Calibration - "Failed to execute script main"

Those numbers all sound very reasonable :grimacing:

Okay, spending a little time fudging the numbers I can use all of the dimensions for my machine as is. But my calibration cuts have to be within 1mm for the calibration to complete. For instance if my 1-2 cut is 1929, I can use 1930 for my 3-4 cut but as soon as I go to 1931 it cannot complete.

1 Like

I have absolutely no idea why that would be the case, but we will look into it and figure it out!

Thanks! You guys rock!

1 Like

Regarding the lengths of cut 3-4 relative to cut 1-2, this is “expected”. If you perform the calibration with the chain sag set to 0, which I think you did, then threw algorithm believes there is no way that the distance between cuts 3-4 can be longer than 1-2. If it were this would require a negative chain sag value, which isn’t allowed. So that would explain why the calibration routine fails.

The real question would be why you’re getting 3-4 longer than 1-2. My only guess would be a mismatch between chain sag values in GC versus Firmware.

As a test, try setting your chain sag value to something like 10, then repeat.

3 Likes

@arnoldcp I simulated your calibration process and got the same result, that it could not be completed. I also confirmed that it is trying to use a negative chain sag value, which is why it is failing.

In looking through your logs I see where the parameters are set, matching the values you provided here. With all of that, there has to be something else going on here. Is your linkage kit functioning smoothly at each cut point? It would be interesting, if possible, to see a picture of the sled at each cut point to make sure we aren’t experiencing linkage issues. The values you measured are outside the bounds of what the calibration routine can understand, given the current physics models used, so I’m trying to track down what’s going on.

2 Likes

Could you shoot a video of the Maslow running through the calibration and share it?

Thank you

Arnold, are you actually making cuts and measuring them? or are you just putting
in values to simulate what you think would happen?

I actually made the cuts. I just left this AM for vacation but I will recut on Sunday evening and take a video. I haven’t really noticed any binding but perhaps it is.

The odd thing is I had cut 3 or 4 things on it after calibrated successfully on 1.04. then I ran this after moving it in my garage and upgrading to 1.06 and it kept failing.

BTW thanks everyone for the tremendously quick responses, I really do appreciate it.

3 Likes

Guys. Just got home so I won’t be doing a calibration cut tonight. Gotta get the car unpacked and recover from chasing my little guys around a water park for 2 days lol. I apologize.

3 Likes

Finally got back to the Maslow. I updated to 1.07 GC and FW and recalibrated. After calibration (the routine is way better now!) I ran the test cuts and it calculated!! Interestingly my motors we’re within 0.5mm and the chain sag was with just a few decimal places of what it was before all of this.

Thanks again to all of you for your help. I think it’s time to practice with a raised dog bowl holder for my ageing dog.

4 Likes

That is great news!

Just don’t use the “set chain lengths automatic” button, I broke that fixing everything else :roll_eyes:…but I’m working on fixing that right now.

3 Likes

He likes it!

My circles aren’t quite round so I’ll have to figure that out but overall a successful first project. 20 minutes in Fusion and I used a slow 20 IPM cut speed so about an hour to cut.

4 Likes

circles not being round is a calibraton/machine config error. This is one of the
things that is MUCH easier to deal with if you are using triangular kinematics.
(i.e. one of the linkage or ring kits)

2 Likes

Dawwwww!!! I like it too :grinning: Nice build.

1 Like

I do have a linkage kit. Forgive my ignorance but is there still the calibration routine that uses the circle inside the square so I can fine tune what i currently have?

No, with a linkage kit the calibration is far different, far simpler, and far
more accurate.

you are now actually figuring out what the machine will do rather than trying to
introduce error in one parameter to counter possible error in 5 other
parameters.

Okay, so basically, the culprit is most likely human measurement errors in my measurements for cuts 1-5. This is why I still have a little bit of an oblong circle.

Makes complete sense based on my solo use of a standard tape measure to take the measurements.

1 Like

The calibration process will build on itself to get more accurate if you run it again, so it might be worth doing that if it’s an issue. You should be under a mm of distortion across the whole sheet at this point.

1 Like

Thanks, that’s exactly what I needed to hear. I will run it a few more times.

2 Likes