Freesteel Blog » Knobbly Knees

Knobbly Knees

Thursday, September 20th, 2007 at 1:41 pm Written by:

The world is fat and the resulting crumbling knees presents a business opportunity.

I’ve got a got a small piece of ray-tracing software feeding in to a project that Liverpool University are doing in the field of “Rapid Manufacture of Industrially Relevant Hierarchical Structures”, where Relevant is for Stryker Orthopaedics. This builds your implants using lasers and powdered metal to create a porous product which your bones can grow into.

I’ve watched this project happen over years, so it’s interesting to learn of the competition which is cutting its knee implants on a machine tool using Camworks software for Viasys Healthcare who then etches it. Viasys also sells TrabeculiteTM, a fully-engineered, porous, titanium structure. I thought these things were patented. Anyway, this means someone out there has written the same ray-tracing function as I have. It’s quite likely because it’s rather low-level. Shame it’s not all open source, or it would all be better.

Where else can I find a familiar-looking algorithm?

Teksoft is showcasing CAMWorks 2008 on stand G18, hall 6 at the EMO 2007 exhibition in Hannover… New toolpath strategies and controls have been implemented for 2-5 axis milling, turning and wire EDM to improve quality and decrease machining time. For example, significant enhancements in 3 axis milling include:

* Adaptive roughing strategy provides the ability to cut using the full depth of the tool and safely run your machine at optimum speed, which can reduce machining time up to 40% over conventional roughing with less wear.

In January, the CEO of Teksoft said:

We believe that the CAM developers need to get rid of the ‘not-invented-here’ philosophy and transition to a license the best-of-class technology and integrate it strategy.

In September in an article promoting the idea of modularity among developers of programming software the CEO said:

“I’d need around 400 people if I were to match all of the R&D engineers that I leverage into my products through these ISV [Independent Software Vendor] and third-party associations around the world… In the past, the strategy was to develop everything yourself… The downside of doing that is you can’t offer the best of class in everything. You typically have a technology that you excel at, and are mediocre at everything else. So the customer might get the best three-axis milling package available, but get a lousy lathe or EDM package with no sheetmetal capability. That kind of a solution no longer fits today’s market expectations…

“Rather than pretending that a couple of guys in the back room can come up with everything that we need, we buy our HSM algorithms from a third party that devotes 10–15 programmers to developing just the HSM modules.”

I suppose it’s satisfying to see the businessmen finally catching up to what I was thinking five years ago at the start of this adventure. While the CAD market had gone through this process in the late 1990’s with companies being proud to say they were using the Parasolid kernel or the ACIS kernel, the CAM market has largely chosen to keep everyone in the dark whilst a kind of consolidation has been going on in the background forced upon them against their will owing to a dire lack of suitable software developers.

Of course, when a couple of the few very capable software developers makes themselves available, it always helps to completely ignore them for years and deal only through hard-nosed businessmen whose diabolical business plans don’t leave enough room to buy lunch, because the mere creators of software are just too lowly to even give the time of day to. Companies do not make software. A small number of people at their desks make software. And they do a lot of other things as well, after having discovered that working for companies is a totally sh!t way to spend your life.

3 Comments

  • 1. Garry B replies at 21st September 2007, 9:30 am :

    I’d be impressed to see a team of 10-15 programmers at any company just working just on HSM. Sometimes I get the feeling there are only 10-15 people worldwide working on it.

  • 2. Neel D replies at 21st September 2007, 12:07 pm :

    I think Liverpool University project is based on Direct Metal Laser Deposition (DMLD), which is also referred to as Laser Cladding, a laser beam is used to form a melt pool on a metallic substrate. Powder is then fed into this pool which melts to form a deposit that is fusion bonded to the substrate. Both the laser beam and nozzle from which the powder is fed, are manipulated using a robot or gantry system. DMLD is already an established technique for the repair of turbine components in the aero-engine and power generation industries.
    At the moment current commercial cam systems will not be able to handle the process as the algorithms are designed based on cutting process and not deposition. The toolpath generated are used to clean excess stock.
    For deposition process we will need a completely new set of algorithms.

  • 3. Julian replies at 24th September 2007, 11:03 am :

    I saw that machine a few years ago. The results were ropey. Now they’re doing just straightforward laser sintering (powder beds layers and laser melting). While the full CAM algorithms aren’t reusable, the primitive and highly optimized functions which these CAM algorithms depend on are.

Leave a comment

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