Backlash

[Ed_R]

So how do you get rid of it? I've got a lot in my x Axis.

[Max-DK]

Hi Ed_R

It really depends, on what kind of screw you are using. We need more info.

[Ed_R]

As far as I can tell it's an ACME thread going through arubber grommet. BUt the backlash is somewhat inconsistent-it's a different amount every time. No idea what's going on there.
THe other two axis are the same design but no backlash.
Yet.

[Max-DK]

Are you sure that the screw cant move, so that the backlash is not comming from the screw mowing back an forth. You could try to put a dial gauge on the end of the screw, just to see that theres no play there.

[Ed_R]

Not sure wha tyou mean there. THe crew moves when teh motor does, but the carriage doesn't follow along at first. I may have to get a replacement grommet.

[Max-DK]

If the screw isn't propely attached in one of the ends, and when it start to either push or pull the table, the screw can do the movement in the x axis instead, before the axis starts to move.

Hope this makes sense :-)

[Walt@SGS.Inc]

I believe what they are trying to tell you is this.....
Mount an indicator with the stylus against the end of the threaded
rod. Mount the base of the indicator on the base of the machine.
Then move the axis in one direction a few thousands then set a zero
on the indicator, reverse the direction of travel and measure the
amount that the threaded shaft moves or "chucks" . It could be
the preloading of the lead screw is not tight enough. If no movement
is noted with this method, mount the indicator base on the axis in question
and arrange the indicator so the stylus is in the thread of the lead screw.
then increment the axis in one direction .0001" at a time until the reading
remains constant. Then reverse the movement while watching the indicator.
This would tell you the nut is loose of the screw or the nut is not tight to
the axis. You have to isolate the problem before ytou can fix it. May be a
loose motor coupling or something like that. Best of luck..
Regards Walt...

[Ed_R]

I'll see if the screw is rotating in the mounts before it moves the carriage. It's definitely rotating before the carriage moves, so it's either a motor mount or the rod mount or the tapped thing that rides the thread attached to the carriage, right?

[Walt@SGS.Inc]

If you use an indicator, you should be able to isolate where the lost motion
is coming from. If you can see the screw turning before the slide is moving,
you are half way to the problem. Just find where the lost motion is at.
If the screw is turning and the slide is not moving, you have to determine
if the looseness is in the nut, in the screw mounting, or where the nut is
secured to the slide assembly.
How much "lost motion are we talking about and how much is there in the
other axis or axies. However you spell it or them (1 or 2)
Regards Walt.

[Ed_R]

What I just did was to move the X axis close to the end of its travel , measure a one-inch gap between the carriage and the stop with a digital caliper. Then I zeroed the x axis at that point, then sent gcode to move it to -.5- and it went to -.483. Repeated measurements fell in the .017 to .020 range. The screw is defintiely moving with the motor, and the carriage is not. The nut is mounted to the carriage securely, and I believe the nut is where the problem is. I could be wrong but everythign else seems to be operating normally. The Y axis has the same nut and thread types and .040 consistent backlash. I didn't measure the z yet. I suspect it's the nuts in both directions because everythign else is tight.
Replace the nuts?

[Walt@SGS.Inc]

Depends on what the back lash should be but I would think .001 would
be closer to what you want. I work mostly with CNC Metal Milling
machines and backlash or lost motion should be under .0002".
Thats right .0002", that is 2 of them little ones.
You would have to be the judge of what is acceptible.
Regards Walt...

[Ed_R]

I have no measurable backlash on the z axis, and after backlash compensation in Turbocnc I have around .002-.005 measurable on x and y, That ought to be close enough for what I need to do, I think.
I hope.

I'd rather have 0 backlash but this will work for now.