Perhaps this is the post:

http://www.fieldlines.com/story/2005/11/10/171555/99

I have a 5m diameter bladeset modeled also - I think it is using the S809 airfoil.

I have a couple of projects that will require new profiles and twists. I doubt I will get to them anytime soon.

Ron, I've opened the IGS file and initially it looks like GibbsCam sees the entire piece as one surface so I guess now it's time for me to learn how to seperate the surfaces. Your file will work well as a learning experiment. If I get it right I'll post a generic G-code file of the part. I see CNC as a good way to make proper molds for blade fabrication or running off complex profiles. There's not much we can't do at the shop with a 4th axis involved it's just a matter of how to hold it and how many operations are involved and what tooling to use and what speeds and feeds and... 5 axis would be nice.

Thank you,

Rich, I've followed your CNC experiments in the past and compliment you on your achievements. I know it's not easy and I realize the tremendous amount of code required to cut the blades you've shown us. Longhand programming a 3d surface with the stepovers you need is a big job and I can imagine but would be reluctant to do it. I envision realizing I maybe should have used a different sart point and then the sinking feeling would come. You could program incrementally but that's not that fun either.

I'd realy like to see a sample of your g-code so I can get a view of how your machine runs. Can it handle canned cycles, not that they would help for blade carving. Can you loop a sub program or even use a sub program. Some controllers will allow looping within the program without a sub call. This could be used to loop an incremental line that could impart a twist and taper although it would be constant. Will your controller allow for macro programming or is that how you allready progam? A macro could be used to scale  each stepover by a specific percentage and impart an incremental move as well this would allow for deminishing twist and taper. Don't think I could write that one though with out a bit of research.

Lot's of questions and maybe some suggestions.

Thanks,

CNC is fun.

I don't have any pictures at home of some of the parts we 've run at work but I grabbed this image from the National Oilwell site.

I've programmed and run the indicator head (housing the digital meter) and the couplings on Either side of the activator as well as the activating levers on either side of the lower coupling on the assembly to the left. The activating levers are .225 thick and 1.01 long. Fairly basic shapes but challenging never the less. We send them out uncoated and National has them plated to make 'em all shiny.