Errors in Mendel STL files

Re: Errors in Mendel STL files

mccoyn — Thu, 17 Dec 2009 21:01:02 -0500

I just spent an hour trying to get one of these files to work. Could someone check the corrected files into sourceforge? I downloaded them here.

Re: Errors in Mendel STL files

gerard94 — Wed, 18 Nov 2009 11:39:39 -0500

Sorry, I meant thicker, not wider.

Gérard

Re: Errors in Mendel STL files

gerard94 — Wed, 18 Nov 2009 08:10:40 -0500

Yes, I have. I've added a new slightly wider version of z-drive-pulley-rim_4off.stl, that I've called z-drive-pulley-rim_4offN.stl.

Gérard

Re: Errors in Mendel STL files

letsburn00 — Wed, 18 Nov 2009 06:46:11 -0500

Has anyone made a collection of mendel files that are already flipped, because I've found most of the bigger parts are upside down.

Re: Errors in Mendel STL files

gerard94 — Sun, 25 Oct 2009 21:28:38 -0400

Wow, 6 months! I'm not surprised, but it's long. Perhaps I'll do that some day, but, for the moment, it's nice to meet someone with your skills.

And good luck for your RepStrap. Beginning with scanners is not common.

Gérard

Re: Errors in Mendel STL files

Grogyan — Sun, 25 Oct 2009 20:51:25 -0400

Blender is harder to learn, but for me it feels more natural than any other program, even against the pro packages like 3D Max and Maya. I got into Blender because it was the only program for which a complex plugin called PyPrp which essentially allows you to design a world with puzzles for the game Uru from Cyanworlds.
Blender took me around 6 months of tutorials, a couple of hours a day 2 - 3 times a week.

I'm building a RepStrap based on two old scanners because i'm a robo nut, can't be too bothered to redesign it, will do after the printer is made, i've st myself till December to get it finished, else i'll ask Vik to just print out Mendel parts for me (backup plan).

Re: Errors in Mendel STL files

gerard94 — Sun, 25 Oct 2009 18:43:39 -0400

Thanks again.
You are certainly right, but, when I tried to learn a little of Blender, I found it much more dificult than AOI.

Do you plan to build a Mendel?

Re: Errors in Mendel STL files

Grogyan — Sun, 25 Oct 2009 18:08:54 -0400

Bed Spring fixed.

I had to remake the mesh because of the irregularities in the mesh, plus I fixed the small errors that Blender makes with CSG modelling (non manifested vertices).

This version is now correct with the same dimensions, with no > 45 degree overhangs

Let me know of any other problems

AoI in my opinion is not being improved, or if it is, its being improved at an unbelievably slow rate, Blender is much more powerful, with features native to it that can fix problems, but Blender does have its own bugs which will be fixed, eventually

Re: Errors in Mendel STL files

gerard94 — Sun, 25 Oct 2009 18:02:18 -0400

Thank you very much. I've tried to do it myself, but didn't understand what to do.
I also can't understand why parts that look correct in AOI, and that are made by specialists, don't export correctly in STL format. Is it the exporter module that fails?

Re: Errors in Mendel STL files

Grogyan — Sun, 25 Oct 2009 17:11:23 -0400

Here is the Y bearing, though the bed-spring will take some work to fix.
The y bearing had missing faces.

The bed spring is intact as a complete mesh, but it will not be printable due to the fact that it has unprintable overhangs > 45 degrees, so i'll fix that later this arvo

Re: Errors in Mendel STL files

gerard94 — Sun, 25 Oct 2009 06:54:39 -0400

Yes, great! It works.
Could you also have a look to these files, which have similar problems?

y-bearing-180-inner_2off
bed-spring_4off

Thanks.

Gérard

Re: Errors in Mendel STL files

Grogyan — Sun, 25 Oct 2009 04:11:37 -0400

I've fixed them, and skeinforge reports no errors back for me.

