Problem: Probing was skipped randomly, for the 1st tool change, before going for tool change in atc_manual macro. #1427
Comments
|
It will be hard for me to help if I cannot repeat the problem, so... Attach a gcode file that exhibits the problem. The smallest and simplest example would be great. What do you normally do between jobs? Do you manually send any gcode, like to return the last tool to the rack? You mentioned a soft reset button and a start button. Are these UGS buttons? If you do further testing check some values between jobs. See if anything looks wrong.
|
|
Thanks for the quick reply.
Thanks for tips given, I'll use the commands to check on it further. |
|
N100 (Exported by FreeCAD) |
|
I doubt this is the problem, but I don't recommend the use of M2 or M30 at the end of a file with grbl style firmwares. It can reset some things like modal values that you may have set before running the file or while running the file. https://linuxcnc.org/docs/2.8/html/gcode/m-code.html#mcode:m2-m30 I can try to reproduce your problem tomorrow. |
|
I got you.. thanks for the update. I'll wait for your result then. my machine is taking a break after a long run, just to check everything. |
|
The best way to reset before a new job is to send M6T0. Manually sending G49, etc, is problematic if everything is not done perfectly. It need to reconfigure a machine to with a toolsetter. I won't be able for about 4-5 hours. |
|
It works by comparing new tools to the one that was used to set the work zero. If you run a second job, you need to let the system know that the original work zero is no longer valid. You do this by sending M6T0. Here is the workflow you should be doing.
|
|
In my case here, it's not a 'new job', it's the same job, repeated ones because we have lots of parts to be machined. But I understand what you're suggesting. I did just the same as your were suggesting in the first job run. No problem with this. On the next or new job, now I need to send M6T0 (to reset the ATC). Cool. Then I need to set the right tool for the job, so M61Q<tool_num> or M6T<tool_num> doesn't matter right because it's a manual ATC..? But it's getting the right tool for 1st operation in the job again.
So, when do I need to send G49 to clear the previous tool offset..? I've seen from my observation, where I sent a G49, after a soft reset (from UGS), and the Z height do changed back to it's zeroed value. So, the soft reset didn't clear tool length offset, just to let you know this. |
|
It is the same job as far as the gcode. If you zero the new workpiece is is basically a new job for the controller. After the M6T0....M61Qx will change the current tool number. M6Tx will move to the tool change location and change the tool number. You should not send a G49. Let the firmware control the TLO. If you make a mistake. you could slam a tool. This would happen when you cleared the TLO and the job started with a longer tool than the Z zero was set with. If the gcode sees an M6Tx where X was the same as the previous tool, it skips the toolsetter to save time. It assumes the TLO is correct. |
|
Okay.. i learned a couple of new things on the subject. Thanks for the update. I haven't got any batch job for the moment, but I already make a note on the commands and its sequence. I'll come back to update the status once I got the chance to do so. |
Wiki Search Terms
probing skipped atc manual
Controller Board
6-pack external
Machine Description
Gantry router with external HBS860H all closed loop, dual X motor, single Y, single Z, Fuling/Fohlinn DB600 VFD 3.7KW, 3.5KW 1200Hz 8pole ATC spindle.
Input Circuits
No response
Configuration file
Startup Messages
User Interface Software
UGS 2.1.9
What happened?
** All axes homed **
1- load my nc file.
2- typed in G56 - my work offset
3- M61 Q10 - load the 1st tool used in the program
4- Hit Start button
5- Reached next tool change. Program paused, spindle turning down.
++normal case++
6- Z moves up, spindle move on to fixed probe for tool length offset probing, and then go to tool change location.
7- Insert new tool, hit cycle start button, spindle goes to fixed probe to complete tool length offset probing.
8- Comes back to the last location at the last tool height, and continue with the rest of the program. I smiled.
9- Subsequent tool changes exhibit no problems at all.
<>
6- Z moves up, spindle move on to tool change location.
7- Insert new tool, hit cycle start button, spindle goes to fixed probe to complete tool length offset probing.
8- Comes back to the last location at the wrong tool height, and crashed on the work piece. I died.
This <> almost always happen when you ran the same program for the next part (2nd, 3rd.. on), regardless if you hit the soft reset button or not. But a power cycle will confirm the ++normal case++ to happen on the fresh run of the program, but not on the subsequent run of the program. Not without a power cycle.
GCode File
No response
Other Information
No response
The text was updated successfully, but these errors were encountered: