Re: [CAD_CAM_EDM_DRO] another stepper scaling question

Hi Murry,

If I understand you correctly, you DON'T have two stepper driven axis?
So you can't just run a controller program?

Your ratio's seem correct, that's basically what a controller program
does.
The "minor axis" distance is divided by the The "major axis" (one with
the furthest to move) distance, and this fraction of a step is added
each time a major axis step is taken.

If only one axis has a stepper you can drive, and the other axis has an
encoder, then a "tight program loop" could watch for encoder pulses, and
increment a value for the other axis (by a fraction). When the value
was greater then one step past it's present position, the stepper could
be advanced. This works well as long as the encoder has the major move,
a little more difficult if the encoder is the minor move!

Does this address your question? Choosing the encoder CPI, and/or using
a Gecko G210 to get choices of microstep could ease these ratios.

This would work on the threading example, and would seem to apply to
your rotary table example too.

HTH

Alan KM6VV


"Multi-Volti Devices (Murray)" wrote:

>
> Hello:
>
> I have a different stepper scaling question . I don'tknow what kind of
> software I'll be running, probably something from Compumotor because I have
> one of their 2-axis indexer/controllers.
>
> Anyway, my question is generic -
>
> If I have one axis that has to move proportionally to another, say perhaps
> analogous to how I imagine screws might be cut...actually that's a good
> analogy, because my application has to move the pitch of one leadscrew
> relative to the rotation of a shaft.
>
> My thought is that if I have a 200 step motor on the rotating shaft, and a
> Gecko 201 (1/10 ustep), I have 2000 steps per rev. The other axis has a 5
> pitch (0.2 inch) leadscrew, so each motor rotation moves 0.2" or 0.2/200 =
> 0.001" per step.
>
> So the question is - suppose I want a particular rate of motion on the
> linear axis (lead screw). I imagine I could multiply two factors to come up
> with a least common denominator then divide the pulse/rate info for each
> axis by the respective factor from the other axis.
>
> I don't know if I'm making any sense, but I hope someone either agrees or
> straightens me out.
>
> Thanks
>
> Murray