Solved Glass bed mysteriously shattered during failed print.

The title pretty well states the issue. I monitored the first few layers of the print, then left it unattended for a few hours. When I came back to see if it was done, I found the print failed and the glass print bed shattered. As far as I can tell, there isn't any obvious cause. I've attempted to include as much information as possible in order to help determine the cause.



The pieces of glass on the floor are not in there original positions-- I didn't have the presence of mind to take any photos of them in situ.

A few pieces remained stuck to the painter's tape, and a few others are still just barely attached to the rest of the bed, though the glass is cracked all the way through.



The extruder head was way over on the other side of the bed from where it was supposed to be printing; this is a far larger x-axis shift than I've ever seen before.



I checked the belt tension and it seemed fine. Strangely, there is a much smaller axis shift on the print that only seems to have lasted for 1 layer.



The filament offers a clue as to the likely cause of the massive shift. The end was shredded and, more notably, not in the extruder. I would guess that it somehow got stuck on the spool holder, pulling the extruder head all the way over to the side, where it was shredded as the hobbled bolt continued to turn.



That would certainly explain the shift; as stated at the beginning of the post, I'm trying to include as much information as I can, just in case. Anyways, the mystery deepens.

In the area where the extruder head was, the nozzle had scraped off the painter's tape and deposited a small amount of plastic on the glass underneath.



An inspection of the z-axis rods revealed that the nuts had come out of the x-axis carriage and were all the way at the bottom of the threaded rods. The nozzle must have been resting on the print bed, but this wouldn't have been nearly enough force to shatter the glass.



I have no idea what could have caused the z-axis rods to reverse direction during a print, but that seems to be what happened. As far as the software was concerned, the nozzle was 45mm above the bed. I don't think it's a firmware issue, as the printer has been running fine since I replaced the Arduino MEGA2560 because of USB connectivity issues, which was right after I got it. The problem doesn't seem to be in Repetier-HOST either. I scrolled all the way up to the beginning of the log, but there wasn't anything out of the ordinary. Could the Gcode have somehow been messed up and caused this? I don't know how to tell if there's anything wrong with it, but I've attached the Gcode file just in case anybody's really determined.

As far as repairing the printer, I can't justify spending $89 on a new bed, I even if I could I wouldn't be able to install it, since I have the first generation printer with the drawer sliders. I was apparently mistaken when I assumed that the linear bearing upgrade for the y-axis would be available for more than a brief time. I might replace the bed with a piece of aluminum, but I don't have any way of cutting it to size with reasonable precision.

Any ideas on what might have caused this and/or suggestions for replacing the bed would be greatly appreciated!

Thanks in advance,
Cybus

 

Did you happen to have the spool holder mounted on the top?

You can buy a replacement bed here, But they don't have screw holes anymore: http://www.robo3dprinter.com/collec...cts/replacement-heated-borosilicate-glass-bed

 

No, I never got around to printing a top-mount spool holder. I think that was the cause of the problem. I just switched to a new filament which is on a different sized spool, and it doesn't match up as well to the length of the side-mount spool holder. I think the filament got too much slack in it, so a few coils slipped off the spool and got wrapped around the spool holder rod. The extruder continued to pull in filament, which forced the extruder assembly all the way to the right. Once it was all the way over to the side, it pulled itself up, until it chewed all the way through the filament and fell, causing the nozzle to hit the bed and break the glass.

I considered just making my own replacement bed with tempered glass and a heating pad, but it wouldn't have been much cheaper than buying an official Robo replacement, and I want to upgrade the y-axis anyway. For the time being, I have a temporary bed set up, with a wooden platform screwed to the rails and a piece of ordinary glass I happened to have laying around clipped on top. It's ugly, but perfectly serviceable for printing the upgrade parts.



All the upgrade parts will be printed at 250 micron resolution with 40% infill.



However, the piece of glass in the above photos isn't big enough to fit the bed mounts, so I picked up a bigger piece intended for a picture frame at Michael's for $6.99.



Because of the size of the wood I happened to have on hand when I made the temporary bed, it's only 12" long, so I had to tighten the belt, which restricts the print area to 180mm in the y-axis. It's a tight fit, but the bed mount just fits on the available print area when rotated 37˚.