Problem when trying to run Autoleveller on Mint 19.3 + Linuxcnc.

More
26 Nov 2020 14:48 - 26 Nov 2020 15:28 #190337 by aparecido
Hello, I did some tests just now and had some problems.
The probe and port are configured correctly in Linuxcnc and it sees status change 0 and 1.
When I ran gcode-ripper **. Py via terminal, these messages returned:
Exception in Tkinter callback
Traceback (most recent call last):
  File "/usr/lib/python2.7/lib-tk/Tkinter.py", line 1550, in __call__
    return self.func(*args)
  File "g-code_ripper.py", line 2068, in Entry_recalc_var_Callback
    self.menu_Mode_Change()
  File "g-code_ripper.py", line 2294, in menu_Mode_Change
    self.Open_G_Code_File(self.NGC_INPUT,Refresh=True) #V0.8 needed so auto arc2line is active during reading
  File "g-code_ripper.py", line 2114, in Open_G_Code_File
    if error_message(MSG) == "ignore":
  File "g-code_ripper.py", line 180, in error_message
    error_report.grab_set()
  File "/usr/lib/python2.7/lib-tk/Tkinter.py", line 690, in grab_set
    self.tk.call('grab', 'set', self._w)
TclError: grab failed: window not viewable
Exception in Tkinter callback
Traceback (most recent call last):
  File "/usr/lib/python2.7/lib-tk/Tkinter.py", line 1550, in __call__
    return self.func(*args)
  File "g-code_ripper.py", line 2077, in Entry_units_var_Callback
    self.Open_G_Code_File(self.NGC_INPUT,Refresh=True)
  File "g-code_ripper.py", line 2114, in Open_G_Code_File
    if error_message(MSG) == "ignore":
  File "g-code_ripper.py", line 180, in error_message
    error_report.grab_set()
  File "/usr/lib/python2.7/lib-tk/Tkinter.py", line 690, in grab_set
    self.tk.call('grab', 'set', self._w)
TclError: grab failed: window not viewable

When I select an option, it returns an error message, when I open a gcode file it also returns an error.
What am I doing wrong? From what the command returned, I have Python 2.7. * Installed ...

I attached the gcode files I used for testing, I generated them to machine a pcb from a layout made in Eagle and exported by the ULP "pcb-gcode.ulp".
Attachments:
Last edit: 26 Nov 2020 15:28 by aparecido.

Please Log in or Create an account to join the conversation.

More
26 Nov 2020 18:37 #190346 by jbraun
Remove Q.001 from the ngc file. I think if you remove the G64 line entirely it won't matter as g-code_ripper will add it's own G64 according to it's general settings.

from the user guide description of Naive Cam Detector:
Successive G1 moves that involve only the XYZ axes that deviate less than Q- from a straight line are merged into a single straight line.
www.linuxcnc.org/docs/html/user/user-con...c:trajectory-control
Yeah, clear as mud to me as well :)
The following user(s) said Thank You: aparecido

Please Log in or Create an account to join the conversation.

More
27 Nov 2020 16:06 - 27 Nov 2020 23:06 #190457 by aparecido
I followed your suggestion and changed some things in the pcbgode configuration and the errors are gone.
But now another problem is occurring, I do the whole procedure in Gcode Ripper, it saves the file to open in Linuxcnc, Linuxcnc makes all measurements, but the last lines it does not execute:
G0 Z5,000
G0 Z5,000
G0 X66.778Y103.832
(PROBECLOSE)
(G Code)
(G-Code Postamble)
M2
It stops running gcode as if the last line of code had arrived, I tried to tell the software to run from that point and it also didn't work.
As I understand it, he does not save the file for Gcode Ripper to adjust with the surface of the part to be machined.

>> I solved the problem above, Linuxcnc did not have all permissions enabled, I enabled and now generates the txt file, but it only generates in the Linuxcnc installation folder, there is no way for me to choose where it will be created.
Last edit: 27 Nov 2020 23:06 by aparecido.

Please Log in or Create an account to join the conversation.

More
28 Nov 2020 06:25 #190486 by jbraun
Good to see you have this solved.
Please ignore the suggestion in my previous post of removing the G64 line entirely. G-code_ripper doesn't post G64 by default.
I'm not sure how to change where the probe data file is created. This is the probe.ngc in case someone else knows a way.
( G-Code Modified by G-Code Ripper )
( by Scorch - 2013-2019 www.scorchworks.com )
G20   (set units to inches)
G90
(G Code)
(PROBEOPEN probe_points.txt)
....................
a bunch of co-ordinates and G38.2's
........................
(PROBECLOSE)
(G Code)
(G-Code Postamble)
M2
The following user(s) said Thank You: aparecido

Please Log in or Create an account to join the conversation.

