search

prusa3d / Prusa-Firmware-Buddy

Firmware for the Original Prusa MINI, Original Prusa MK4 and the Original Prusa XL 3D printers by Prusa Research.
Other
1.12k stars 219 forks source link

[BUG] 4.4.0 release issues with mesh bed leveling #2798

Closed prongato closed 2 months ago

prongato commented 1 year ago

Printer type - MINI/MINI+

Printer firmware version - 4.4.0 RC1, 4.4.0 release

Original or Custom firmware - Original

Optional upgrades - Prusa Filament Runout Sensor, E3D Revo Micro, Octoprint

USB drive or USB/Octoprint - micro USB/Octoprint

Describe the bug

In Octoprint or any other software package that allows you to see the terminal output from the mini, after mesh bed leveling, the returned values are sometimes missing or just plain wrong. Sometimes an entire line is missing; sometimes it’s just a value it two on one line. See the attached photo for an example.

I’ve verified this issue across 3 separate Buddy boards (rev 1.0.7, rev 1.0.0, rev 1.0.8), Prusa Mini firmware 4.4.0 RC1 and Prusa Mini firmware latest release

How to reproduce

It doesn’t make a difference if the mesh bed leveling is triggered by starting a print, use Bed Visualizer or trigger with G29 via the terminal, it happens with all of them.

Expected behavior

when mesh bed leveling completes, the output to the terminal, Bilinear Leveling Grid: sometimes is missing values or even missing and entire line (ex, returns values for lines 0,1 and 3 but no values for 2.

G-code

it happens with all gcode and even with no gcode

Crash dump file

N/A

Photo - https://forum.prusa3d.com/wp-content/uploads/wpforo/default_attachments/1670228502-ED9AC854-78F8-4991-8DAA-60DA7DA60A04.jpeg

Tommy544 commented 1 year ago

Same issue here, following...

Prusa-Support commented 1 year ago

[sorry an accidental combination of keys sent the message closing the issue while I was elaborating the message]

Prusa-Support commented 1 year ago

Thanks for reporting this.

That's very interesting. This issue could be a duplicate of issue https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/2729, or related to it. The missing output may also be related to a known issue with serial communication (https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/2815). Please try FW 4.4.1-RC1.

Moreover, it is curious how in issue #2767, both serial communication and calibration issues are mentioned too. I may start thinking that the two things are related.

Regardless of the missing probed values output, would the Mesh Bed Levelling be successful? Would the first layer accuracy be affected in any way?

Michele Moramarco Prusa Research

ZasoZ commented 1 year ago

I had the same problem with 4.4.0 but downgraded to solve the issue , if you need an additional persons to try it i can upgrade to 4.4.0 again and see if the first layer is affected.

Kind regards

Prusa-Support commented 1 year ago

Yes, more tests could help. We may need more printers and users to be able to reproduce the issue. Also, I wonder if this is only a missing value output or if the calibration and the first layer quality are affected in any way.

Please try FW 4.4.1 (final) though, as it fixes some serial connection problems.

Michele Moramarco Prusa Research

ZasoZ commented 1 year ago

Hi So now that i have had time to test it and i can report that mesh bed leveling fails in the calibration menu as well. Though updating to 4.4.1 fixed that issue and my problems with G29 (and octoprints mesh bed visualizer). So happy to report that it is fixed for me.

Prusa-Support commented 1 year ago

Thanks for your feedback. That confirms the nature of the problem we are looking at, which is fixed by now.

I haven't heard from other users since the possible issue was posted more than one month ago but it seems safe to close this issue as solved. Feel share more feedback though.

Michele Moramarco Prusa Research

Prusa-Support commented 1 year ago

Sorry for re-intervening in this closed issue. According to a new issue (https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/3036), this problem continues happening on the most recent firmware.

I wonder if any of the users in this conversation is still able to reproduce the problem with the most recent firmware version, or if any user recognized a sequence of troubleshooting steps that probably resolved the problem in their case (e.g. a system refresh, disconnection of Octorpint, the use of a different USB cable...).

Michele Moramarco Prusa Research

Tommy544 commented 1 year ago

@Prusa-Support I am still experiencing this problem using firmware 4.4.1 on my Mini+.

In my case, I get missing entries more often than the correct output and my workaround so far has just been to "brute-force" it and start the measuring process enough times until I get a complete output.

I was using the latest version of Octoprint on a Raspberry Pi 3. I haven't tried to switch USB cables to see if that has any effect.

rinkek commented 1 year ago

Yes, i still have this issue in 4.4.1 on the Mini+ I normally use Prusalink to print from PrusaSlicer, but use Octoprint for bed leveling which runs on a RPI 3b+ which needs to be booted before use. So I've only tried this with Octoprint. You can see it in the terminal; missing values.

marek26340 commented 1 year ago

Hi. I have already made some comments in other issues here, detailing my issues with the firmware version 4.4.1. #3042 (2nd comment)

So yes, I also have discovered this issue, among with other things, all of which disappeared the moment I downgraded back to 4.3.4. Feel free to delete my comment if you think that I am spamming. I'm only confirming that I did encounter this issue.

n-dony commented 1 year ago

Hi. Same problem. Randomly missing values ​​in the bilinear leveling grid. Mini+ factory assembled, firmware 4.4.1. 16 points taken without error on the printer, but sometimes missing values in ​​the output to the terminal.

