So what causes this?

[araugh]

See the extra little bite taken out of the bottom left and top right sections of the center of the X pockets? What am I doing that causes this? It doesn't show up in the simulation, I'm running at 1ips at 14000RPM using a 5hp colombo spindle and an Osrund 65-026. In this case I compensated for it (mostly) by having the edges of the pocketing toolpath outside of the profiling toolpath, but in the future when I'm trying to do an interior pocket how do I avoid this? Here's the product after final profiling.




Also, what's the best way to get rid of milling marks on something like this, a vibratory tumbler or media blasting?

Thanks!

[steve_g]

At first glance... If I had to guess... I'd say positively...maybe... bit deflection... or perhaps cumulative machine stress changing at change of cutting direction. Just my thoughts.. I wonder if "onion skin" type cutting strategy... like we do with wood will help here. By that I'm suggesting a very light perimeter cut.

SG

[Thom Dudley]

I get the same problem in wood - rastered pocketing does a light perimeter cut, but I don't like the surface it creates. Is there an easy way to make it go back for such a cut, r does it require making multiple vectors - one for the pocket, a SLIGHTLY larger one for the cleanup?

[Brady Watson]

Keep in mind you are starting to push the limits of the machine with this project. It is a light weight CNC router - not a CNC milling machine. There is a reason milling machines are so heavy - you need the mass to overcome the forces that cause deflection.

I would say from experience, you are cutting too deep and too fast. Hog out with some allowance, maybe .015" then go back and clean up the cuts after they are roughed out with a finishing pass. From what I see, it looks like you did the roughing...now need to go back at a lower speed @ full depth to do a nice sweet cleanup pass on everything.

-B

[Steve M]

What Brady said, plus use climb cutting and coolant/lubricant for the finish pass.

[araugh]

Got it, thanks guys! I didn't think that a 1/4"x2" bit would deflect that much! Or is the the entire z axis that's deflecting from true? Good to learn!

[billp]

Another factor- the longer the bit, the longer the lever arm...If you are only planning to cut .25"on a regular basis, get bits whose over all cutting length is only .5". Shorter bits = less deflection, and get you more useful cutting life than longer bits where you never use part of the edge.