More
28 Nov 2020 14:56 #190519 by aparecido
So, I realized that if I leave the G64 line, the error.
I put it manually so that Linuxcnc doesn't 'round' the vectors and perform machining exactly like the drawing.
I can just leave without the G64 line and run Gcode Ripper, fix the gcode according to the surface read and after adjusting the original gcode, I insert the G64 line.
I see if I can find something on how to change where the TXT of the probe file is saved.
Thank you again.

Please Log in or Create an account to join the conversation.

More
08 Dec 2020 15:38 #191427 by aparecido
I was able to use the gcode ripper to correct the imperfections of the board's surface, but now there are other problems, but I believe that with Linuxcnc 2.8, so as not to be creating several topics here on the forum, I'll post it right here.
1-I give the command for the machine to go home on the 3 axes, on the X axis it goes to "0.0" which is the configured position of the sensor, on the Z axis it goes to the position of the sensor which is "26.0". However, on the Y axis, the sensor position setting is 0, but it finds the sensor and marks it as "-2.691".
2-When machining the pcb, it seems that the machine is losing pace and makes some overlapping tracks and some holes in the wrong position, however I lowered the speed of the machine well, checked the board (CNC3AX) and sensors, the power supply is not overloaded , the mechanics of the machine are also ok.
I give the command to do homming on the 3 axes and have it run again from a certain point in the gcode, where there is already a problem in the machining position and the machine runs 'with error'.
So I'm guessing it might be a problem in Linuxcnc 2.8.0 interpreting the code, or maybe it is the computer that has a problem with latency? Although I did the tests that I even posted here on the forum.
I attached photos of the board and the gcode.

Please Log in or Create an account to join the conversation.

More
08 Dec 2020 16:10 #191429 by tommylight
1 it looks like you have active offsets as can be seen from the green round/spiky thingy where the tool overlay is in the GUI.
2 one axis is loosing steps, badly and progressively, so check for mechanical issues like binding first by removing the motor and spinning/moving the said axis, then check the drive for what current it is set to, make sure the motor connected to it can handle the set current.
The following user(s) said Thank You: aparecido

Please Log in or Create an account to join the conversation.

More
08 Dec 2020 16:24 - 08 Dec 2020 22:13 #191431 by aparecido
1-How do I disable offsets? I was testing and didn't activate anything, I don't know how it was activated.
2-Okay, I'll check the mechanics and electronics.

News:
I did some tests here, generated a gcode for the machine to move 30 times from the point "x 0 y 0" to "x 200 y 200" and return "x 0 y 0", at the end, it gave a difference of 0.03mm in x axis and 0.1mm on the y axis.
The axles are not weighted, removing the motor, the spindle of these two axes rotates freely in the hand. The motor current is adjusted correctly, I am using the CNC3AX driver board.
Last edit: 08 Dec 2020 22:13 by aparecido.

Please Log in or Create an account to join the conversation.

More
21 Dec 2020 22:03 - 21 Dec 2020 22:05 #192733 by aparecido
I was able to machine the plate a little while ago, practically all the holes were centralized, I think some 3 were left out but it is almost imperceptible, I believe it is something with the plate creation software (Eagle).
Well, detailing what I did, it might help others with the same problem:
-I reinforced the structure of the machine, welded some angles locking the 4 legs even more, to prevent the structure from having any balance when the machine moves.
-I changed the type of foot that was in the structure, to a rubber that fixes better on the floor.
-I adjusted the current adjustment trimpots of each axis of the CNC3AX board again.
-The handle that I use to probe the surface and adjust the imperfections by Gcode Ripper, I disconnected it from the part that was being machined, I got the impression that the spindle when it made the cutter touch the surface of the PCI, induced something by the probe that could interfere with the CNC3AX cycles responsible for controlling the motors.
Follow the photos of the last machined plate, thanks for your help.
Attachments:
Last edit: 21 Dec 2020 22:05 by aparecido.
The following user(s) said Thank You: tommylight

Please Log in or Create an account to join the conversation.

More
21 Dec 2020 22:35 #192739 by tommylight
The following user(s) said Thank You: aparecido

Please Log in or Create an account to join the conversation.

Time to create page: 0.152 seconds
Powered by Kunena Forum