Well, I more mean when you actually run this on your Grbl device, show us that G92 stays turned on as a work coordinate system.
The grbl-section-display still shows G54 as WCS … hm … why?
So you’re saying G92 is off after running the Touch Plate? Or are you saying g54 is on but you’re not sure whether g92 is on? g92 can be on while g54 is on so you have to independently query for g92.
See linked video. When I was doing my touch plate I don’t see any G92 commands at all. Just a G43.1. And things weren’t coming up right. Video here: https://youtu.be/gvjScU1ebi0
I think you guys are going to need @Jarret_Luft 's help here. My guess is that there could be a bug in the touch plate on runs after the first run. The touch plate widget does do some subscribing of events and it’s possible that on a 2nd or further run it subscribes a 2nd or more time to callbacks and thus it gets kinda whacky. Can you try running the touch plate from a fresh load of the browser? I do sort of remember fixing a bug a long time ago in the TinyG workspace of the touch plate widget being run multiple times and perhaps Jarret never fixed that in his version. The TinyG workspace surely gets a lot more attention than the Grbl workspace, so it could be that its an old bug.
Well, that seems even weirder then that Jarret’s experience is fine. The plot thickens.
I can’t quite understand why after you use the touchplate and it pulls up… wouldn’t the coordinate for Z always be exactly 2mm + the touch plate height then? If you are zeroing to that point and then moving up 2mm, shouldn’t the Z coordinate be 2 + the touch plate height? I never see exactly that. Sometimes it is close, but even in those case, I can hit go to zero and see that it isn’t perfectly on the surface sometimes. I guess I just don’t understand what that tool offset is doing since I don’t see any G92 commands at all for Z.
When I watched your video, that final position value is definitely wrong. When you then went to Z 0 it was doubly wrong. So something is amiss there. I wonder if it’s a difference in the Grbl version from what Jarret is using to what you’re using.
well, i don’t think it is not moving the amount i command… it says I am 17 mm above the surface but the bit is at the touch plate height plus 2 mm… so roughly 22. So when I tell it to go to zero, it is only going to go down 17 of those and leave the bit in the air. I think that part is fine. What I don’t understand is what that tool offset is doing to make the coordinate come out like it does, and why it never seems to give the same number (or close to the same) twice. It seems to me it should do the probe… stop… do a G92 Z then command up. But, I don’t know what I’m talking about when it comes to GCODE.
But I don’t hitnk it is an issue of the machine moving the right distance when I command it - I think that is fine. But the touch plate widget or the GRBL G43.1 implementation must have something wrong.
UPDATE: looking at the video, it looks like I’m running 0.9j.
so, I typed in $i (and $I) in the serial port console, but I don’t see any output. $$ works. But $I just basically returns “ok”. Am I doing something wrong?
how do i turn on verbose mode?