Re: [CAD_CAM_EDM_DRO] EMC and stepper driver

Wally K wrote:

> I have stolen the following post from a emc message site.
>
> http://www.nist.gov/itl/div896/emaildir/emc/msg01620.html
>
> It is from our very own Tim Goldstein, and deals with Steppers and
> servos under EMC. acording to Tim the G320 servo drive runs great at
> 30,000 steps per second and steppers with a stepper drive cant run
> any way near this fast.
>
> My question is. Is the front end of the G320 somehow conditioning
> the signal or are servos just more forgiving.

Yes to both. Servos can not lose sync or miss steps, as there are none.

The Gecko servo drives can lose steps, in blocks of 128, I think, if I
understand his design. Those would probably be easily detected, unlike
an occasional missed step. If the computer gets 128 steps ahead of the
physical position as read by the encoder, then the counter will overflow

and start over at zero error. A true servo system will never lose
position,
as long as the encoder is not disconnected from the control, or the
program
restarted. (When this needs to be done, the reference can be
reestablished
from the home position.)

> If the G320 is
> conditioning the signal then we should ask Mariss at geckodrive to
> put this conditioning in his line of stepper controllers.

I believe his line of multiplier stepping drives have this feature.

Jon