Notifications
Clear all

Mystery crash? possible corrupt board?  

  RSS
jason halter
(@jason-halter)
Active Member
Mystery crash? possible corrupt board?

Woke up this morning to see this on the my new mini.  It was frozen in this state, I attempted everything, no joy...  a 32+ hr print destroyed.  The file was over 220Kb. I am wondering if there is a limit to how large the file can be? like say 199Kb?  honestly IDK, hoping someone on here can shed some light. I have not attempted any consecutive prints after his crash.  

I do have some additional questions:

1.     Why am I getting tonnes of zits when using my MacBookPro --> pronterface --> USB-A cable into the mini.  When using the USB solo into the mini, the results is no more zits? What is up with that?

2.     Has anyone modified the cooling fan outlet. seems to be too high[elevation]. Presently the air seems to be driven straight into the silicon sock I installed, resulting in stringing when using PLA? 

Posted : 12/08/2020 3:19 pm
pab
 pab
(@pab)
Active Member
RE: Mystery crash? possible corrupt board?

Looks like this problem: https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/771

Can you upload your crash dump to that issue?

 

Edit: Seems like another bug, so better create a new issue @ github.

This post was modified 1 year ago by pab
Posted : 12/08/2020 3:31 pm
jason halter
(@jason-halter)
Active Member
Topic starter answered:
RE: Mystery crash? possible corrupt board?

@pab

IDK what a crash dump is? or how to...

Posted : 12/08/2020 3:51 pm
pab
 pab
(@pab)
Active Member
RE: Mystery crash? possible corrupt board?

Its a feature added with firmware 4.0.3:

Crash Dump feature

The advanced 32bit SoC running the Buddy board enables logging the system events. In the case of a system crash, the information is written in the SPI flash. After the printer is restarted, users can download the Crash Dump file onto a USB drive and provide it to the support team/developers for debugging purposes.

To download the Crash Dump file, insert a USB drive into the printer first and then head to “Settings -> Save Crash Dump”.

Posted : 12/08/2020 3:53 pm
jason halter
(@jason-halter)
Active Member
Topic starter answered:
RE: Mystery crash? possible corrupt board?

Ok, I have the bin file on the USB now.  But I can not upload the >bin< file to github.  I tried to open it with C++ as well, still no-go?  How do I get a bin.file onto github for assessment then? 

Posted : 12/08/2020 4:19 pm
pab
 pab
(@pab)
Active Member
RE: Mystery crash? possible corrupt board?

Just zip the .bin file (if you are on a recent windows, just right-click it and select "send to" and "compressed (zipped) folder").

Then github will accept that file.

Posted : 12/08/2020 4:25 pm
jason halter
(@jason-halter)
Active Member
Topic starter answered:
RE: Mystery crash? possible corrupt board?

G1 X157.713 Y107.263 E0.00530
G1 X157.733 Y107.444 E0.00324
G1 X157.706 Y107.721 E0.00495
G1 X157.558 Y107.993 E0.00552
G1 X157.371 Y108.117 E0.00401
M204 S1250
G1 X157.175 Y108.040 F9000.000
G1 E-1.40000 F4200.00000
G1 F7200.000
G1 X157.141 Y108.183 E-0.13235
G1 X156.916 Y108.117 E-0.12976
G1 X156.729 Y107.946 E-0.14045
G1 X156.688 Y107.856 E-0.05490
G1 X156.642 Y107.658 E-0.11254
G1 E-0.03000 F4200.00000
G1 Z121.300 F9000.000
G1 X158.600 Y105.035
G1 Z120.800
G1 E2.00000 F2400.00000
M204 S800
G1 F1500.000

Posted : 12/08/2020 4:28 pm
Crawlerin
(@crawlerin)
Prominent Member
RE: Mystery crash? possible corrupt board?

Quick question - why do ypu print from Macbook through Pronterface? Why not from USB stick from MINI directly, or from deficated print server like Octoprint? You get zits because your computer is not fast enough, it is not feeding data to USB port in time and printer skips parts of gcode. It is not recommended to use any other than dedicated computer for printing, because any activity taking place that moment may disrupt real-time stream of data to USB. Not to mention if your computer decides it needs update and reboots in the middle of long print.... Besides, Pronterface is not actively developed anymore.

Print from USB stick directly, or if you insist on USB then get a Raspberty Pi and install Octoprint print server on it.

Posted : 12/08/2020 9:47 pm
jason halter
(@jason-halter)
Active Member
Topic starter answered:
RE: Mystery crash? possible corrupt board?

@crawlerin

Primarily because my printers are upstairs and my primary work station is downstairs.  It is all about eliminating the trips up and down stairs. If you need to do a quick slice you have the capacity to do that there on the spot. These laptops are indeed dedicated to just the printer. They are old school MacBook Pros that are set to never sleep, dedicated to just pronterface to the printer. I am aware that the printer is reliant on the laptop, and aware that a causation of zits is due to the laptop not sending info @ a proper rate.  With pronterface you still have some manual controls [ minimal compared to the UI on the printer]. Pronterface was something that I seemed to be familiar with and worked well for me previously.  Anyways... I have read somewhere, can not remember specifically, that 0.10 mm settings are prone to zits.  <-- IDK? As of now I have only seen zits on curves, nothing on any straight edges. 

Posted : 13/08/2020 1:37 am
Crawlerin
(@crawlerin)
Prominent Member
RE: Mystery crash? possible corrupt board?

@j-halter

I see. Well, that's one way to solve it ... though I still think Octoprint would be more beneficial for such purposes, because it includes not only printing itself, but many useful plugins (webcam, spaghetti detective, exclude failed region,...). There seem to be unofficial Mac port of Octoprint, https://community.octoprint.org/t/setting-up-octoprint-on-macos/13425 you can give it a try, if that does not work well, there's still RPi and official Octopi.

You get zits when printing curves because of that exact reason of computer not keeping up sending data to printer. Straight motion is (basically) one command to X or Y and E motor, but for printing curves Slicer calculates it as series of small X&Y then X&Y then X&Y ... movements. Instead of one or two commands computer needs to send hundreds small instructions to printer while extruder keeps extruding. Any small stall causes filament to keep coming out while head stands still because motors wait for their instructions.

Posted : 13/08/2020 9:43 am
Share: