Problem calibrating Y axis

I have had an issue with the Y axis getting locked up. Increased the VRef on the OpenBuilds Blackbox. It seems ro have smoothed things out.

Trying to calibrate the Y steps and having an issue. Even with the acceleration turned down from 150 to 50 I am getting a weird result.

Here are my movements/measures:

10mm measured result is 9.77mm
Another 10mm on top is 19.88mm
Another 10mm on top is 29.91mm
Another 10mm on top is 39.97mm
Another 10mm on top is 50.01mm
Another 10mm on top is 60.06mm

so I start under the expected result but then around 50mm travel it’s pretty much right. Any more than that and it goes over the expected measured distance. This is all done at 50 acceleration.

I don’t understand how it starts under then goes over. The results are repeatable with the same outcome.

Any ideas what could be causing this or how to correct it?

 

This is also happening on the X axis, just checked

 

Have also confirmed it with 2 individual different branded digital measures with exact same results.

 

It's all super rigid chain driven with almost no slack or backlash at all. I have adjusted chain tension and still getting the same result, which is why I was so stumped Checked all couplers and rail systems for any movement and can't find anything that would put off the positioning.

 

I have just cut a 2200mm to test the X and 1000mm on the Y then measured and adjusted from there. Might still be problems with the machine but I figure the difference over those large sizes should be enough for now. Thanks for the help mate, it was probably good to check over all that stuff anyways.

 

I have had it all checked out. Setting it off a longer piece seems to have given me back more accuracy in the measurements. Getting some really weird results now when cutting a big job. A symmetrical lock hole is no longer symmetrical even though it is in the GCode and no matter where I move the piece on my machine.

Also some slots now have have little jagged parts on the same corner on slots in different positions

 

It’s just weird that this is showing up now. I redid the post process (OpenBuilds GRBL for Fusion 360) to make sure there wasn’t a problem with the export. It just keeps doing the same thing.

 

Anticlockwise

 

Machine slowed down and specifically made the moves so it’s not like it grabbed. Running the job in the air does the same things.

 

2 GCode files there. One is the full with the predrilled holes and stuff. The 2nd file I just cut operation 1 and 2 out while still leaving the spindle, delay and mist on with startup.

Both files have the same results. I am using OpenBuilds Control with OpenBuilds Blackbox. This exact file has cut fine before and my GRBL settings are all the same as when it was cut originally. I am completely up to date with latest version of Control but the working cut was before the latest update. I do not believe there to be a link between the update and the issue with what is on GitHub I don't think there is any possible link there.

I have recently updated the X and Y steps but only by 0.139 and 0.089 at the most on either.

I don't feel that it is the hardware of the machine doing something specific and I have tried moving the cut piece on the X and Y to ensure there isn't like a specific dead point somewhere. The error with the slots having a pointy bit in them happens on all 3 of the small ones, but the larger one works out fine.

Let me know if you need more information, I would rather do any leg work to save you time and I really appreciate the help. This problem has my anxiety going all sorts of crazy.

 

Forgot to mention, the full gcode file will move to like X50 Y50 after it does the initial screw down holes. I then go to OpenBuilds Control and press pause then resume after I do all the screw downs and the machine resumes. This however shouldn't impact anything as the shorter file doesn't do this - but also has the same issue with the slots and the hole for the lock being short on one side.

The other weird thing is that even though the lock hole is shorter on one side, it perfectly completes the cut. Like if it just shifted then it wouldn't complete the contour either.

Again, thanks

 

Here are my GRBL Settings from OpenBuilds Control if that helps.

