Just finished converting a lathe. Gecko's, Nema 34's. Mach3 Lathe.

I turn a diameter and dial in to the correct size and the next part is perfect. THEN! the next part is .001 larger than the first. Hit cycle start again and the third part is .001 larger than the second. Now I know your thinking i'm loosing steps but I have the rapid speed way down and acceleration ramp looks like a triangle. Z axis is a 1200 Oz and X axis is a 600 Oz motor with a 2:1 belt drive. Way more motor than I need. The weird thing is the error is exactly the same every time. Is there a setting in Mach Lathe that could cause something like this to happen? Could the Gecko be configured wrong? Anybody ever heard of something like this happening?

stevehuck,
Frankly, the actual incremental change you are looking for is 0.0005".
The big picture question is ....Is it software, the lathe axis system, or setup along with actual machining the cause?

I am curious if the change is the same for reducing or enlarging a diameter.?

I suggest you do nothing other than do a simple test and then only make ONE change at a time
and see what happens for each change.
A gcode file for reducing and enlarging stock using the same tool, pullout distance, etc / keep it simple.
Air cut only and no need to run the spindle. You can setup to actualy measure and compare to DRO display or just visualy check the tool position.  

RICH

Words to the wise I'm sure but if you're using a changing parameter value for turning the diameter, make sure it is zeroed (or returned to a fixed, favourable value) every time the code runs. Otherwise it will persist and keep incrimenting itself over and over.

don't know, maybe this will be helpful.  I had a fanuc driven engine lathe, start making parts in the morning and I would see a small change, maybe .0002 on each consecutive part.  a little worse if the shop was cold.  go to the shop a couple hours early in the morning, put the lathe in a warmup program, and go back in the house and watch the news and drink coffee.  after that, the parts would usually repeat perfectly.

Thanks for the reply's. Yesterday the lathe really went haywire. It would run good in one direction and then chatter and make really bad noises in the other. Then it quit working all together. I suspect the breakout board had a lot to do with it. I'm thinking the step and direction signals were not getting passed along thru the BOB. I ordered a new one and will see it at the end of the week. If that doesn't improve anything it will be new Gecko's. I run a mill with the same computer so I am sure that's not it. It was running beautifully a week ago.

I'll keep you posted.

don't know, maybe this will be helpful.  I had a fanuc driven engine lathe, start making parts in the morning and I would see a small change, maybe .0002 on each consecutive part.  a little worse if the shop was cold.  go to the shop a couple hours early in the morning, put the lathe in a warmup program, and go back in the house and watch the news and drink coffee.  after that, the parts would usually repeat perfectly.

It is cold here. I will consider the temperature after I get it back up and running. Thank you!

Just to follow up, it was the breakout board that was causing the problems. I replaced it today and ran 100 pieces. The first piece was the same size as the last.

Great news  Happy turning and merry christmas.