Eyeball Engineering

[rob_b]

Very nice work. Using a single tool?

[Duuuuuuuude]

Yup. Double flute 1/16" endmill.

[danielk]

I love the asymmetrical design a few posts back

[Duuuuuuuude]

Originally Posted by danielk

I love the asymmetrical design a few posts back

You mean the one where I randomly threw some lines down and sent it to the router? Me too.

[Duuuuuuuude]

Yesterdays results:

[rob_b]

Awesome

[Duuuuuuuude]

They're getting better. It took some time to work out how to do spokes, but now its fairly easy. You just have to make sure the gaps aren't too small to cut.

A thing happened last night with the last three wheels...instead of the machine making nice smooth arcs along the curved lines, it was making small, jittery steps. I don't know yet if it was an equipment issue or a coding issue.

I'm also experimenting with the kinds of cuts I make. Pocketing the holes makes them cleaner, but takes forever. Doing a profile cut is not as clean, but much faster, and there is a risk of the waste material bouncing back into the bit and causing havok. I had a chunk wedge itself in and knocked the wheel up out of the fixture a bit. I think it may have also caused the machine to lose a few steps. I'll know more when I zero it in again, but I'm pretty sure it did.

[rob_b]

Make sure it didn't slip into "exact stop" mode. Mine does that randomly. It's great for drill cycles but to want "constant velocity" for any profile milling as it smooths out the motion. There is also a setting that controls the allowable position error for constant velocity mode so you can tweak the smoothness as required.

[Duuuuuuuude]

Now that you said that, I'm pretty sure it started after running the bolt pattern wizard for the clamps. Thanks for the heads up.

[Szczerba]

Digging everyone of those designs.

[danielk]

You should do a Tartan design. I'm curious how well plaid would turn out as a wheel.

[XSRCdesign]

Man.... That's done on a router?

Sweeeeeeeet

[Duuuuuuuude]

Blanks cut on the lathe, design cut on the router, then back to the lathe for some profiling. Got the first production prototype set done tonight and they look pretty bitchin'. Exciting times...I've been wanting to do this since forever...

[Duuuuuuuude]

[XSRCdesign]

That is so dope!

[Duuuuuuuude]

Couple more done.






I might have to have a prototype sale to fund the R&D work.

[Duuuuuuuude]

So my wheel fixture wasn't working out so well...kind of a pita to use, couldn't lock it down or even instal the wheel blank if I didn't cut the OD dead nuts. I liked rob_b's super simple fixture, so I dug around for material and made something similar. 1.25" rod bolted to the table in some existing holes, and an eyeball engineered cam lock to hold the blanks in place.



I also went ahead and surfaced as much of the table as the machine would reach, so now I've got a perfectly flat 10x12 work area. Doing so helped me sort out some more vibration and chatter, which was coming from the table itself. It was resonating enough that it would bounce the cutter around quite noticeably. I managed to knock quite a bit of it down by placing a chunk of rubber and the heaviest aluminum block I had on the table outside of the work area, but I'm thinking I may ditch the top entirely and replace it with a solid 1/2" aluminum plate.

[Duuuuuuuude]

I'm in need of the RCC machinist brain-trust.

Last Monday I took a break from making wheels and did a short run of aluminum parts. When I was done, I got set up to do wheels again and ran into a bit of a problem. It's cutting everything egg shaped instead of round.

I went through everything and made sure it was tight and as it should be. A few things needed attention, and the cuts were noticeably better once I took care of them, but the issue remains.

The best way I can explain it would be to use the 12 beadlock bolt holes I drill as an example. They are all within .001" - .002" of each other except for three or four of them, so it's a fairly abrupt shift to the worst of the bunch, which is a full .010" out of line. I've zero'd the work a dozen times.

I've triple checked backlash on both axis, and a .0005" movement in mach equated to about .0004" or so on my indicator in either direction. When I physically push the gantry around, even though it's a bit flexy, it always comes back to near 0.

It's not a change in programming, any of the bolt hole programs I do, new and old, will net the same results. Along with the bolt holes, the programs also contain code to cut a .020" deep channel that I use as a guide when cutting the bead grooves on the lathe. They too are slightly egg shaped. Same with the wheel designs and the drill cycle for the hubs.

What am I missing? What should I be looking at? Tomorrow I'll run the cycle again and take note of exactly where and when the problem happens...

[eric113]

You have got too fancy and need to get back basic's.

[Duuuuuuuude]

Originally Posted by eric113

You have got too fancy and need to get back basic's.

Growing pains. It'll pass.