[louieatienza]

well it did not fix the problem.

any other suggestions would be great I'm willing to try anything at this point

Just a thought... are your spiders snug with the couplings or is there any play? Also are they a soft compound?

Also I might have missed it, but what is the torque rating of your steppers? And what driver board are you using?

edit: scrolled back... have you considered the driver board itself as a possible culprit?

[dumb_luck]

Just a thought... are your spiders snug with the couplings or is there any play? Also are they a soft compound?

Also I might have missed it, but what is the torque rating of your steppers? And what driver board are you using?

edit: scrolled back... have you considered the driver board itself as a possible culprit?

spiders are super tight and the red ones I think they are the hardest

I'm to the point were I think its the driver board. not sure how to test it though.

[louieatienza]

spiders are super tight and the red ones I think they are the hardest

I'm to the point were I think its the driver board. not sure how to test it though.

Also, I think another member of the forum had problems due to the parallel port on the computer, and was resolved by swapping computers...

[dumb_luck]

I'll drag the computer from office down to shop and see if that works

[JerryBurks]

spiders are super tight and the red ones I think they are the hardest

I'm to the point were I think its the driver board. not sure how to test it though.

I had loss of steps when I used maximum microstepping (1/50). According to the specs of the controller and the driver board that should have been possible but I had random small position return errors. No problem with 1/8 or 1/16 microstepping and so I left it at that. I did not investigate but maybe the pulse width got too small to be reliably transmitted across the driver optocoupler input.

[dumb_luck]

WOW Vista is a pain to run with Mach3.

finally got everything running and the error is still there. It must be the break out board. I've ordered another one.

[dumb_luck]

I just wanted to go through what I've done to date.

The problem: step .01 one direction and the machines moves .01 step the other direction and the machine moves .0096. continue stepping back and forth and machine moves away from gauge.

I don't believe it is mechanical because if I switch the motor direction by changing the phase wires the error is reversed.

I have tried different stepper controller with same breakout board error remains the same.

I have tried different computer error remains the same.

I have tried all combinations of pulse width and step width including sherline mode.

I have tried turning micro stepping down to 1 and all combinations in between.

I have a new break out board on the way from cncgeeker.com. hoping this will be the solution.

any other thoughts and Ideas would be appreciated.

[ger21]

What about the step pin active low setting? Try changing that.

[CarveOne]

It sounds like mechanical slippage in a motor coupling or a motor shaft to pulley connection is slipping only in one direction by 0.0004" each reversal, not that the motor is going the wrong way.

CarveOne

[dumb_luck]

It sounds like mechanical slippage in a motor coupling or a motor shaft to pulley connection is slipping only in one direction by 0.0004" each reversal, not that the motor is going the wrong way.

CarveOne

If I reverse the motors wires on the control board the error is in the opposite direction. That is why I feel it is electrical and not mechanical. If it were mechanical it would not matter what direction the motor was wired. when I jogged back and forth the error would always be in the same direction.

[RomanLini]

It could be edge-triggered noise on the STEP pulse signal.

Can you try the setting in Mach3 that inverts the STEP pulse from normally HI to normally LO? That may fix it, and even if not fixed if that reverses the direction of error again you know for sure that you have a problem with the step pulse.

You could also put a small capacitor from the STEP input in the driver to GND, a value of 1nF (0.001uF) should do. That can clean up any edge noise on the step pulse. If you had access to test equipment you could put an oscilloscope on the STEP pulse signal and look at the waveform.

[dumb_luck]

I don't have scope and I'm not sure where to find one in my little town. I have tried switching the pulse to active low it did no change the direction of the error. I'm not sure what that means

[tpworks]

I had a problem gaining steps and it was due to line noise on the parallel line at the breakout board, some filter caps cured the problem.

http://www.cnczone.com/forums/410814-post11.html

[dumb_luck]

Well I'm very sorry it took so long to get back to everyone on this. I did resolve my problem I replaced the cncgeeker breakout board with one from keling cnc and it did the trick. the machine is running great, backlash of .0005 or less not sure my gauges are accurate enough. running smooth at over 140 inch per minute.