Freesteel Blog » Slicing of STL now online

Slicing of STL now online

Thursday, December 4th, 2008 at 4:34 pm Written by:

Check out the online version of our STL slicing application.

I’ve used ajax for calling a server script to do the actual slicing.
The link should also work in Internet Eploder (I’ve tried it in version 7).


  • 1. Matt Bennett replies at 15th September 2009, 10:58 am :

    Hi Martin

    I’m just ‘experimenting’ with cam software coding and various algorithms for generating toolpaths. Nothing as fancy as adaptive clearing mind, I’d be happy with rastering and z-leveling for now, with raster based area clearance to follow.

    I’ve written my z-slicing code today, and your question/response (Why has my Z value been moved slightly?) addressed a fundamental issue before I even knew it existed so a big thank you there.

    I appreciate this probably isn’t the place for asking questions but I’m wondering if the winding order of STL triangles is guaranteed (at least if they come from a reliable source such as Solidworks), since it affects so much of what I’m doing.

    Anyway, keep up the good work, and I look forward to your (and Julians) blog updates.


  • 2. Martin replies at 15th September 2009, 3:34 pm :

    Hello Matt

    I am afraid that you can’t rely on a triangulation being consistently having the triangle’s normal vectors pointing “outwards”. The concept of “outwards” only exists for solids, but what do you do if the model you’re meant to machine is not a solid?
    Maybe you can explain why you need to rely on this assumption for your machining algorithms?

  • 3. Matt replies at 16th September 2009, 9:39 am :

    Hi Martin, thank you for your response.
    At the moment this is all proof-of-concept stuff so my approach is pretty naive.
    I appreciate the surface orientation problem but I know very little about stl (not that there’s much to know) and need to understand where I can safely make base-line assumptions. For example; in my test stl files, the triangle vertex order exactly matches the winding order required by OpenGL. Is this a coincidence, a written rule (I can’t find an authorative stl spec online) or unwritten yet expected behaviour?
    Since my output systems are limited it would be helpful to hear from someone used to dealing with any anomalies.
    Thanks again

Leave a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <blockquote cite=""> <code> <em> <strong>