Print failure when printing vertical columns?

[jjenco]

Okay, so I designed this prototype that could generally be described as a 7" tall vertical hollow column with several proprietary features. I have tried to print it twice on my Rostock MAX v2 w/ upgraded 280 print head. Other smaller (shorter) pieces seem to print fine. I'm printing Carbon Fiber PLA. The print job is set to 'Stardard' and still calcs out at around 19 hours and change to complete.

In both cases, when I check the print job first thing in the morning after printing all night, I find the print head stalled at a completed height of 3-4 inches. The print head and bed have returned to ambient temperature and the print head is stuck on the top of piece where it stopped. I have to reheat the head in order to free it.

First question, what the heck is going on? I've never had this happen before, but then I've never tried printing a piece that took this long either. At first, I thought it might have something to do with a 'middle of the night' software update, so I turned off my network connection for the second attempt but it still failed sometime during the night.

Second question, at present I have a 3-inch tall semi-completed piece on the bed. Is there a way to have the printer pickup where it left off and complete the piece? I figure this piece has about 9 hours of effort into it already.

Sorry that i can't post a picture but any help would be appreciated.

Thanks!

John

[Xenocrates]

Are you printing via USB or from an SD card? It does sound like it's USB. You may want to try using an SD card to avoid a reset USB connection killing the job (Loose cable, driver glitch, polling rate mismatch, etc. USB to serial is hard) If you want to debug the USB, make sure to open the serial console to the printer before starting to print, and look at the last few entries in it.

There are ways to start a job from an intermediate point, however my recommendation is to start a new print, without the completed part, and solvent bond the two together (Super glue debonder is a relatively available solvent for PLA, and the carbon fiber is not continuous anyways, so while it will be a bit of a weak point, it will produce a part.)

If you feel an absolute need to keep your design proprietary (At least a few of us understand that small bussiness's need to worry about that), you may want to try with a cheaper unfilled PLA, and a similar but simple design (Such as a tower with hollows in it that relate to whatever interior radii or overhangs you have, but lack the functional features) that you could share for troubleshooting. I'm betting this is a relatively simple error in either the slicer or communications (Or possibly a power glitch in your home/office)

[jjenco]

Xenocrates:

Thanks once again for offering some great observations that suggest a pathway forward! You're right, a USB connection. I did't think about that. I'll try reprinting from the internal SD card. Will let you know how that works out.

I would like to slice and print with Cura, but I get a 'printer not found' error when trying to use that software so I continue to use MatterController. Sorting out that issue is my next challenge.

[geneb]

You would slice with Cura and print from the SD card. Cura isn't really meant to be a printer host other than for Ultimaker printers. That being said, the network link from Cura to the Duet works great!

g.

[jjenco]

Thanks! Having trouble at present printing from the SD card. If it's not one thing, it's another! Working through the issues...

When I slice in MatterControl and save the g-code file to the card, then insert the card into the printer, the display shows "SD Card Inserted" but it does automatically show available files to print, per the user guide instructions. I've tried navigating to the SD Card menu on the printer but nothing shows up there either. Also, when i remove the SD card from the printer, the status does not change; it still says 'SD Card Inserted". Any thoughts?

[geneb]

Might be a dodgy SD card.

g.

[jjenco]

Might be a dodgy SD card.

g.

Is there a maximum SD card storage size that the RostockMax v2 can read? I think i'm using a 32GB card.

Thanks

JJ

[geneb]

I've never used larger than an 8, but I don't know of any inherent size issue. It does need to be FAT formatted, though. The system can't read an NTFS-formatted card.

g.

[jjenco]

I've never used larger than an 8, but I don't know of any inherent size issue. It does need to be FAT formatted, though. The system can't read an NTFS-formatted card.

g.

Reformatted the card to FATS and it is recognized fine. THANKS!!

[geneb]

\o/

You're welcome!

g.

[jjenco]

\o/

You're welcome!

g.

So, in the continuing saga of trying to print this vertical column, changing to direct-print from the SD card in the printer seemed to be a solution. The first time printing from the card the job got about 80% of the way done and then the printer somehow knocked over the print job in the middle of the night after about 15 hours of the 19+ hour print job. The result can be seen in the attached pictures. I woke up to find the column knocked over on it's side on the bed, with a ball of filament all over the bed.

I tried a second time and once again I found the column lying on its side the next morning with the printer spitting out filament all over the bed. In both cases, the printer was able to complete about 5.5" of the vertical column before whatever happens happened.

I used a raft for the piece. Attached are photos of the mess.

Thoughts?

Thanks

John

[Mac The Knife]

I would try printing with a brim attached to the print,,, possibly as high as 3 layer thick.

[geneb]

Make sure you've got your slicer configured to lift the nozzle on retract - z-lift or z-hop.

You can also add a brim to your print that will give it increased surface area on the bed to improve the part's hold on the bed.

g.