z axis being an arse

[daniellyall]

have problems getting z axis to be accurate had to change ball screw and stepper motor what has made it better but the velocity is 310.3, acceleration is 220 the Z is within 0.02 mm
I run 1000 moves plus the z will be 194.85 then after 000 of moves 194.58 then 000 of moves 194.79 and so on
help please
I have been at this for 3 days tried all different acceleration values its not coming right

Similar Threads:

• Unwinding the rotary axis ( 4 axis rotary axis milling fanuc control)
• Need Help!- 4th axis rotation on x axis chaging to Y axis in simulation
• calculating pulse equivlent factor for a rotary axis in place of the x axis
• 5-axis, 4-axis, and 3-axis CNC Router manufacturer

[neilw20]

Is that measured or on the display?
A mechanical problem (loose coupling,backlash,stiffness) or microsteps too coarse.
Pictures? What is it?

[awerby]

It sounds like your acceleration is very high. If you're having problems, reducing that is the first thing to try (well, second, after checking for mechanical problems.) Start with an acceleration about 10% of your velocity value (which would be 31, not 220) and if that doesn't work, cut it in half. Once you get it running reliably, then try making it faster by inching up the acceleration.

Andrew Werby
www.computersculpture.com

[daniellyall]

tried from 10 to 300 20 was OK it was the closest but was still going up and down all over the show but far to slow 200 works just as good as 20.
there is no mechanical problems fix all off that the z axis go up and down very easy with the belt taken off, it falls down if its up high.
run 200 for 1000 moves it was -.012 down from start height then 1000 moves more it was -0.02 from start high then it was -0.025 from start height then -0.11 from start height then when back up to 0.01 of start height so you can see it goes up and down by very little but never goes back to start height it seems to stay with in 0.2 mm + or - of start height. is this normal or just this machine

[neilw20]

What sort of machine. Show us a picture pleas.

[awerby]

Are you sure that something isn't slipping? If you have the leadscrew couplers attached to the motor with set-screws, for instance, they can move a little without you noticing. Try making marks with a felt pen over any connections like that and running the axis up and down for a while. You should be able to see any slippage.

Andrew Werby
www.computersculpture.com

[daniellyall]

its a 1200 mm x 2400 mm x 150 mm router made in china

[warrenb]

I'm guessing mechanical problems. Check the fasteners and the coupling. I had the same problem on a servo drive coupler that drove me nuts because it "appeared" to be just fine and I couldn't figure out what was causing the problems because they weren't consistent. .

[daniellyall]

ill recheck that every thing is nice and tight.
will post results.

[neilw20]

Put a dial indicator under the spindle and push up and down, and disturb the X and Y axis loads.
Might be some movement there. What vertical restraint in the X and Y?
A photo or a model number and a link would go a long way to helping you.
What is the smallest z step that makes it move?
Do you have linear guides on all axis? Your tiny picture on webs.com doesn't show much.

[daniellyall]

its got linear rails on all axis`s x an y on r & p, z axis on ball screw (harwin) and fancy bearing mount, all axis belt drive, all axis have nema 32 frame motors gecko g213v on all steppers, MB-02 V6 breakout board. have attached picks of machine.

[neilw20]

Does that motor have air bearings?

[daniellyall]

its a 3k water cooled spindle

[awerby]

It could be RF interference from your spindle. Have you tried dry-running the process with the spindle turned off? High frequency spindles give off a lot of random noise from their cabling; this can be particularly problematic if the spindle cable is run alongside the motor control cables. It should be covered with a braided shield that's grounded to the motor side, but not to the other end. Here's a thread that discusses HF spindle grounding in more depth:

http://www.cnczone.com/forums/genera...y_spindle.html

Andrew Werby
www.computersculpture.com

[daniellyall]

all tests where done with spindle off.
all wires that need to be are shielded.
everything that needs to be done to make machine a very good machine has be done, but z is still is being a pain that's the only problem we have with the machine. thanks for suggestions everyone give me more suggestions no suggestion is stupid.

[thething84]

have you tried swapping an axis motor with a know working one to rule out the stepper playing up

[daniellyall]

did that and it was part of the problem its still out by a little bit there where 2 lose nut on the bearing housing every nut an bolt on z axis is very tight now still just slightly out have the acceleration back down to 30 now