pauldebra
Posts: 4
Joined: Fri Sep 02, 2016 4:16 am

model looks fine, prepare to print shows rubbish

I designed part of a (Flashforge Creator Pro) spool holder in Sketchup.
I tried many different fixes to make sure the faces are in the right "direction", that there are no hidden internal planes, etc.
Flashprint has no trouble to generate usable gcode for this and it prints fine.
Simplify3D shows the model correctly but generates completely bogus gcode.
If I try a few of the repair actions and then prepare to print Simplify3D crashes.
For instance: open stl file, remove duplicate triangles, prepare to print ==> crash.
Does anyone understand what is going on here?
I am relatively new to Simplify3D but so far have managed to correct all issues (like wrong face direction) and got Simplify3D to produce correct prints, but this model is stubborn.
I can email the sketchup file if desired. (The forum does not let me upload a file with extension skp.
Attachments
spool-holder-base.stl
STL generated from sketchup.
(217.14 KiB) Downloaded 316 times
CompoundCarl
Posts: 2005
Joined: Wed Aug 05, 2015 7:23 am

Re: model looks fine, prepare to print shows rubbish

All you have to do is go to View > Cross Section to see there's lots of issues with your model. There's a bunch of internal surfaces and self-intersecting bodies. It definitely isn't a proper water-tight solid model, so you need to fix that first.

You need to actually create valid 3D solids instead of just surfaces, and then make sure you truly boolean those together when you want to combine features. So either fix it in the CAD package, or potentially try using an STL repair service
User avatar
dkightley
Posts: 2405
Joined: Tue Mar 10, 2015 4:09 pm

Re: model looks fine, prepare to print shows rubbish

Using such a poor 3D package as Sketchup (in my opinion, that is), you're bound to get issues as you've described.

S3D is a slicing package....and as such is good at what it is designed to do. To sort out defects in the shape file, you need to use something that is designed to sort out defects.....and there's a free-to-use service that can do exactly that!

https://netfabb.azurewebsites.net/

I've run your file through this service....and the repaired file is attached.
Attachments
spool-holder-base_fixed.stl
(27.33 KiB) Downloaded 272 times
Doug Kightley
Volunteer at the National Tramway Museum http://www.tramway.co.uk
Railway modeller and webmaster at http://www.talkingtgauge.net
pauldebra
Posts: 4
Joined: Fri Sep 02, 2016 4:16 am

Re: model looks fine, prepare to print shows rubbish

dkightley wrote:Using such a poor 3D package as Sketchup (in my opinion, that is), you're bound to get issues as you've described.

S3D is a slicing package....and as such is good at what it is designed to do. To sort out defects in the shape file, you need to use something that is designed to sort out defects.....and there's a free-to-use service that can do exactly that!

https://netfabb.azurewebsites.net/

I've run your file through this service....and the repaired file is attached.
Many thanks. This repair service is a great find.
I do know that Sketchup is rather limited but it is easy to use. I do believe the sketchup to stl converter must be messing things up.
The repaired file slices perfectly. I guess I have to get used to Simplify3D being much more picky about what's in the stl files than Flashprint or even my old ReplicatorG version 24 from something like 6 years ago.

Still, no matter what is wrong with my stl file it's not ok that Simplify3D crashes (on Mac OS at least) when trying to repair and then print. Whatever it does wrong it should not crash.
dorsai3d
Posts: 237
Joined: Mon Jan 11, 2016 9:01 am

Re: model looks fine, prepare to print shows rubbish

I've noticed that happens with operations that remove triangles from a model. I'm not sure why, but it happens with remove duplicate triangles as well as mesh reduction, but only if you have just done it to the model. If you save the model and reimport it, it works fine.

My guess is that it's some sort of issue where what's in memory isn't matching the new reduced model and something's referencing triangles that aren't there anymore when you go to slice it. This fits with the exception type (EXC_BAD_ACCESS on OSX, invalid address at all 0s) that looks like trying to reference a null pointer.

Return to “Troubleshooting and Bug Reports”