Makerware stops slicing at 58% and hangs, different objects same outcome.

[Andrzej Raczynski]

Howdies,

One of our loaner Replicator 2 units (we lend them out to schools to try out) has apparently started doing something strange.  Its been working fine up until today when Makerware starts the print process and stops/hangs the slicing progress at 58%.  She's tried different objects (2 from thingiverse and one from the Makerbot digitizer that they also borrowed).  They've reinstalled the Makerware software (they're on Macs) but nothing has helped.

I probably should've asked for a log file from her, but I haven't spent enough time on the Mac version of Makerware to be able to tell her over the phone where the .log is at.  (if anyone has a link handy for a website with a walk-through for that I could forward it to her).  The unit is about a 100 miles from my office, so not readily available.

In any case, anyone run into this weirdness before?  cursory search here didn't turn up anything.

TIA
andy

_______________________________________________________________

This e-mail (including attachments) is covered by the Electronic Communication Privacy Act, 18 U.S.C.  2510-2521, and is confidential.  In addition, any student related information contained in this e-mail (including attachments) is protected by the Family Educational Rights and Privacy Act of 1994. If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution or copying of this communication is strictly prohibited. Please reply to the sender that you have received this message in error, then delete it. Thank you.

[Ryan Carlyle]

Let it run all day and it should finish slicing. MW very often looks like it's hanging when it's actually just stopped updating the progress bar while it works. I've had prints "hang" for 3-4 hours and then finish just fine.

If it's truly crashing the program, that's a bit different.

[Jetguy]

PS, it has NOTHING to do with the bot.

This is all the model STL you are trying  to slice. If they developed from a CAD software, likely it's full of flaws. At a minimum, my experience has been at the typical school they had better run EVERY file through Netfabb before even thinking about trying to slice it with MakerWare.

[Andrzej Raczynski]

Thanks guys, I'll ask her to let one of these run for a couple hours if need be to see what happens.  Now that you mention it, I do recall seeing something similar once when I was printing something from thingiverse, didn't think anything of it at the time and forgot all about it.  Didn't take hours to finish slicing, but did 'stop' for an unsettling amount of time.

[Andrzej Raczynski]

that turned out to be the case, she started the process at 3pm, and said when she came in in the morning, it was finally printing (must have started only an hour or so before she came in as it was printing bulldog feet on about 8 copies, but ran out of filament :))

thanks.

[Scottbee]

So she was printing over USB in lieu of writing to the SD card?

[Andrzej Raczynski]

Yes.  I generally explain to the schools that they should use USB vs the SD route to force them to take the objects they find (Thingiverse) or create (Autodesk Inventor/etc) through Makerware for visual verification of scale/placement/etc before actually printing.  I'm not sure how 'intelligent' the directly-from-SD route is in terms of what gets 'fixed' automagically for them, so I err on the side of caution and just tell them to do it that way as a confirmation.

I realize they can export from Makerware onto the SD card and pop it in, but then of course they'll quickly deduce that there's no point in even going to Makerware and just dump everything directly to the SD and bypass the above process.  Now I could be completely wrong with my logic, which I'll admit, and you're more than welcome to tell me I'm smoking crack rocks here :)  I just 'figure' it will lead to less issues overall (?).

The USB route, in a way, just 'tethers' them to the process above.

[Scottbee]

Unless you are directly downloading x3g files from Thingiverse (not too common) you can't directly print them anyway.  You need to have some process to convert the STL (common download format) to x3g for your printer (which is also where it gets its temperature, layer thickness, speed, and etc.. info).  That has to happen in the slicer (MakerWare, S3D, etc.).

The only reason I asked is because the pros here will tell you that direct USB printing is more error prone due to communication weaknesses in the bot.  That's all.

[Andrzej Raczynski]

Ah, thanks for the clue-in.  I was under the impression that the STL had all the info the printers cared about.  Glad you corrected me.  The printers/Makerware tend to be hooked up/paired to laptops at schools/here in the office, and that's usually at best a hit-miss proposition for SD slot availability (vs USB ports).  For example I tend to pair my bots with a Surface Pro when I work with them, which has no SD slots.  I realize there's plenty of SD/USB adapters out there, I have a couple, but they tend to annoy me to deal with, and would almost certainly disappear quickly enough when sent to a school.

I remember reading about the SD over USB recommendation, but I find it works 'well enough' 99% of the time.

Anywho, thanks for clarifying about the STL/X3G, appreciate it.  I think I will reevaluate my process in light of your info.

[James V]

You're best off doing the slicing in makerware (or sailfish(!!)) and loading it on a SD card anyway. This prevents the scenario you're encountering now, as well as preventing unrelated errors from stopping the print.

When you're printing, you must maintain constant contact between the computer and the printer. This is resource intensive and it leaves the door open for another program to hang the system, aborting your print. When you export the process to the SD card, you don't have to worry about any of that and can continue working as you normally would. This is especially useful for laptops, as you enable the user to resume mobility rather than being chained to a printer.

As far as the slicer hanging, as the others have said, an error free file is critical. Also important is the quality of the computer in question. Generating toolpaths for larger objects consumes a great deal of computing power, and if you don't have much to begin with, you'll have slow or no results. The method I'm recommending now is best because you can do all your slicing on your most well equipped computer and let the printer do all the rest of the work.

On Wednesday, March 12, 2014 3:38:45 PM UTC-5, Andrzej Raczynski wrote: