> What we use (in our very limited code space = 1K) for SoftDMC is over-drive 
> detection, This is implemented with a 16 bit counter that is incremented by N 
> when Drive (PWM etc) is 100 % and decremented by M when Drive is less than 
> 100 
> % each sample period. The counter is dead-ended at zero. If the counter ever 
> wraps, we do a shutdown on that axis.
> This allows us to program the trip time in sample periods, yet not have false 
> triggers when drive is near 100% (a value of M much larger than N will reset 
> the count if drive is ever less then full)
> With this simple system (and some reasonable amount of integral term in the 
> PID loop) you can pull an encoder connector and have the motor shut down in 
> 10 
> mS or so

this sounds doable in HAL. It would be completely independent of motion 
planning, but a safety measure nonetheless.

The duty cycle limiter would look at the DAC signal and effectively 
limit its duty cycle (time integral) to some value. I think you only 
need to define two parameters: the time-window, i.e. how much history is 
taken into account, and the duty cycle limit.
If the duty cycle ever reaches the limit the component triggers E-stop.

 From looking at the motor specs I can probably figure out what DAC 
value corresponds to the rated continuous current. But if the servo amps 
are voltage mode, the DAC value will not determine current directly...

So a question for Jon: are your PWM servo amps voltage or current mode ? 
i.e. when I set the DAC to a certain value, does that result in a 
constant motor voltage or in a constant motor current ?

I'm going to test something like this when I get my motors and amps going.

AW

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to