.0001 and send it to the PLC and it is fine. What I can't do is then read it back from the PLC because it only "sees" 0.000 and that is what it will write back if I don't catch it. Same behavior opening the view. Zero gain is not nice.

It works great from Data View.

So is there somewhere to change the default resolution for PID View?

I could go about it lots of other ways to get around this, but all my units are reference pulses and encoder counts representing 0.001" so 1000 is 1 inch. I don't want to compromise with mixing reals with DInts.

I notice I can even put 1E-005 (0.00001) into the gain in PID View and it writes that fine. But again, it only reads 0.000 back.

With an input error of 1000 and a gain of 1E-009, the PID is still working so this looks to be a strictly a PID View issue.

Latest firmware and software.

Title: Re: Do-more PID Gain "Gotcha" in PID View
Post by: franji1 on May 03, 2016, 08:56:51 AM

Sadly, no.  The formatting is hard coded to 3 decimal places.   :-\

We can make that be configurable.  Sorry about that.

Title: Re: Do-more PID Gain "Gotcha" in PID View
Post by: Mike Nash on November 17, 2017, 06:48:05 PM

Quote from: franji1 on May 03, 2016, 08:56:51 AM

Sadly, no.  The formatting is hard coded to 3 decimal places.   :-\

We can make that be configurable.  Sorry about that.

Did this ever get made configurable?

I ask because the thermocouple cards and modules read in degrees *10 (450.0 degrees F is read as 4500) and since the Scale functions in the PID instruction add to the scan time (especially with a large number of loops) and MATH instructions do as well, I would prefer to simply have the raw values for PV and SP. But... I sure don't want to keep nailing myself with this 3 decimal problem.

Also, does anyone here know whether the T1F-14THM is 12 or 16 bit? AD's website says 12 and the PDFs say 16.

Title: Re: Do-more PID Gain "Gotcha" in PID View
Post by: franji1 on November 20, 2017, 08:50:48 AM

It did not get into 2.1.  I bumped its priority.

Title: Re: Do-more PID Gain "Gotcha" in PID View
Post by: Mike Nash on November 20, 2017, 10:03:20 AM

Thank You.

Title: Re: Do-more PID Gain "Gotcha" in PID View
Post by: franji1 on March 19, 2018, 03:27:06 PM

Implemented in 2.2

Title: Re: Do-more PID Gain "Gotcha" in PID View
Post by: Mike Nash on March 19, 2018, 06:05:52 PM

Thanks! I see the BRX analog in unipolar was also tweaked. Believe it or not, I haven't yet started that one up, so I may flash it in a couple of weeks.

The weird thing is how these updates come out right about the time the latest project has or needs to ship. I turned over a multi H2 system today to be shipped, so if it gets updated, it will also be on-site. It's the one I started this thread about.