First upload rejected for volume/size.

I'm uploading this OBJ file, see attached.

It seems quite small, but it keeps getting rejected for size/volume. Since this is my first model, I was wondering if someone might have a look-see and tell me where I'm going wrong.

Thanks

Bounding box: 0,03 x 0,05 x 0,07 mm

That's not "small", that's "tiny". ^^
Probably, it's getting rejected because it's way too small to be printable.

How big (or small) was it intended to be ?



btw,
it has 32 border edges, 8 holes, and 9 shells... no good.

You should download a free copy of netfabb (basic) to check an repair your files.

Very helpful, thank you. First model, so I'm learning the ropes. That software will probably help. Also blender made changes to its editing/geometry methods which take away a lot of control.

What changes are you talking about exactly?

Quite a few, but the one I'm referring to is when you are modelling, there isn't as much direct polygon manipuation - it assumes a lot of stuff for you. Its quite confusing and I've had my doubts about it since it started, especially because I hate tris!

See link:

http://wiki.blender.org/index.php/Dev:Ref/Release_Notes/2.63 /BMesh

Your model seems to have been setup with measurements in meters.. if you pick "Meters" when you upload it, you'll have a reasonable sized object. Otherwise, you need to re-scale.

To help hilight the issues, please see attached image. I separated the shells by a bit of distance and added a bit of color to help explain.

The yellow areas are holes.. the two red areas are single-sided planes with zero thickness.

The two red shells should be removed..The holes should be filled, or you may want to union those two larger parts together

Very generous use of time! Thank you. Since going through all of this and seeing thread remarks I've decided to remodel my object over again with some simpler and more predictable shapes to avoid those "gotchas" which I seem to be running into.

A bit off-topic: I haven't played much with the new version of Blender, as I have been doing some strictly academic investigation (yeah, that's the ticket!) into a certain series (cough Mass Effect).

Regardless, I think you may be misunderstanding the BMesh features? Overall, this seems to be improved, and not in the "less control" direction. Perhaps I can get back to you after I play with it some more on my next batch of projects.

Alrighty - back to the drawing board and ready to go again! This time I was a little more careful.

See attached OBJ file if you like.

My concern is that this is accurate in 3d, but perhaps it will be brittle at its size, or even a little inaccurate due to printing limitations? I'd love to know.

I plan to drill in the axles myself for the balls to move on. How does this material react to that?

Apparently it was successful, straight from blender, and the size is accurate. https://www.shapeways.com/model/608997/4ee0ad5e7e24bb15ea554d 6b47b1553c

Aahm BMesh... sometimes with Blender the best is to run older versions instead of the last. That system is too untested, people is reporting problems with subdivide and knife in really basic cases. Converting to triangles and then quads to get rid of the ngons just makes things even worse, with inner edges and faces. And the tesselation just sucks in general. You are not alone, see comments by Philippe ROUBAL for example, or any other there showing a bit of vision beyond "wow, flat surfaces without edges".

They better fix the bugs, finish some features and provide extra tools to be able to work in tri/quad mode without having to fix ngons side effects every time you use a tool. Otherwise it is too cumbersome for anything where real surface shape and topology matters. If you search in this forum, you will see people bitten badly, in other sotfware, for similar reasons. Bevel or vertex slide are nice, but not by getting weird surfaces (the non planar quad issue, but to the N power) or worse things backfiring.

You've got the model marked private in your shop.. other members can't see it.

Which material were you going to try to print it in? I've had good luck drilling both WSF and FUD.

The flat side of your rollers are not "flat".. there's a bit of a convex curvature to them.. I'm concerned that they may fuse down toward the center. But.. for your design here, there's no real nead to have the rollers "in place".. you could move them out as independent peices so that they won't fuse to the frame.

Another option you have is .. depending upon how much pressure would be applied to the final product.. you could actually print the axles in place so that no drilling is needed.

Take a look at: http://shpws.me/3Rrl the whole thing is less than a 1/2 inch cube, but all four axles turn freely.