Re: INCONSISTENT STEPPING - WIERD SITUATION

Bob:

I had a similar problem but with radically different equipment. I'm using EMC
with Camtronic drives. I initially had opto-isolaters which inverted the
signals between the parallel port and dirvers. It turned out the timing
between the step and direction pulses was "off" enough to cause one lost step
each time an axis went + to - but not - to +. This stacked up over a large
number of moves to a noticable error.

You could try this: Put a dial indicator on the Z axis and zero, write code
to move up and down say .1" 100 times, execute. Does the error indicated
divided by the number of moves match the step distance for that axis? For
mine each step is 0.00025", 100 moves would have given 0.025" error. If so
you're losing one step in direction change.

For mine I took the opto-isolator out and it worked fine. I haven't tried to
track it down any further. For yours it may give a direction to look in.

For what it's worth.

Hugh Currin
Klamath Falls, OR

On Sunday 10 March 2002 06:20 pm, you wrote:
>
> I do a lot of 3D engraving which means my machine is doing a lot of short
> moves in the Z direction. After about 10 minutes of operation, my Z height
> has moved up by about .020".
>
> I am using the CNC controller from Microsystems. I have also tried it with
> cncpro. With cncpro, it goes down about .020".