Prusa-Support commented 11 months ago

Feel free to give it a go with the new FW 5.1.0 but please mind it is an alpha version. A factory reset may be needed.

However, our developers plan to focus on serial line communication issues in the next few firmware releases (probably after 5.1.0) so please let us know if the problem persists.

Michele Moramarco Prusa Research

jakl-159 commented 9 months ago

Printer type - MINI/MINI+

Printer firmware version - 5.1.0 BETA

Original or Custom firmware - Original

Optional upgrades - Prusa Filament Runout Sensor, Octoprint

USB drive or USB/Octoprint - micro USB/Octoprint

I do have the same problem with my Prusa MINI+. After doing maintenance I wanted to to check the bed level but getting similar response as above/other threads on this issue (MBL). This is the response after G28 -> G29 commands from pronterface/octoprint:

Default storage slot 0 selected. Mesh invalidated. Probing mesh. echo:endstops hit: Z:-0.10 echo:endstops hit: Z:-0.11 echo:endstops hit: Z:-0.11 echo:endstops hit: Z:-0.13 echo:endstops hit: Z:-0.15 echo:endstops hit: Z:-0.14 echo:endstops hit: Z:-0.10 echo:endstops hit: Z:-0.10 echo:endstops hit: Z:-0.10 echo:endstops hit: Z:0.03 echo:endstops hit: Z:0.03 echo:endstops hit: Z:0.03 echo:endstops hit: Z:0.13 echo:endstops hit: Z:0.13 echo:endstops hit: Z:0.13 echo:endstops hit: Z:-0.12 echo:endstops hit: Z:-0.12 echo:endstops hit: Z:-0.12 echo:endstops hit: Z:-0.23 echo:endstops hit: Z:-0.23 echo:endstops hit: Z:-0.23 echo:endstops hit: Z:-0.35 echo:endstops hit: Z:-0.34 echo:endstops hit: Z:-0.34 echo:endstops hit: Z:0.01 echo:endstops hit: Z:0.01 echo:endstops hit: Z:0.01 echo:endstops hit: Z:-0.02 echo:endstops hit: Z:-0.02 echo:endstops hit: Z:-0.02 echo:endstops hit: Z:-0.10 echo:endstops hit: Z:-0.11 echo:endstops hit: Z:-0.11 echo:endstops hit: Z:0.13 echo:endstops hit: Z:-0.09 echo:endstops hit: Z:-0.09 echo:endstops hit: Z:-0.09 echo:endstops hit: Z:-0.17 echo:endstops hit: Z:-0.17 echo:endstops hit: Z:-0.19 echo:endstops hit: Z:-0.19 echo:endstops hit: Z:-0.19 echo:endstops hit: Z:-0.24 echo:endstops hit: Z:-0.24 echo:endstops hit: Z:-0.24 Extrapolating mesh...done Unified Bed Leveling System v1.01 active

Factory reset did not help.

ricarva commented 9 months ago

Hi, everyone

I'm experiencing this issue, apparently: it manifests by Octoprint's Bed Level Visualizer not being able to get/refresh values, even though the procedure completes on the printer with no errors.

Is this working for anyone? Is it universally broken?

marek26340 commented 9 months ago

Hi, everyone

I'm experiencing this issue, apparently: it manifests by Octoprint's Bed Level Visualizer not being able to get/refresh values, even though the procedure completes on the printer with no errors.

Is this working for anyone? Is it universally broken?

Yes, it's a widely known problem.

Go ahead and upgrade to 5.1.0, it's supposed to finally fix it. 4.4.0 was just plain unusable in my opinion. It's been great so far!

edit: Apparently it's still not fixed. Sorry!

jakl-159 commented 9 months ago

Still ocuring in 5.1.0 release for me.

marek26340 commented 9 months ago

Still ocuring in 5.1.0 release for me.

Oh. Goddamnit! Should I just delete my comment then?

ricarva commented 9 months ago

Yup, can confirm that the issue is still present in 5.1.0 for me too 🫤

ricarva commented 9 months ago

Feel free to give it a go with the new FW 5.1.0 but please mind it is an alpha version. A factory reset may be needed.

However, our developers plan to focus on serial line communication issues in the next few firmware releases (probably after 5.1.0) so please let us know if the problem persists.

Michele Moramarco Prusa Research

@Prusa-Support

Yup, still present in 5.1.0

n-dony commented 9 months ago

Hi, For me, no more missing values ​​with FW 5.1.0. But I need to ask for a report with G29 T to get a matrix usable by the plugin.

G28 ; home all without mesh bed level G29 ; mesh bed leveling @BEDLEVELVISUALIZER G29 T; mesh bed leveling report

github-actions[bot] commented 3 months ago

This issue has been flagged as stale because it has been open for 60 days with no activity. The issue will be closed in 7 days unless someone removes the "stale" label or adds a comment.

github-actions[bot] commented 2 months ago

This issue has been closed due to lack of recent activity. Please consider opening a new one if needed.