Re: [CAD_CAM_EDM_DRO] Re: Measuring backlash more accurately?

Randy Gordon-Gilmore wrote:

>At 11:24 PM 2/16/03 -0600, Jon Elson wrote:
>
>
>
>>You might make sure the steppulse polarity is going the right
>>way for these drivers, too.
>>
>>
>
>I think you nailed it, Jon. I had not even paid attention to the pulse
>polarity. But near the back of the 29A55 brochure, it says "The step
>occurs on the rising edge of this signal." CNCPro was set "High-to-Low" as
>it was, and I flipped it to "Low-to-High" and the anomalous behavior seems
>to have stopped. When I cycle back and forth, there is no accumulation of
>movement now whether I have the backlash set at .0008, .001 or .0011. I
>will leave the compensation on all axes as I had originally measured it
>(.002X, .001Y and .002Z) and try my contouring program again tomorrow.
>
>I wonder if that has been the source of my "drifting" all along?!? Thanks
>for your insight, Jon! At least I now have a *very* carefully wired
>controller box... :-)
>
>

Well, the wrong sense of the step pulse could certainly do it!
Glad it seems to be working, now!

Jon