Re: Errors in Mendel STL files

Grogyan — Sun, 25 Oct 2009 03:39:54 -0400

Yep, found the problem, the problem is that there are missing faces so it can't work out how solid it is

I'll fix them

Re: Errors in Mendel STL files

Grogyan — Sat, 24 Oct 2009 18:59:57 -0400

I'll look at those files tonight with blender as I can check the normals and missing faces easily

Re: Errors in Mendel STL files

gerard94 — Sat, 24 Oct 2009 17:51:01 -0400

y-bearing-180-outer-right_1off seems very strange too. I'm not sure, but I think there are problems inside.
bed-spring_4off gives a division by zero with Skeinforge and the result with the reprap program looks very odd.

Do you get the same problems ?

Gérard

Re: Errors in Mendel STL files

gerard94 — Sat, 24 Oct 2009 16:16:49 -0400

Well, my ideas were wrong! There are really errors in the two files y-bearing-180-inner_2off and y-bearing-180-outer-left_1off in Mendel's set of RP parts. I can't convert them to gcode files without getting clogged holes and broken shapes. I used both Skeinforge and the Reprap program on the original STL files, and got similar results.

Help!

Gérard

Re: Errors in Mendel STL files

Grogyan — Sat, 24 Oct 2009 05:20:39 -0400

What I forgot to say is that when doing CSG boolean operations in Blender, it creates problems in which direction the normals go.

I've also had this error with correct normals but, some new edges created overhang, again Blender has problems with CSG, and I think they are bugs that have been fixed for Blender 2.5

Re: Errors in Mendel STL files

gerard94 — Fri, 23 Oct 2009 15:42:56 -0400

No, I used AOI.

Well, I've still not verify this point, but I think that I understand the problem now. I wanted to rotate the object, for a correct printing without overhang. But, instead of rotating the original object, in AOI format, I loaded and used the STL files. Therefore, I think that small errors have accumulated.
I'll have some time soon to test this idea.

Thanks anyway.

Gérard

Re: Errors in Mendel STL files

Grogyan — Fri, 23 Oct 2009 15:04:28 -0400

I'm not sure which programs you have, but I know that this error comes up after modelling with Blender, when you do CSG boolean operation

Re: Errors in Mendel STL files

gerard94 — Thu, 22 Oct 2009 20:12:03 -0400

Thanks.

I'll see that tomorrow.

Gérard

Re: Errors in Mendel STL files

Triffid_Hunter — Thu, 22 Oct 2009 20:01:15 -0400

from your description, sounds like internal faces of drill holes are missing or something.. open up in AoI or blender or something and have a look, they should both have some way of finding missing faces

Errors in Mendel STL files

gerard94 — Thu, 22 Oct 2009 19:23:05 -0400

Hello Reprappers,

I'm printing Mendel parts from STL files with a Rapman. I convert STL to gcode with Skeinforge. On two pieces, I get annoying errors from this program:

For y-bearing-180-inner_2off.stl, I get

This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
587 [354] [185, 205]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
18.9600000381
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
369 [180] [139, 146]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
19.3600000381
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
596 [363] [128, 191]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
19.7600000381
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
596 [363] [128, 191]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
20.1600000381

And for y-bearing-180-outer-left_1off.stl:

This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
626 [341] [189, 260]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
10.45
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
767 [469] [189, 265]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
10.85
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
767 [469] [189, 265]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
11.25
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
647 [363] [63, 84]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
11.65
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
647 [363] [63, 84]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
12.05
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
106 [38] [84, 85]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
12.45
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
106 [38] [84, 85]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
12.85
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
106 [38] [84, 85]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
13.25
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
769 [469] [245, 265]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
13.65
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
769 [469] [245, 265]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
14.05
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
789 [491] [85, 260]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
14.45
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
789 [491] [85, 260]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
14.85
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
789 [491] [85, 260]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
15.25
This should never happen, there is a hole in the triangle mesh, each edge should have two faces.
789 [491] [85, 260]
Something will still be printed, but there is no guarantee that it will be the correct shape.
Once the gcode is saved, you should check over the layer with a z of:
15.65

In either case, the holes of the part are clogged, and there is something broken in the shape.

Do you know how I can solve this problem?

Thanks.

Gérard