Garolite bed crashes  

RE: Garolite bed crashes

Posted by: Vojtěch

Posted by: charles.h13

Posted by: Vojtěch

 

1. Buy a bare steel sheet, some 468MP adhesive, a 0.2mm or 0.5mm Garolite sheet, and make yourself your own garolite flex sheet. Or buy one pre-made.

That is fantastic.  Thanks for sharing.  I have been looking for one for months.  

I cannot vouch for quality or fitness for purpose of the pre-made one, though, I have made one for myself from the Aliexpress parts.

I will order and test one.  

RE: Garolite bed crashes

Bear in mind, it's sensing range is not as good as a PRUSA sheet, and it is not as flexible, But, I get really nice first layers (Consistent and more even than a PRUSA textured sheet) on the same FYSETC steel sheet as shown in the link.

I got mine through Amazon. Arrived 3 days after ordering.

When you get it, clean it thoroughly with IPA, and use a plastic scraper working across to push any air bubbles out as you lay your sheet on top.

Try to overlap on all sides then simply run a sharp knife along the edge to trim any overhang to the steel sheet.

RE: Garolite bed crashes

Posted by: chocki

Bear in mind, it's sensing range is not as good as a PRUSA sheet, and it is not as flexible, But, I get really nice first layers (Consistent and more even than a PRUSA textured sheet) on the same FYSETC steel sheet as shown in the link.

I got mine through Amazon. Arrived 3 days after ordering.

When you get it, clean it thoroughly with IPA, and use a plastic scraper working across to push any air bubbles out as you lay your sheet on top.

Try to overlap on all sides then simply run a sharp knife along the edge to trim any overhang to the steel sheet.

Amazon?  which link?  

RE: Garolite bed crashes

In the US: https://www.amazon.com/FYSETC-Printer-Platform-Printing-Buildplate/dp/B07LBQ57WV

In the UK: https://www.amazon.co.uk/Printer-Platform-FYSETC-Printing-Buildplate/dp/B07MLZ6D9P/ref=sr_1_27_sspa?keywords=FYSETC&qid=1559742080&s=gateway&sr=8-27-spons&psc=1

in Germany: https://www.amazon.de/FYSETC-Federstahlblech-Heizbett-Plattform-Druckplatte/dp/B07MLZ6D9P/ref=sr_1_7?__mk_de_DE=%C3%85M%C3%85%C5%BD%C3%95%C3%91&keywords=fysetc&qid=1559742122&s=gateway&sr=8-7

Just search for FYSETC on Amazon.

RE: Garolite bed crashes

Ok, you were talking just the metal bed.  I think t might be worth it to by the applied one from Etsy.  

RE: Garolite bed crashes

Quick correction: The cheap BuildTak-like surfaces I purchased were from FYSETC, but the actual "spring steel" sheet is from Blurolls. This thing is a monstrosity, and with the 2 FYSETC surfaces applied, comes in at nearly 2mm thick. I find it unusable. It sounds like FYSETC is a good choice. Others less so.

RE: Garolite bed crashes

I may have to check this garolyte out myself as polycarbonate WELDS to buildtack, but from a couple of reads this garolyte might be a better choice, but not 1mm thick.

RE: Garolite bed crashes

 Of course Voj still hasn't explained HOW he got the words to display on the screen ... too funny dude.  Useless, but funny.

RE: Garolite bed crashes

Posted by: Tim

 Of course Voj still hasn't explained HOW he got the words to display on the screen ... too funny dude.  Useless, but funny.

It works for simple gcode files that don't contain anything useful. All what I've done is I've put those two lines (M0 text and M84) that I've included in my post into a gcode file and ran that gcode file. It displayed the words. I'm sorry, I assumed that was obvious from my post.

But it breaks for anything bigger, and it looks like a firmware bug. I've spend a couple hours reading the firmware sources today, now I'm back at home and I'm going to do some debugging.

RE: Garolite bed crashes

I found why the Prusa Marlin firmware doesn't display the M0 message. It's because the print progress status and filename display takes precedence when printing from a SD card. This M0.patch fixes the problem and messages are displayed even on long gcode files.

I still need to continue debugging, because there is another bug in the firmware that skips the M0 gcode entirely, not waiting at all.

RE: Garolite bed crashes

In my gcode above I discovered a minor flip of data: 

G0 X200 Y0 Z100 ; this is a good waiting point for bed flip

should be

G0 X0 Y200 Z100 ; this is a good waiting point for bed flip

The idea is to put the bed forward for easier handling.  And there's probably some other moves that can be helpful for cases where a user bumps the bed forward or backwards when flipping the sheet.

RE: Garolite bed crashes

Debugging finished. The next bug was that the M0 command would be ignored if the text contained the capital letter 'G'. So when I tried

M0 Install Garolite!

then the printer would just ignore the statement. Plus I found another bug in M117, which I also fixed. The Marlin code is pretty crusty.

Anyway, here is the combined M0-v2.patch. Apply using

patch -p1 < M0-v2.patch

 to the checked-out Prusa-Firmware git tree, recompile, flash into the printer and the M0 and M1 commands (as well as M117) will work as they're supposed to.

RE: Garolite bed crashes

What's your github index for the bug you entered?

RE: Garolite bed crashes

Posted by: Tim

What's your github index for the bug you entered?

Well, at the moment I'm figuring out how to do that properly, hopefully including a git pull request. I'm not a github user yet.

RE: Garolite bed crashes

You don't need a pull request, just get an account, and post an issue; someone will eventually look at it.

RE: Garolite bed crashes

Posted by: Tim

You don't need a pull request, just get an account, and post an issue; someone will eventually look at it.

Well, there already was one https://ghttps://github.com/prusa3d/Prusa-Firmware/issues/1898ithub.com/prusa3d/Prusa-Firmware/issues/1898 so I just added a link to my solution.

RE: Garolite bed crashes

And pull request done, too. https://https://github.com/prusa3d/Prusa-Firmware/pull/1900github.com/prusa3d/Prusa-Firmware/pull/1900

RE: Garolite bed crashes

Thanks.  Worth noting is the M117 defect is still there and my bug report for it is still hanging fruit. Maybe your sample code will be enough to get someone off their fanny to dig into and actually fix the two issues.    

Sounds like another case of poor interrupt handling (pop to the vector, clear the flag, then return).  It's my guess why the filament sensor is always having issues: they appear to clear a register of interrupt flags without really knowing what the other flags are doing - even if one is pending.

RE: Garolite bed crashes

Posted by: Tim

Thanks.  Worth noting is the M117 defect is still there and my bug report for it is still hanging fruit. Maybe your sample code will be enough to get someone off their fanny to dig into and actually fix the two issues.    

Sounds like another case of poor interrupt handling (pop to the vector, clear the flag, then return).  It's my guess why the filament sensor is always having issues: they appear to clear a register of interrupt flags without really knowing what the other flags are doing - even if one is pending.

What number? I've searched the bug list for M117 and can't find anything matching your description.

RE: Garolite bed crashes

https://github.com/prusa3d/Prusa-Firmware/issues/1443