[BigAl1]

Hello All,
First post here, other than in the photos section. I would like some feedback regarding the images of my 5 axis milling machine design uploaded yesterday. Hoping for thoughts about now to build this, kinematics, and strength. I hope to cut aluminum freeform shapes with a KaVo spindle and 1/4" cutter. Any and all ideas are welcome.
TIA,
Allan

[derekj308]

Hi Allan

I think it is an excellent 'first cut'. I too have thought about a horizontal Z-axis 5a machine but mine was based on A and C being on Z and the spindle moved on X and Y. I like your design and will give it more thought and when I have suggestions I will post them back to this thread. First idea is to counterbalance the z-axis assembly to reduce the need for higher torque on the y-axis motor. I have a question, is the additional worm gear driven by the bevel gears on the c-axis a way of tuning out backlash on the primary worm? If it is can you post a picture with a close up of that mechanism.

Cheers

[BigAl1]

Pic added to gallery of the double worm drive. Note - this is not my original idea, I believe it was somebody in Japan that came up with several versions of zero-backlash motion systems and I copied one of them. Don't know who to credit for this as the link is dead now. The worm gears have a preload wave spring washer on one end in my version, but I seem to remember a torsion spring somewhere on the original idea - need to think about this more.
Thanks for the comments - I'll look at some kind of support or balance for the Y axis travel. Any thoughts about structural strength? The main baseplate is 1" A50 steel with 2x4x0.25 tube welded underneath, and the column is 2 of those tubes with 3/4" plate face for mounting the rails. C axis table is 1" thick and the support structure is all 3/4" welded plate.

[derekj308]

Hi Allan
Thanks for posting the pic, I'll look into that design for future reference when I get around to building a 5a machine. I think your design is structurally OK and possibly expensive unless you have access to offcuts of your plate and RHS. The area I would focus on is making the B and C tables (for those of you who wonder where the A,B and C convention comes from; A rotates about X, B about Y and C about Z, Z is the tool axis, use the 'right hand rule' to define X and Y positive directions) and associated components as light as possible to minimise the pendulum effect during rapid accel and decel of the x-axis. I would also try to move the x-axis screw to be centralised between the x-axis guides. That would involve a redesign of the position of your drive for the B-axis but would enabling a more balanced system under heavy accel/decel. Ideally you would want the screw postioned under the COG of x-axis assembly (incl. B and C assemblies) and the two guides spread equidistant from that centreline. The way it is at present will work but you will notice performance issues at high rapid speeds. If you are planning on going slow (under 1000mm/min) then your screw position wont be an issue if you have good quality guides. Again I think its a great concept and I hope you get to build it and share the final result with CNCZone.

Cheers

Derek

[FPV_GTp]

Hello All,
First post here, other than in the photos section. I would like some feedback regarding the images of my 5 axis milling machine design uploaded yesterday. Hoping for thoughts about now to build this, kinematics, and strength. I hope to cut aluminum freeform shapes with a KaVo spindle and 1/4" cutter. Any and all ideas are welcome.
TIA,
Allan

Hi

can we have a link to the images in the photo sections , have no idea where they are

cheers


"edit mode , sorry guys found it " cheers

[Splint]

And for those who've yet to find it here it is.

http://www.cnczone.com/gallery/showg...ser/36464/sl/b

[BigAl1]

Thanks for the feedback about the screw not being centered between the rails for my X axis. I don't plan to go fast, and I want the weight and strength for rigidity. At this time I can't see a different position for the X ballscrew, so I'll continue with this for now.
Any other ideas out there?