Hi Charles.

If I'm not wrong in HAL you can setup "threads" as you need.
So I believe that perform the PID every 1ms is too fast and not needed for
a "relatively slow"
temperature control.
Also because you have a 20ms reading temperature cycle.
If you look at the HAL User Manual you will find the info you need.
As an example from the manual with a period of 1ms :

halcmd: loadrt threads name1=test-thread period1=1000000

Alex



On Thu, Jun 27, 2013 at 4:56 PM, Chris Radek <[email protected]> wrote:

> On Thu, Jun 27, 2013 at 09:42:37AM -0500, Charles Steinkuehler wrote:
>
> > I was trying some small I gains, mostly between .001 and .024.  I
> > _think_ the PID thread rate is being accounted for, but if not that
> > could easily explain things (I'll have to dig through the code).
>
> The pid manpage tells you how the scaling works, and even gives an
> example for Igain.
>
> > I suspect, however, that given the slow rates my temperature moves
> > around, I may have been dealing more with integrator wind-up than with
> > bad gains.
>
> Yes I assume you have a step command in temperature, say from 0 to
> 180 degrees.  You'll immediately get huge error and the output will
> saturate.  This saturated output is actually what you want.  But
> this causes I to start accumulating and it'll wind up quite a bit
> before your feedback catches up.
>
> I doubt you want I gain at all in this situation where you expect to
> have big errors.  You'll probably want P and D only; D will be the
> one that you tune to prevent overshoot and oscillation.
>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Windows:
>
> Build for Windows Store.
>
> http://p.sf.net/sfu/windows-dev2dev
> _______________________________________________
> Emc-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to