[alevitt53]

Hello,
I've built a 30x40 CNC loosley based on CNC Router Parts unit. I'm running nema 23 motors running an x and slaved a axis with ball screws.
Running Mach3 and a g540 driver.
The bottom line is...if I run a quickie little program to cycle the x axis back and forth about ten times, the two sides are no longer aligned with each other.
The X axis is no longer aligned with the A axis so therefore the gantry becomes skewed. I've check to make sure the settings in Mach are the same for each axis and
checked all the electrical connections. Mechanically, as far as I can tell, nothing is binding on either side.
I'm at a loss to further diagnose the problem as I feel I've tried everything I could think of.
Is it possible the motors are not "exactly" the same?? Hard to believe....
Any suggestions would be greatly appreciated.
Thanks
Art

[MadeInTheShed]

Sounds to me more likely you are missing steps.
Try slowing your velocity and accelration right down - half or even a quarter to what you have it set at now. Does it still do it?
You could try disconnecting one of the ballscrew nuts and driving it with just one screw. Run your program and test if it returns to the same place. Then try the same with the other ballscrew. To see if the motors / mechanics are the same, while you have one ball screw disconnected, tell it to move a certain distance and then measure that distance accurately. (Mach3 can do this for you on the diagnostics page I think) Is it the same distance on the other side?

[handlewanker]

Hi, I'm building or will be starting soon a build for a moving table router or bridge mill with 30X25 work area and chose to design it as a moving table type mainly due to this problem of gantry slewing.

It seems you are having the problem all twin screw drives are susceptible to and if not apparent then you are lucky, but as you are having the gantry slew then you could try synchronising the screws with a twin toothed belt drive from one drive motor in the middle.

This means you only have a problem with missed steps if you drive too hard etc, but it will be evenly applied to both screws so the gantry cannot go out of square ever.

A single screw in the middle will also allow the gantry to slew when the spindle is at the far right or left of the table, hence the leaning is to twin screws or rack and pinion, but without any form of synchronisation it is a step full of problems as you have found out.

I think that if a passive twin rack and pinion were fitted, that is just a pair of racks on either side and two pinions coupled together on a common shaft not driven, that just move with the gantry sides but give them a synchronised drive to both sides of the gantry when the single central drive screw drives and the spindle is to one side or the other.

If one side of the gantry tries to move ahead of the other side the rack and pinion will prevent it from going out of square.....one side then drives the other etc. and you only need one central ball screw and stepper motor.

The cost saving alone would be a couple of hundred dollars......as in a ball screw and nut, ball nut housing, ball screw bearings each end, stepper motor and mounting bracket, coupling and a reduced power supply against the cost of two racks, two pinions, two bearing housings and a shaft.
Ian.

[Dragonfly]

@alevitt53
Are you saying that you drive each ballscrew as a separate axis (X an A)? And how did you even manage to make it under Mach3?
If this is the case it is not correct. You may use two motors with ballscrews each with its own motor driver but the step pulses must come from one pin for a single axis. Just feed them to both drivers which must have identical step, currents, etc. settings.