pid bug in 8.2.1 output stuck at mininum value

[johnnyt]

Hi, first post here.  I recently upgraded from version 8.2.0 to 8.2.1.  I'm using the advanced PID vi to do closed loop control.  I'm starting my control output to .5 volts minimum to get a head start on the heater and capping at 3.5 volts maximum.  Everything was working fine, controller starts operating at the .5 volts the way I wanted, until I upgraded to version 8.2.1.  Now, when I first run the app, the controller stays stuck at .5 volts and won't climb above that  when the setpoint exceeds the feedback.  If I set the controller to manual mode and move the output value above .5 then set back to closed loop, it starts working.  I solved the problem by removing the PID toolkit and reinstalling the old 8.2.0 PID toolkit only.  So I'm going to assume something is wrong with Labview and not me.  Thanks.

[GiGe]

Hi Johnnyt,

Thank you for contacting National Instruments.  I am sorry that the upgrade wasn’t working for you.  I want to reproduce the error here, so I just wanted to clarify a few things from you.  If I understand correctly you saw the problem when using version 8.2.1 of the PID toolkit with LabVIEW 8.2.1.  And it was corrected when you went back to version 8.2 of the toolkit, but you are still using version 8.2.1 of LabVIEW.  Let me know if I misunderstood your set up. 

Have a great day,

GG

[johnnyt]

Yes, that is correct.  I'm still using Labview 8.2.1 but removed the control directory found here: 

C:\Program Files\National Instruments\LabVIEW 8.2\vi.lib\addons\control

and reinstalled only the PID toolkit from 8.2.0.

Also, I found that I could get the controller to start working if I switched the advanced PID vi from closed to open and then back to closed, if that helps any.  But, ultimately I had to remove only the PID toolkit and resinstall the old one.

I've attached the vi that uses the PID vi.

[nathand]

I suspect this is the same problem I reported in this thread: http://forums.ni.com/ni/board/message?board.id=170&message.id=239235
There's a revised version of the PID VI there that seems to work.

[johnnyt]

Just to clarify, the AO vi I attached was the working one with the old 8.2.0 toolkit reinstalled.  I suppose if you sub in the PID.vi from 8.2.1 it will break, but I do not have the time to try it.

nathand, thanks for the link to the other discussion. sounds like this may be a recurring problem.

[nathand]

I've just run into this same issue that johnnyt saw, and I now understand it's not the same one as I first reported in the thread linked above.  The problem is the following:  if the output of the PID Advanced is set to a value below the minimum PID loop, and then the Auto input is switched to true, the PID output will stay at the minimum output limit.  This happens if Auto?is off and the manual input is left unwired or is set to a value less than the minimum output.  Perhaps the same thing happens if the manual input is more than the maximum, I haven't checked.  This occurs with both the array and single-point versions of the PID toolkit, version 8.2.  I think I can get around this by modifying PID Advanced to force the output into range when Auto is first turned on, but is there a corrected version of the PID toolkit available?

[Angela M]

Hello All,

There is not yet a corrected version for this particular behavior.  Thank you for bringing this to our attention though, we are working on a fix which should be included in the next version.  As nathand mentioned, you can add some code to work around this for now.

Thank you and please post back if you have any other questions about this.

Best Regards,

Angela M

Applications Engineer

[jamrite]

Hello All,

I wanted to let you all know that this bug has been fixed in the LabVIEW PID toolkit version 8.5. Please refer to bug fixes section of the LabVIEW 8.5 PID Toolkit Readme for confirmation. This bug is refered to as:

452AFCCI

New Algorithm For Antiwindup in 8.2 Introduces Some Problems

Thank You,

Jaidev Amrite

National Instruments