| | |
Hi the problem with my nm-200 was the stepper on the x axis we replaced that and then we found a bare wire grounding rite at the metal 90 then every thing was good then started acting up so i checked the wires all the way back to the computer an the one wire 1b- of the x axis was working its way out oh the holder at the back of the controll panel .
every thing is working good for the last 2 days of use and i keep an eye on the axis couplings they seame to come loose every so often.
other then that problem i like the machine and the support that i get is great thanks khai,rodney ,ron, john and jeff
Check that your motor steps are correct. You can do this by measuring the travel with an indicator. Do this on all axis. The XML profile may not be correct for your motor steps right out of the box. Or perhaps the gibs in the X or Y direction are too tight?
Bob
Oval circles are caused usually a Backlash or loose motor coupler or thrust bearing. Also check where the ball nut connects to the axis.
(BTW, Ihavemofish, if it was a step pulse problem, the end of the circle would't meet up with the start.)
If your rails are rusting, change to a coolant with more oil content. I use DoAll 470 coolant.
Larry K
Manufacturer of CNC routers and Viper Servo Drives
www.LarkenCNC.com and www.Viperservo.com
actually it would, because it looses the same amount of steps on each side. on a more complex program it would be out of position more noticably.Originally Posted by Larken
its an issue i am very familiar with, one that is acknowledged by gecko, and one that i have fixed completely on 3 machine setups using sherline mode.
its the first and easiest thing to check before going off to tear things apart on the machine. discounting it doesnt make any sense.
backlash of course could also cause a small flat spot on a circle, but not an oval one.
Loss of steps can make an oval ? ... I doubt it.actually it would, because it looses the same amount of steps on each side. on a more complex program it would be out of position more noticably.
Manufacturer of CNC routers and Viper Servo Drives
www.LarkenCNC.com and www.Viperservo.com
depends how its loosing them.Loss of steps can make an oval ? ... I doubt it.
point is, its the first place to look, and in my experience, the usual cause. backlash on these machines is insignificant and wouldnt leave a very visual flat spot on a circle.
This is excellent info. Thanks fish!
I'm excited to try Sherline mode - I've been going out of my mind trying to figure out why my mill seems to have shifted .001 or .003" in the X after doing some cutting. I need accuracy because the part gets flipped to cut the bottomside, and the mis-registration between top cuts and bottom cuts is causing alot of scrap.
Out of desperation, I was going to add a gradient wheel to the X leadscrew, so I could tell definitively if steps were being missed - I still may do this...
easy way to test. set the jog speed to .1% and listen to the motors.
other things to try might be running the program faster or slower and seeing if the amount of lost position is the same.
.003" could be backlash of course, but as i said, software is so much easier to check first
It's not backlash (checked @ .0007" , and accounted for).easy way to test. set the jog speed to .1% and listen to the motors.
other things to try might be running the program faster or slower and seeing if the amount of lost position is the same.
.003" could be backlash of course, but as i said, software is so much easier to check first
Tried running the program fast/slower, including rapids and acceleration, doesn't seem to affect things.
To definitively answer the riddle: "am I losing steps?" here's what I did:
I attached a protractor to the X Leadscrew; the indicator spins wildly as the axis moves, while the part that says "PROTRACTOR" remains firmly epoxied to the pillow block. It is not hard to read an angle less than 1 degree; with my 4mm pitch leadscrew, 1 degree is .00044" of linear movement - so lost steps totalling more than .001" should be easy to see.
The idea is to type G00 X2, read the angle, then do some cutting, then type G00 X2, and read the angle again. If the angles differ, then, positively, steps have been lost. Or my epoxy failed. Or that little screw holding the indicator slipped.
I'll let you know what happens....
That should work.
The test I'd do is run left/right rapids with a dial indicator off the column zeroed on something on the table, it should never read higher than the backlash.
You can usually hear a stepper miss.
the software lost steps issue will only show up in slow motion, so:
g1 x2 f0.1
Posting Permissions
| | |
LinkBack URL
About LinkBacks
