coding for roller cams

John,

I'm pretty sure sheetcam can generate the correct profile. Do you have a dxf of the cam profile that you can post? If so I'll check it out and post the result so you can check it out.

John

Congrats on your 72th... I hope I make it that far.

I'm at work and can view your cam and see what you mean about the cam but can't run it through sheetcam from here. I can see where a table driven cam would be much better. I know I can to that in Solidworks but that is high dollar software.

John

grey1beard wrote:

I'm running diy 3 axis using ubuntu 8.04/EMC2, and I want to machine some nylon roller cams.
Does anyone know of an open source ( ie free !) piece of software that will help in simplifying the process of going from a specification to g-code ?

The mathematics of roller-lift to cam profile are quite complex, as the roller contact point "leads" or "lags" the line between the centres.
Things get even more complex if the roller moves through an arc.

If I was trying to design such a cam, I think my approach would be to use Excel (or Open Office) and have that produce the G-code directly.

This approach actually works rather well for mathematical shapes like cams.

Alternatively, if you choose a cutter diameter to exactly match the roller diameter, you can "solve the mathematics in hardware". With a rotary acis it then becomes a very simple hand-coding excercise. If you don't have a rotary axis you can convert in Excel/OO again.

(I even think that there is a demo kinematics module that emulates a rotary axis in a cartesian machine)

Andy, did you take a look at the dxf file?

John, here is what I got from Sheetcam... I could not get an outside arc entry with cut on line (you would have to use cutter offset). There is some extra code in there for my plasma you can just delete. Any how you can see the output...



John

Hi John,

If you get a chance post a photo of the cam, it sounds real interesting to see the mechanics of it.

John

Sure.

John