That seems off, but I’m not sure how it could be causing the issue that we’re seeing. I think that the issue that you are seeing and
Might be the same thing. They seem quite related.
I think I just had a similar problem occur while running the machine. Everything was good for 1+ hours as it ran a detail program. Then I noticed the router had rotated in the mount (see other thread) so paused the job, twisted the router in the collar, then unpaused the job. It continued to run for awhile until halting.
I’ve attached the logs (this is running the 66.4 firmware I think - it is possible I downloaded 66.3). In the logs, you can see that it halts when the position error all of a sudden jumped up a lot. I don’t think it is a gcode issue but I included that too.
HaltErrorLogs.txt (233.1 KB)
Miller Lite_detail.nc (795.8 KB)
Let me know if you’d like me to test anything else and I can pick it up tomorrow.
Bob
1 Like
That could be related. It’s certainly the same error message. When you started the run did the tension in all four belts seem good / consistent? There wasn’t one belt that was extra slack or anything like that?
No, in fact when it started the belts were almost too tight - the machine was groaning a bit until it had moved a bit. And this only occurred after more than an hour of lots of movement.
1 Like
Is there an option to retract all then extend all 14.5 feet (I believe I read that is the length of the belts) THEN retract all with EQUAL tension on the majority of the belt roll?
Not even sure it would matter but I wonder.
there was one person who posted thay were having problems and discovered that
they had the toller with the magnet in the wrong place.
David Lang
I read that and when I reassembled one time, I double checked all magnets are in the right spot. Good thought.
1 Like
I just created a fix for the Take Slack → Play Button → Red Light issue that I was able to replicate, but it won’t fix the issue in your case where the calibration values seem off.
I’ve been digging into it and I haven’t found anything wrong there.
The only thing that I can think is that somehow the calibration process is producing funky results?
Can you try running calibration back to back twice and see if we get the same coordinates each time? If we’re not getting the same answer every time that would be a good indicator that it’s something in the calibration process.
If we get the same (or similar) coordinates every time then the issue has to be elsewhere (or is more likely to be elsewhere I think)
I can’t get machine to do calibration. It keeps pulling sled off the 4’x8’ board. I had this problem before with 3,000, 2,000 settings for which at that time you recommended upping to 3,100, 2,100 which worked at that time.
1 Like
It sounds to me like maybe your config file is getting reset somehow and the machine has different values from what we want.
0.66.5 adds a save button to the settings tab so that you can stop worrying about the .yaml file entirely and edit your settings directly on the user interface which might help.
Trying that right now. THANKS @ronlawrence3
2 Likes
I’ve been using it today with all my config changes, but let me know if you see any issues! 
1 Like
I can’t calibrate anymore. Retract, Extend, Calibrate. Belts will get pulled and get center 0,0 - I thought it was supposed to pull all the way to the bottom and then to the left. When it goes to the left, the belts seem like they are tightening too much. The sled will then go off the left side.
Serial Messages.txt (5.2 KB)
1 Like
I have truly no idea why that could be.
Can we try a really small calibration grid right in the center? Something like 2x2 points with big offset from the edges (Maslow_calibration_offset_X: 1300, Maslow_calibration_offset_Y: 850) that way we should stay far from the edges and it should be a quick process
I watched your calibration video again and as you can see in the “Serial Message,” it doesn’t go all the way to the bottom and then to the left as it shows around 8:00 in the video. It give predefined points which is new - did something change with calibration?
I’ll try a smaller grid. I did this yesterday with the 4x4 grid and the score was too low to use the values.
1 Like
Yeah, I would expect the results to not be useful but it should get us back to where things are working and we can build from there.
The calibration process has changed quite a bit since that video, but I would still expect it to go to the bottom and then over.
Ah I think that now it takes a measurement at the center, a measurement at the bottom and then a measurement over on the left so there are some extra points now I think?
The goal there is to help with keeping the belts organized.
Didn’t we run into this before and we solved it by increasing the size of the frame in the settings?
Yes, and I already tried increasing to 3200 and 2200 with same results.
1 Like
Darn 
Would this cause any problems with calibration? It was caught on the gear when calibrating 2 days ago. It doesn’t seem to be too problematic but maybe causing the problem?
It seems like the Top Right belt is pulling the machine around out of proportion to the other belts/motors. I can’t even calibrate with small field.
1 Like