Daily Archives: September 8, 2012


Help! My GCode is borked! 1

Do you have prints that don’t print?A pronterface that looks wrongterface? Well, you might have a bad mesh.

I ran into this when attempting to slice a whistle from Thingiverse: Slic3r would give me the ominous “Non-Manifold mesh” message and the printer would spaz out.

In order to undertand what’s going on, there should be a little information on how tools like slic3r work. Slic3r and other G-code generators tend to work on the values of face “normals” — the direction the face, well, “faces.” Let’s take our whistle for example: As downloaded from Thingiverse, it looks like this:

Bad STL file from Thingiverse

The tool I use is ModuleWorks STLView (also for android!) which allows us to show the face normals (green lines) and faces which face the wrong way (gold surfaces). Slic3r will produce some very, very strange GCode for this model:

Ugly GCode as a result of bad mesh

The tool here is the RepRap GCode Simlulator from opencode.eu.The source is included, and depends on openGL. This lets us see what the actual path for each layer of the model looks like.

So what do we do? First: don’t panic. Most of the horrible problems with ugly looking meshes can be fixed — all it takes is some time with the NetFabb cloud service. NetFabb produces a commercial 3D printer, but also runs a cloud service at http://cloud.netfabb.com/¬†which allows you to upload STL meshes and download the fixed version.

Here’s what the model and resulting GCode look like afterwards:

Fixed, good model with all its normals pointing out

Good GCode as a result of a clean mesh



Wednesday Workshop: Pin-Hole Cameras, long exposure photography, and easy at home developing

Wednesday September 12th, Im going to be teaching a small informal workshop on lone exposure experimental photography via pinhole cameras. we will be going back to the very roots of photography.

As a long time photographer its easy to get caught up with the latest and the greatest in technology. But a lot can be learned by going way back to the start of it all. I wanted to get into some pinhole Solargraphs(photos of the sun streaking across the sky) but i didnt have a good darkroom handy.

So I developed a technique for using every day metal food tins to make small pinhole cameras, and then loaded them with blueprint paper. (blue print paper is nice because it does not need precise timing to develop, and can be developed with simple and fairly safe household chemicals, and you don’t have to work with negatives!

Here are some examples.

more examples and some of my cameras can be seen here,
https://secure.flickr.com/photos/killbox/sets/72157626481399715/

This Class we will talk a little bit of theory about How long do these long exposures need to be.
Camera making, (select your tins, and pierce your pinhole)
About the paper, and its handling.
Developing the film

Everyone signed up will go home with atleast one camera, and several sheets of paper (10 or more photos possible per sheet).

Date:Wednesday September 12th.
Time:7-9pm
Location: 1112 2nd street NW
Cost: $15 (you can choose to pay with cash if you rsvp look for the “Show other payment options” which will save you a little bit of service fees!)
RSVP now! http://pinholecamera.eventbrite.com/