Mini wannabe datron build

lcvette one, sorry

i tried the update / upgrade, found nothing new.

So where do i find this elusive real time kernel? There doesn't seem to actually be one for 12

forum.linuxcnc.org/9-installing-linuxcnc...-ubuntu-23-04#269653

thanks ill check that.

However, when i look at the package manager, it DOES have preempt rt installed already.... (6.1)

rt kernel is a red herring here i think.

"X Error of failed request: BadWindow (invalid Window parameter)"

This is the error. it must not be installed correctly. I will start again after lunch with a new install. if anyone has a complete instruction set with no missing elements, please point me to it, cause clearly some steps are missing in the lcvette method.

Pretty!
Has a DECI sticker, so assume that's the "real" manufacturer. It is always hard to tell from these sellers. I need to make some spacers for the mount as it hits the servo. just some 1/4" drilled holes in some ground 4140 so no big deal. Still swamped with some other things though so I wont get back to this til next week.

 

So, I have an issue. The machine seems to be losing position in the X over time. Not in every program, but in one it is doing it by a LOT. over a mm over 13 mins of cutting.

The load is light, and I have ruled out anything mechanical - coupling is not slipping, no play in the axis.

So that leaves:

1: Servo
2: Software.

Servo I *think* is unlikely. It errors correctly at 20 pulses out, which is only 0.05mm and way smaller than the error I am seeing. Could be a glitch where is misses reading the steps I have seen in the past with mach3 an issue where the pulses are not being sent or read correctly by certain drives. I have never seen this with linuxcnc, and it was fine under the DDCSV3.1 control.

Anyone have any thoughts on where to look, what type of test to perform? I am a bit at a loss and need to actual mill a run of production parts on tuesday. Eep!

changed the servo setup and hold to 20000 from 10000, but nothing changed. I got one really bad out of position where the x "moved" several mm. hard to say if it moved too far, or stopped moving and "crashed" as a result. in any cause it made a bad noise and blew through the part wall.

it seems to happen in almost the same spot and same amount.

Here is the motors suggested timing.
I have a suspicion the motor/drive went bad but i do not have a spare to test and swapping 2 axes will be bothersome. its only the one axis. it was fine yesterday I *think*. Not that anything really changed though.