I really like this approach, but I think one second is far
too long of a delay.

If your PID loop is saturated during normal operation, that
means you are no longer accurately controlling tool position,
and are probably making scrap parts.  So you should adjust
your accel down until it never saturates during either cutting
or rapids.  Just use halscope to look at the pid.saturated
pin while you are tuning and make sure it never trips.

Once it is adjusted so you never get saturation under normal
operation, then any saturation must indicate a problem - either
the gantry is jammed into something, or the encoder failed.
In that situation, I would have just a tiny bit of debounce
to avoid false trips - maybe 10 milliseconds.  That will stop
it before it gets going too fast.


On Wed, May 23, 2012, at 10:25 AM, andy pugh wrote:
> On 23 May 2012 07:08, Jeshua Lacock <[email protected]> wrote:
> 
> > I am guessing there is not an easy way to detect this condition,
> 
> It should be possible to check if your PID is saturated for more than
> a second or so.
> 
> loadrt debounce cfg=3
> addf debounce.0 servo-thread
> 
> setp debounce.0.delay 1000 #1 second delay in a 1uS servo-thread
> 
> net x-check debounce.0.0.in <= pid.0.saturated
> net y-check debounce.0.1.in <= pid.1.saturated
> net z-check debounce.0.2.in <= pid.2.saturated
> 
> net x-fault debounce.0.0.out => axis.0.amp-fault-in
> net y-fault debounce.0.1.out => axis.1.amp-fault-in
> net z-fault debounce.0.2.out => axis.2.amp-fault-in
> 
> Note that I am not sure what amp-fault-in actually does, but it was
> conveniently available and sounds about right.
> 
> -- 
> atp
> If you can't fix it, you don't own it.
> http://www.ifixit.com/Manifesto
> 
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and 
> threat landscape has changed and how IT managers can respond. Discussions 
> will include endpoint security, mobile security and the latest in malware 
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Emc-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/emc-users
-- 
  John Kasunich
  [email protected]


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to