[txpto]

Hi, I'm a little problem.

When mill only in X direction -> perfect -------
When mill only in Y direction -> perfect -------

When mill in x,y direction (diagonal) -> /\/\/\/

Fix this foto for view.
If anybody have solution...please mail me
[email protected]

Use CNC Amateur with MACH3. No Backlash.
Thanks a lot.

[fizzissist]

Sure looks like a bad parameter setting to me!
Something is set to move only one axis at a time, in little increments too.
If the command line has both X and Y coordinatates ....I'd suspect parameter setting.

Are you programming this manually or with CAM? If using CAM, check your preferences or configuration.

Almost looks like a series of small arcs?

Post some code so we can see what's going on.

[Fifty_ohm]

Could it be a bent screw? Look at how the waves are lined up with the waves directly above and below. I'm not sure how that would work but they seem to be repeatable by location. Maybe the nut dragging on a bent screw. But I'm just guessing. When you get it figured out be sure to let us know what the cause is. I'm really curious.

[ger21]

How fast is your PC? Check your CPU load while cutting.

[txpto]

Excuse me for my english.

a simple code G:

G01 Z25
G01 X0 Z0

G01 Z-2
G01 X10 --> perfect, draw a rect line without problems.
G01 Y10 --> the same situation. all ok.
G01 X0 Y0 --> problem.... line not perfect diagonal.


PC is a good PC (dual core 2 duo 2Ghz, 2Gb RAM,Windows XP)
Charge of CPU 51% in MACH3 when milling.

Software of control MACH3

Is not posible backlash because the movement is in the same direction X+ and Y+, no there are movements in oposite direction.

Mecanichal is good, because is a proxxon table XY modified.

If can help me, I be happy with a solution.

Thanks a lot.
Charles

http://blog.txpto.com

[Delw]

Theres a few things
ball screws really sloppy or in a bind
your motor feed back maybe off
your motors dont move together at the same time ( controller problem)
the noise from one motor can be affecting the other motor ( controller problem)

Try making a circle and see what it looks like

[txpto]

I have XY axes with passage of 2mm, ie 1 = 2milis per round.

I put the comparator in advance axis X. .. 1 millimeter and everything ok.

I put the comparator in advance axis Y. .. 1 millimeter and everything ok.

But ....

Fix the Feedrate to 10%, that is suuuper slow. (to timing)

I am taking the same test ....

I put the comparator in advance axis X. .. 1 millimeter and everything ok.

I put the comparator in the y-axis. Avanzo 0.5 millimeter and I takes 2.5 seconds.

I put the comparator in the y-axis. Avanzo 0.5 millimeter and takes me 2 seconds.

Conclusion:

-- Not because endiablé reason, the Y axis does not have the same step in the entire journey. the bolt problem is not because it is repetitive each 1 / 2 back.

Deduction: The nut Y axis is wrong screw. Have more step to the principle that in the end.

I think: how we can make it with a different way, I was not the key screw Y axis? easy .... as a backlash has allowed a game inside the nut.

Solution: NO, which is what I most CABRE. I can not tell the MACH I do two different speeds each 1 / 2 laps shaft.
I parameterize a curve around the axis (correction shaft), but not in the nut itself.

The problem with stamina and learn to live with it.

Never will I leave smooth surfaces that do not correspond with the x-axis alone or with the y-axis

Moral: Do not cheap prefabricated buy tables. This demonstrated that eventually emerging sides.

Greetings to all and thank you very much for the help.