Hello Guest it is October 20, 2019, 03:43:35 PM

Author Topic: Seeking some help on "drifting" X axis  (Read 1232 times)

0 Members and 1 Guest are viewing this topic.

Seeking some help on "drifting" X axis
« on: August 28, 2018, 09:59:12 AM »
We recently got a table second hand that had only been used a couple times. We got this table from an estate sale so the past user is....of no help. It literally only had a couple cuts on it based on the tray below the work piece.

First few cut files we ran were fine, but they were very basic shapes of plate (square, triangle, rectangle). Recently we tried cutting a file that had a series of holes in it and it seems to cause the X axis on the machine to "drift" to the right. In the picture below, this was the cut file run with a marker and cardboard to really see what was going on. The file was run 3 repeated times with each run time started from the finish position of the last run. You can see in the circles that it DOES NOT drift a consistent amount or with a consistent pattern each time-the hole overlap is not the same (thinking it rules out motor step count is wrong in setup as it would be a consistent error). We already swapped X and Z motor output cables from the control box to the motors and reassigned the pin locations on the parallel cable to reflect that-the problem stayed in the X. Also threw a new parallel cable at it just to check.

The computer is an older gateway running XP and IS NOT connected to the internet in any way
Using sheetcam for cam and nesting and Mach3 licenced under CandCNC for control.
The table is a Precision Plasma LLC, kit table that someone put together with a Gecko g540 board.

Open to any thoughts at this point!!

Re: Seeking some help on "drifting" X axis
« Reply #1 on: August 28, 2018, 02:54:47 PM »
Hi,
your X axis is either gaining or loosing steps.

Go though and double check that the couplings are snug and the stepper motor shaft is not slipping.

Can you swap drivers, say take the z axis driver out of service and use it in the X axis drivers place?

Craig
My wife left with my best friend...
     and I miss him!
Re: Seeking some help on "drifting" X axis
« Reply #2 on: August 28, 2018, 03:02:53 PM »
Hi,
re-reading your post it already sounds like you've tried that.

Craig
My wife left with my best friend...
     and I miss him!
Re: Seeking some help on "drifting" X axis
« Reply #3 on: August 28, 2018, 03:05:17 PM »
Hi,
your X axis is either gaining or loosing steps.

Go though and double check that the couplings are snug and the stepper motor shaft is not slipping.

Can you swap drivers, say take the z axis driver out of service and use it in the X axis drivers place?
 

Craig

Thanks fort the thoughts.  Everything was thoroughly checked when we got the table so I'm pretty sure the couplings are all tight/snug.  From what I can tell, the stepper motor shaft is not slipping either.  Are you reffering to it slipping at the gear or internally in the motor?  

We did essentially swap drivers also.  We swapped output cables for X and Z axis at the driver box (connector out of the driver box that cable to the motor plugs into) so the X motor was plugged into the Z driver and vise versa, then reassigned those drivers to the new axis via the pin assignments on the parallel cable.  The problem stayed in the X.  Ever stepper motors randomly loose steps when they begin to fail?  These motors have virtually no run time, but coming from the automotive field, I have seen plenty of new parts fail right out of the box.  
Re: Seeking some help on "drifting" X axis
« Reply #4 on: August 28, 2018, 06:50:37 PM »
Looking at your photo it can’t be a step loss problem because the circle centers are moving closer together and yet they are still round! Look at your G-code because it has to be the problem.

Offline RICH

*
  • *
  •  7,367 7,367
    • View Profile
Re: Seeking some help on "drifting" X axis
« Reply #5 on: August 29, 2018, 06:34:05 AM »
Post your gcode file. A quick backplot of the code  will reveal if it's a code problem.
Post your xml file you are using.


RICH
« Last Edit: August 29, 2018, 06:55:47 AM by RICH »
Re: Seeking some help on "drifting" X axis
« Reply #6 on: August 29, 2018, 07:17:55 PM »
Maybe I should clarify.  The location the marker is currently in is the end position.  I ran the program 3 times, without changing home or anything on the table.  In other words, these 3 patterns should be drawn directly on top of each other instead of off setting each time to the right.  The initial home location is at the lower left corner of the square.  All the circles were created from the "bottom" of the work piece to the "top", so the X is essentially "drifting" to the right as it moves up the work piece cutting the circles.  Then when it starts the outside edge, it is further to the right than it is away (downside of no encoders).  It does not however have any issue cutting the square.

As for pasting the Gcode in, I am a fabricator that wound up with a plasma table and is learning as I go.  I'll get it up here but have to move it from one computer to this one (we do not connect the XP machine to the internet).
Re: Seeking some help on "drifting" X axis
« Reply #7 on: August 29, 2018, 09:09:19 PM »
Still looks like a G-code problem to me. Actual axis drift would have resulted in ovals not circles.  It sounds like your program might have generated in incremental code rather than absolute and all the increments plus and minus don’t add up to zero. Incremental is common for nesting because you can put the motion code in a subroutine and move to a new starting location for each instance.

Offline ZASto

*
  •  178 178
    • View Profile
Re: Seeking some help on "drifting" X axis
« Reply #8 on: August 30, 2018, 09:02:18 AM »
This looks definitely as G-code problem (variables?).
First column of circles is aligned in Y direction, second gains In X direction, as well as third column, more than second. And, pen does not return to origin :).
Incrementing definitely, in code.
Make no mistake between my personality and my attitude.
My personality is who I am.
My attitude depends on who you are.
Re: Seeking some help on "drifting" X axis
« Reply #9 on: August 30, 2018, 10:43:31 AM »
This looks definitely as G-code problem (variables?).
First column of circles is aligned in Y direction, second gains In X direction, as well as third column, more than second. And, pen does not return to origin :).
Incrementing definitely, in code.

Can you put the incrementing definitely a little more in laymans terms?  And if there is an issue with the gcode, would it mean there is an issue with our Sheetcam being that is what we convert DXF from our cad software over to gcode with?