Re: [CAD_CAM_EDM_DRO] Re: 8-wire stepper: series is 1/2 the amps.Strange but True

On Sun, 01 Jun 2003 19:54:53 -0500, you wrote:

>Stepper Driver circuit design
>
>I have gotten my samples from Allegro, and they're the 3967 single chip
>bipolar stepper translator/driver. Nice chip...inconvenient package for
>a homebuilder (24 pin SOIC), but a nice chip nontheless. that's what
>PCB CAD libraries are for.

ouch, not a convenient package for me....

>
>So...I have a chip, a power supply, and a stepper motor. Now what? ;-)

(punt).... <grin>

>
>No...I'm not starting out quite that basic -- the circuitry is not going
>to be a big problem, but a motor is an inductive load, which sometimes
>complicates things. I assume there will be backcurrents to deal with,
>possibly interference from traces laid parallel to each other at certain
>frequencies, etc. I understand the parallel port can operate at 50KHz?
>

perhaps...

The main problem I think you will have is the isolation between the
stepper motor currents and the logic paths.. Please check the app
note and design notes on the 297 and 298 for a hint of how they
control the spikes and currents.

>So...any tips on stepper driver board design, from those who have done
>it? Do I need to isolate the grounds for the various parts of the
>circuit? Diodes to prevent backcurrent from reaching the chips? That
>sort of thing. I have only the white sheets from Allegro, at this
>point, on which to base my design.
>

If their suggested designs have snubbing diodes (the 298 needs them),
then go to it. If not (and there's a DC motor controller chip that
doesn't), then you don't need them...

The main thing that I would do is to try to have separate grounds for
the logic and the motor driver, then tie those grounds together at one
point. If you are making one board, then that's going to mean that
you can tie the grounds together, but only at one point on the board.
If you need separate boards, then you can try to split the grounds
*if* the chip allows it.

>Has anyone ever designed a circuit based on this chip? I understand
>it's new...single chip translator and microstepper for bipolar drives.
> If I can get it to work, I think it'll be nice.

Sounds good. I did a design on the 298 and 297, and I've got a few
bypassing things to think of... but it could be the power supply as
well, since I'm not sure about the bench supply's behavior on current
limiting....

Harvey

>
>THANKS!
>
> -- Chuck Knight