First Layer consistently uneven/tilted Voron 2.4 350mm

[josh284]

Printer Model

Voron 2.4 r2

Extruder Type

Clockwork 2

Cooling Type

Afterburner

Hi All
I have been printing with my Voron 2.4 for nearly a year and have gone through many kilo's of filament without any major issues then recently the first layer changed to have a 'tilt' from being too high on the back left corner to dragging the nozzle on the PEI sheet on the front right corner.

I previously had perfect first layers, have been using Klicky and was only performing QGL and no bed mesh because my build plate is/was pretty flat and didn't need it.

As part of my trouble shooting I have:
- Twice tried loosening the z belts and re-tensioning using the soundspectrum app on my phone to find the belt tension/frequency and adjusted them to be as close as i can get.
- Added the adaptive bed mesh macro, that does a bed mesh over the area that the print will be performed
- I have even tried selectively tensioning some of the z-belts more to try and make the problem better or worse to identify where my issues is and it makes no noticeable difference after the QGL and bed mesh, the first layers are always the same amount of inconsistency with a gradient in the same diagonal direction.

I have attached some photos of the first layer (0.3mm layer height) from a test print and the bed mesh from the area where the the print was on the bed. I dont want to print closer to the front right corner of the bed because the nozzle drags on the PEI sheet and i would prefer not to damage the sheet.

Can anyone suggest what i can do to correct this issue?

Thanks

Josh

 

[Sanity Agathion]

What setup do you use? Single MGN12 rail, or 2x MGN9 rails? Are you running mesh when printer is fully heat-soaked?

Can you try to run this script to test probe accuracy in 4 corners of the bed?

GitHub - sporkus/probe_accuracy_tests

Contribute to sporkus/probe_accuracy_tests development by creating an account on GitHub.

github.com

 

[josh284]

Hi Sanity,

I have a single MGN12 rail on the gantry.

I have attached a mesh from when i had heat soaked the bed for over an hour.

I have also added the data from that script, i couldn't work out how to get the charts but the data looks like the sensor is ok. Is that correct?

Thanks in advance for your help. It really has me stumped.

 

[josh284]

Hi @Sanity Agathion, I was able to find a way to get the images, scp wasn't working for me but i downloaded the Cyberduck program and it made it much easier.

I have looked at these charts and they all look like the printers belts are tight and the klicky probe is reliable.

Am i missing something in these charts or is there somewhere else i can look to troubleshoot because it the printer is still printing with the same layer variation across the first layer.

Thanks

Josh

 

[Sanity Agathion]

Hi, just in case - which Klicky do you have? NG or standard? Or PCB? If it's not of the PCB kind, just in case, check if your magnets are not going out. That can cause some wird inconsistencies.
Also check if Klicky sits on the toolhead straight and well placed on magnets. If it's tilted or has any angle, it can cause wrong measurements too.

 

[RYGRR]

i'm having the same issue, i'm using tap, but my mesh is a taco along the x axis.

Hi, just in case - which Klicky do you have? NG or standard? Or PCB? If it's not of the PCB kind, just in case, check if your magnets are not going out. That can cause some wird inconsistencies.
Also check if Klicky sits on the toolhead straight and well placed on magnets. If it's tilted or has any angle, it can cause wrong measurements too.

^ love the commander keen avatar. i used to play that on an old tandy 386/486 pc off of a cd that i had to run from dos in the 90's.

 

[josh284]

Hi RYGRR, I have finally found a solution to my problem.

The solution came about when i found that i could tape a small shim to my bed and tell the QGL to test the location at that new point that was made artificially higher and then i was able to print again as per normal.

What i found was that i had a small piece of failed print stuck in my front right belt pulley. See attached photos.

This piece of junk wouldn't be in between the belt and the pulley when the print head had the klicky probe on because the nozzle was above the surface but when i dropped the Klicky probe and lowered the nozzle the piece of junk would get in between the belt and pulley effectively increasing the size of the pulley and moving the gantry out of level. It drove me crazy and once i found it the symptoms all made sense.

I still haven't printed the belt covers, though i think this piece could have gone down the hole in most of the covers i have seen. From that i dont think i can be bothered making a cover that closes the gap between the belts.

Hopefully this helps you. Good luck, because i know how frustrating it can be.

Also thanks @Sanity Agathion your input helped me rule out a bunch of things using tools i didn't know existed.

Josh

 

[Sanity Agathion]

oh FFS that's a headache to figure it out!

 

[RYGRR]

Hi RYGRR, I have finally found a solution to my problem.

The solution came about when i found that i could tape a small shim to my bed and tell the QGL to test the location at that new point that was made artificially higher and then i was able to print again as per normal.

What i found was that i had a small piece of failed print stuck in my front right belt pulley. See attached photos.

This piece of junk wouldn't be in between the belt and the pulley when the print head had the klicky probe on because the nozzle was above the surface but when i dropped the Klicky probe and lowered the nozzle the piece of junk would get in between the belt and pulley effectively increasing the size of the pulley and moving the gantry out of level. It drove me crazy and once i found it the symptoms all made sense.

I still haven't printed the belt covers, though i think this piece could have gone down the hole in most of the covers i have seen. From that i dont think i can be bothered making a cover that closes the gap between the belts.

Hopefully this helps you. Good luck, because i know how frustrating it can be.

Also thanks @Sanity Agathion your input helped me rule out a bunch of things using tools i didn't know existed.

Josh

i discovered that my r6 tap carriage warped, replaced with r8.

 

[m00dawg]

Digging up a slightly old thread but thought it was worth it here. I was having a similar issue and it was seemingly getting worse. I ran into this forum post and, sure enough, the front left corner pulley had some (ground up) filament which also seemed to strip part of the belt away. Been wondering why the front corner always seems to have had a harder time calibrating and, well, now I know! Either that or something else ate up a 20mm section of belt.

I think I have just enough length in belt left over from the initial build to run a new one. I'm gonna go ahead and buy more so I have some on hand in case that happens again.

I don't have the covers on mine either (I couldn't sort out how to install them). But I think one thing that might help is, instead of flat covers, perhaps angled shields tall enough to shield the holes from the bed. That would avoid having to have a thin slit to allow the belt to run through. I might give that some thought to see what can be done while still making sure the gantry clears something like that.