Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: Cuemaker on June 30, 2013, 06:49:14 PM
-
For some time now I have had a problem with my A axis on a clockwise turn getting a error.. I have changed the gecko, breakout pcb, parallel port cable, computer, reloaded new versions of mach3, erased the hard drive and reloaded everything, took off all unnecessary programs, swapped the gecko cables, swapped the servo motor with a new one, took the motor off the assy and ran by itself with (G0 a40000) check light... run (g0 a-40000) ok... I have a K2 cnc and they have no clue what could be the problem... Anyone had such a mess...??? Sometimes it works ok......Re grounded box and no help...I thought about going to smooth stepper, but don't want to spend the money if it won't help.... Some of the runs I do set (A) as much as 212,000. I tried changing kernel speed, no help. Motor speed, no help.... H E L P if U can... I am not stopped from working because I can run in CCW directiion, but I have set up to be able to run in CW most of the time...
-
What error are you getting?
Post your xml file that you are using.
RICH
-
This is my XML... red check light on control box...
-
What does the light signify?
This site is to do with Mach3 software and there are lots of different machines set up with it. Unless someone is familiar with your particular hardware then you will not get much advice unless you can elaborate a bit ;)
Hood
-
K2 says any chk light is a servo fault of some kind... that is all I know... I have talked to K2cnc several times about this problem.. Maybe a total new controller is in order... everything works.. wrapping vcarve and aspire files works fine, just when I run long taper cuts or straight rounds in the (what I call) clockwise direction... Of course I ran 4 runs today... 29" @ .05 cut per inch and it ran perfect.... Who knows... ?? not me.... I spent money on servo, gecko, breakout pcb, computer etc...What the significance of only one direction error is beyond me...
-
Ok servo error sounds like a tuning issue, ie it is exceeding the following error limit. As I have never messed with Geckos servos I cant really help on the tuning I am afraid.
Hood
-
Hood, there is no limit on the A axis.... It is set to go and go and go.....
-
It is the servo drives following error limit I am referring to. Servo drives need to be tuned so that they get as close as possible to the commanded position and keep as close to the commanded position whilst moving.
Servo drives have a value called Following Error, if the axis exceeds this value (ie is out of position by that or more than that amount) then the drive will trip. It sounds as if your tuning is not too great and it is close to the limit but maybe just more friction one way than the other and that is enough to push it over the error limit.
Hood
-
I have set and reset the tuning. Lowered velocity and accel, raised velocity and accel... set as per K2. Doesn't seem to help. I worked 25 years as a computer technician so my knowledge base is more than basic, (I think anyway) . I thought as you that it had to be in tuning, but have not had any reason lately to think so... I even put a higher rated servo on the cables and it also failed after tuning... I thought It might be the code disc, so bought new style servo w/o code disc. exactly the same... Think I might just be stuck with it faulting (sometimes)...
-
Are you talking tuning in Mach or tuning of Gecko drive to motor?
I am meaning the latter.
I have experience of tuning industrial grade AC servos (thats what I use on my machines) but afraid I have no experience of Geckos or the like. I have heard people say the tuning options are rather poor and to get the best you really need a scope but as said I have never used Geckos so I do not know from first hand experience.
Hood
-
right... I have been tuning Mach3, not the gecko....
-
After much work rechecking, I believe I have figured it out... I had not set a axis to reset on home... that caused the counter in mach to keep getting larger... I believe a overrun of the field occurred... now that I go to reset the a axis coordinate field it seems to run ok... weird??? or not