[s3r4ph]

Hey all, I've had a Joe's 2006 R2 machine for a bit, but want to mod it for 5 axes. I've mocked up a design that could mount onto the Z-axis plate.

I've used bearings from VXB, mounting hubs from Pololu.com, and some standard 180oz/in NEMA 23 motors.

This is my first attempt at Solidworks and a 5axis design, do you think this would work?

(I can't seem to upload any files, so here's a google drive link, just DL the file(s) and open)

EASM File
https://docs.google.com/open?id=0Bwa...kFlMnpvQ3dCMEE

Solidworks Files
https://docs.google.com/open?id=0Bwa...VFPM2FrTzE4OVU

-s3r4ph

[ger21]

Can you just post some images?

[s3r4ph]

Sure, here's some photos which may help understand the design.

[ger21]

What material do you plan on using to build it?

[s3r4ph]

I didn't really think too much about it, but had pretty much figured 3/4" HDPE would be the best, for price and because some holes need threaded.

[ger21]

I would think that it'll flex like crazy with those long thin sides.

[s3r4ph]

I thought about that, it's a trade off with weight and rigidity. I will probably decrease the hole size, or remove them completely. I dont really have a good feel for how heavy the sides would be (10" x 5" x 3/4")

[ger21]

Never trade rigidity for weight.

[joecnc2006]

you may find it might have to much flex in that camera gimble type design, you would need to Box it in to make it more rigid. This design is used allot in areal photography.

[s3r4ph]

Joe/Ger, thanks for the feedback. I've been messing around and refining the design a bit.

Boxing in the router might be a good idea, I wonder if I can put an arch-type front and back panel to enhance rigidity, but still allow the router to pass through those sides.

I've messed around with my design a bit today, I'll box in the router and post an update.

[joecnc2006]

your router mount need to be two plates spaced or a single plate having a minimum thickness of 2" to avoid the router from having angular movement in the moumt.

[s3r4ph]

Here's a modified design with an additional router mounting clamp, and some brackets on the front and back of the vertical axis.

[s3r4ph]

OK, I think I've gotten as far as I can get without building the part. I've made some adjustment to the edges, to give it better surfaces for gluing if necessary (for MDF), as well as keeping bolt holes (for HDPE).

It's currently designed for a Bosch Router, but if anyone wants to give it a try I am more than willing to provide all files and schematics, just ask and keep me in the loop on your build.

[joecnc2006]

Looks like you have direct drive on the motor, What do you plan for the step down for better resolution.

[s3r4ph]

Honestly, I havent thought that far ahead. With the current design, for better resolution a gearing ratio might be the easiest thing to implement, with a small gear attached to the motor shaft, and a larger gear to the axis of rotation.

Have a better idea? I'm not thrilled about setting up a belt/chain setup, but I've seen some designs which use it, and it might be better for CG moments.

[joecnc2006]

Use a planetary gear attachment to your stepper. Then you do not need to change the design.

[s3r4ph]

Good idea , think a 5:1 would be enough? I'm planning around a 1.8deg / step motor. 5:1 puts me at a 0.36deg/step or 1000steps per rotation.

I can also find a 10:1, both from Kelinginc.net, but for about $50 more.

[Walky]

Good idea , think a 5:1 would be enough? I'm planning around a 1.8deg / step motor. 5:1 puts me at a 0.36deg/step or 1000steps per rotation.

I can also find a 10:1, both from Kelinginc.net, but for about $50 more.

I have absolutely no experience with rotary axes, but if I were in your place, I might as well go with even more reduction if possible. Let's talk mm instead of degrees, and suppose the router rotation in either the 4th or 5th axis makes a 200mm diameter circle (projection of the tip of the bit), then such circle diameter is ~2 x 3.14 x 100 = 628mm. For a 10:1 ratio and a 200 steps/rev motor it would be 628/2000 = 0.314mm/step along the perimeter (let's not consider microstepping as it's not absolute). I assume 400 steps/rev motors would help, too. Now, I don't know how much do cutting strategies and combined movement of all axis compensate for a certain resolution, maybe it can be enough resolution considering a 5 axis machine with a ball nose bit can attack the material in different ways?. I guess this kind of compensation's effectiveness must be proportional to the radius of the ball nose bit.

[s3r4ph]

Nice math there, I haven't thought about it from that angle.

I may go with these, they're pretty inexpensive, and a .02deg step is about as good as I have been able to find while staying under $250/motor.

Phidgets Inc. - 3319_0 - 42BYGH40(M)-160-4A NEMA 17 Bipolar 99.51:1 Planetary Gearbox Stepper