Re: [CAD_CAM_EDM_DRO] Re: Adding friction to an axis? - or some other solution...

Seems like a bit of friction might be my friend - how incredibly
ironic... I will give this a try after the family returns from a couple
weeks away, (leave today). The idea of a rotary damper also seems worth
investigating if the staight friction suggestion doesn't work out.

I also suspected a rounding error could be the cause and haven't totally
ruled that one out either. I'm in the process of setting up Indexer LPT
& G-Code Controller to see if another controller eliminates the problem,
but based on so many people using Flashcut, I can't imagine it could
be the source of the problem. But hey, in this hobby, you can't take
anything for granted.

Thanks, gang!

Cheers, Peter



Steve Blackmore wrote:

> On Thu, 30 May 2002 22:20:11 -0400, you wrote:
>
>
>>Jan, the machine in question is a BP clone with rockford ball screws on
>>all axes and 1800 oz-in NEMA42 steppers direct coupled via helical
>>couplings. The machine is in near new condition and all axes move
>>freely. The problem I have is with long CAM generated files - I seem to
>>experience a cumulative error somewhat in Y & mostly in Z it seems. If
>>after running a 3D contour program for say 90 minutes if I return to
>>machine zero, I could be .015" out in Z and say .005" out in Y and bang
>>on in X. Accel rates etc have all been played with ad-nauseum. It
>>seems to occur more as a function of run time, rather than speed. And
>>while it would seem plausible that this could be a thermal problem or a
>>lack of friction problem - I can't imagine what could be getting so warm
>>as to cause a thermal problem.
>>
>
> I had the same problem recently - I was loosing steps in one direction
> on the Y - Axis. Tried all sorts of changes to IPS, acceleration etc.
> Somebody suggested lack of drag - Then I remembered I'd loosened one
> of the jib screws to fit a bracket for limit switches. I'd put it back
> - but not tight enough. Adjusted it up and the problem disappeared!
>
> It would seem some steppers like a bit of friction to work against.
>
> Steve Blackmore
>