MakerWare stuck when slicing

[Tad]

Hi guys,

I'm trying to print this model http://www.thingiverse.com/thing:283641

Unfortunately MW 2.4.1.24 freezes at 53% every time I try to slice it (no matter what settings I choose).

Can anybody have a look at this and tell my why? If its not possible to slice it on MW - could you guys slice it for me (for Replicator 2x, right extruder, 0.3 or 0.25mm) using other slicing software?

Let me know what you think,

Tad

[Ryan Carlyle]

Just let it run longer. MBI's progress bars are NOT accurate. Some models, particularly ones with complex extruder paths (like your bajillion holes here) just take the slicer a long time to figure out. 

Slicing time that depends on the part orientation is a good sign that the slicer just gets bogged down with the model and needs more time. Slicing in the as-downloaded STL position (upright on edge) it only took me about 3 minutes to slice at 0.2mm with 30% fill, no raft, no supports. But there's no chance in hell it'll print as-is since it's tall and has a tiny contact area with the bed. So I oriented it face-down with similar settings and now it's hanging at 50%. I'm pretty sure it'll work in, say, a few hours.

Leave it overnight and if it's not done in the morning, you might have actual issues.

[Tad]

Thank you Ryan.

It is obviously designed to lay flat.
I'll leave it over night. I have quite a decent rig - 6 core CPU with HT and 8gb of RAM. Yet there is barely any CPU activity after an hour of slicing. I'll report back after leaving it on for 8h.

Tad

[Ryan Carlyle]

Makerware can't utilize processor power for squat. I don't know if it gets stuck in a little loop trying to optimize tool paths or what, but it just seems to chug away for hours sometimes and then DING! It finishes fine.

You can always try other slicers, too.

[Tad]

I gave up after 7h - still stuck at 53%...

I will try again over night (I live in NZ).

I have also tried to process it using Slic3r - this produced 25mb gcode file. Converted the new .gcode file to .x3g format using gpx converter with the following command: "gpx -v -g -m r2x Filename.gcode"

There is something wrong with the new converted .x3g file as my Replicator 2x makes funny noises when I try to print it. LCD also indicates that extruder will only heat up to 120 C - which is not what is written in original .gcode file (230C).

I'm a little bit lost here...

[Tad]

Just to clarify - Slic3r took less than 15min to complete the job with my item in correct horizontal position...

[Steve Johnstone]

Hi Tad,

Slices fine with Simplify 3D - took under a minute on my machine :)

Layer height - 0.1mm

5 x Top Layers

5 x Bottom Layers

2 x Shells

50% infill

[Steve Johnstone]

On Friday, 28 March 2014 22:38:07 UTC, Tad wrote:

[Tad]

Hi Steve,

Thats great news!

I have left MW over night to finish the job - this time its 10h+ and only 44%... I'll stop at 24h mark.

Looks like one has to pay for Simplify 3D $140 and I'm not too keen to buy it just for this one model.

Would you be able to slice it for me again at 0.25mm and 30% infill settings and upload it somewhere?

Cheers,

Tad

[Steve Johnstone]

Hi Tad,

OK I have sliced it as requested, but please bare in mind that I have used a profile that has been tuned to my 2X & ABS filament.

Also note that the start code is different to MW, so the bot will behave differently at the beginning. Mainly once everything is to temperature it will move the extruder head off the front left hand corner, purge the extruder and then wipe to the start position.

Because of the size of the print you may have problems if the build plate is warped or isn't perfectly setup.

Good luck, let us know how you get on.

You can download your sliced file here - Board.x3g

Cheers, Steve

On Friday, 28 March 2014 22:38:07 UTC, Tad wrote:

[Tad]

Thank you Steve,

I was able to start the print with your file. Print failed on first layer exactly as you predicted. 

My build plate is warped way to much to allow build of such a wide object. I have just checked - its 1.5mm when cold and there is nearly 3mm difference when buildplate is heated to 110C. I will look into this within next few weeks as my 2x should still be covered by Australian warranty.

I've always been worried about the shape of my build plate. However due to the timeframe of my original project (which has now been successfully completed) I was unable to send my Replicator back for repair. So far I've been printing pretty much only on the central part of the plate. 

Once again - Thank you very much for your help Steve!

Cheers,

Tad

[Steve Johnstone]

Sorry to hear that Tad,

I would get in touch with MBI support and ask for a replacement. Just bare in mind that that does't guarantee you will get one that's not warped - my second one was just as warped as the first. I didn't bother going for a third - I'm printing on cheap picture glass which works brilliantly. 

I posted my glass build plate system on Thingiverse with some how to videos a while back - LINK

Also have a look at this great solution by Adam P, to minimize the the heat induced warping - LINK

I did a short video following Adams and and others advice which may also help - VIDEO

Let us know how you get on,

Cheers, Steve

 

On Friday, 28 March 2014 22:38:07 UTC, Tad wrote:

[Tad]

Thanks Steve,

Your last post is just exactly what I was looking for - excellent summary of existing solutions to address warping problem.

I'll start with replacement build plate first. Than I think I'll go for Adam solution, and if that fails - glass plates.

I understand that one needs to use hair spray to print ABS on the glass surface?

I wanted also to give you guys and update about my slicing problem. I have left MW on the job for a little bit longer - just as Ryan advised. I have to report that it finally managed to successfully slice the object. It took MW mere... 40h and 2min to get the job done....

Cheers,

Tad

[Ryan Carlyle]

My theory is that every time MW's slicer path hits an obstacle (hole, edge), it has to decide which way to bounce to continue filling the layer -- continue zig-zagging in the same direction, or double back. Optimizing this path requires checking permutations to see which is shorter. When you have a large number of obstacles, the binary decision tree grows exponentially and the slicer gets bogged down.

For a shortcut to check if the file will eventually slice, you can shrink it to 5-10% and if it slices in miniature, it should eventually slice the full-size file.

[helibrdr]

I am with Steve... The MW slicer can get hung up on complex geometry. While slicing a heart model from CT scans I waited over 8 hours to slice a part. Ended up purchasing Simplify 3D and it sliced and post processed in under 5 minutes...