$0=10 ; Step pulse time, microseconds
$1=255 ; Step idle delay, milliseconds
$2=0 ; Step pulse invert, mask
$3=3 ; Step direction invert, mask
$4=1 ; Invert step enable pin, boolean
$5=1 ; Invert limit pins, boolean
$6=0 ; Invert probe pin, boolean
$10=2 ; Status report options, mask
$11=0.020 ; Junction deviation, millimeters
$12=0.002 ; Arc tolerance, millimeters
$13=0 ; Report in inches, boolean
$20=0 ; Soft limits enable, boolean
$21=1 ; Hard limits enable, boolean
$22=1 ; Homing cycle enable, boolean
$23=3 ; Homing direction invert, mask
$24=100.000 ; Homing locate feed rate, mm/min
$25=3000.000 ; Homing search seek rate, mm/min
$26=25 ; Homing switch debounce delay, milliseconds
$27=5.000 ; Homing switch pull-off distance, millimeters
$30=18000 ; Maximum spindle speed, RPM
$31=0 ; Minimum spindle speed, RPM
$32=0 ; Laser-mode enable, boolean
$100=83.796 ; X-axis steps per millimeter
$101=83.225 ; Y-axis steps per millimeter
$102=505.415 ; Z-axis steps per millimeter
$110=10000.000 ; X-axis maximum rate, mm/min
$111=10000.000 ; Y-axis maximum rate, mm/min
$112=500.000 ; Z-axis maximum rate, mm/min
$120=150.000 ; X-axis acceleration, mm/sec^2
$121=150.000 ; Y-axis acceleration, mm/sec^2
$122=150.000 ; Z-axis acceleration, mm/sec^2
$130=3205.000 ; X-axis maximum travel, millimeters
$131=1250.000 ; Y-axis maximum travel, millimeters
$132=50.000 ; Z-axis maximum travel, millimeters

 

Normally I do spiral in, I think I just forgot on this run. The bit is actually an 8mm bit on that job, just selected the generic bit as it doesn’t seem to impact the resulting GCode and it means I don’t have to have each one as a different entry in Fusion. It’s just set to the longest one when cutting through 16mm ally.

I don’t seem to have any problems normally cutting one pass for up to 5mm. 10mm is clean still in 3 passes too.

I am totally doing things in regards to the cuts outside of the recommendations, however I get mirror finish cuts with razor edges up in 10mm ally with the 12mm bits. I just replace them regularly as they cost less than AU$10 each.

I tried rotating the piece and I am still getting the exact same errors. The X and Y are really different in terms of their mechanical movement.

I also disengaged the drive system and just watched what the motors are doing with 0 load on them. They are still going through the motions to create these glitches. It was really easy to see on the slots. I just finished this up and came back to post the results.

The only components in my latest test were the computer, the Blackbox and the motors. Well and that same GCode file.

It’s bugging me that with the motors free of any connection to the mechanical that it still wants to play games.

I am not sure what else to test. This is really frustrating. I appreciate your help ruling out the GCode as an issue. I am really grateful. I will persist and try and work out where things are going wrong. Thanks mate.

 

The machine goes up to about 16000 before it go brrrrrr and stops when I have tried to push it for curiosity. I usually just set it to 8000 to get around the 1500x3800mm bed quick enough.

Would be nice if Control remembered that I like 8k but it always sets it to 4 point something something K each time (cheeky feature request)

 

Also an off topic bone to pick with you two.. you said having an Arduino with GRBL was a good way to test interacting with OpenBuilds Control. It totally was and that isn’t the bone part. I now have 2 Unos, 2 Megas and about 10 ESP32 Arduino like devices and am significantly through building a Mars rover like robot and completed a pen holding drawing machine. The bone to pick part - You forgot to warn me how fun, easy and addictive it was to drive steppers and stuff with Arduino devices. Seriously thank you though.. you brought me a lot of joy and fun times building stuff with my niece.

 

That is **** cool! I live about a 30 second walk from the rainforest so I don’t need monitors for temp and humidity. It’s always hot and always max humidity

I’m kinda homeless, living with family since the flood destroyed my house and all my stuff. I don’t think they want me to modify the house at all. But I am willing to push that line and tell them it’s Peters fault

 

I just woke up and it occurred to me, if it was something mechanical causing my problem the slot wouldn’t know where to complete and the hole for the lock shouldn’t come back to the right place to complete the contour.

I know this might not be right, I just don’t understand how it’s possible.

 

If you draw a line down the middle of that cut is it symmetrical on both sides? Not 100% sure but I think maybe the photo is just hard to see if it’s a mirror or if the right hand side is slumping.

Yeah I rotated the project 90 degrees and reprocess and cut again and it makes the same errors in the same spot and is identical.

How do I reset everything back to scratch in my blackbox? Figure it probably won’t help but might be worth doing. I assume I just backup my settings, and do something to the EEPROM?

 

Thanks again man, I will test it all out and see if it persists after flashing. There really seems to be a bug in the system and I just got asked if I could do my first proper work with the machine today. Great timing.