Homebrew 4th axis build underway

[mountaindew]

I was reading your thread while watching and testing g-code simulation with a 4th axis test part
New version of cam was giving me a hard time but persistence pays off. This g-code simulation feature is crazy useful with 4th axis tool paths.
Clip of simple 4th axis test part. Just a 1" dia pvc pipe.
The operation shown can generate a lot of code or a little code based on settings and surface selected and even how the model was drawn in cad!

I use stuff like this to experiment with the many settings until I come up with some good efficient usable code. Then use those setting on real parts in hopes of good tool paths.

[soofle616]

Sweet.
Now you can relax and unwind 4th axis inside joke

Anyway
Nice work. hog finish on motor mount is fine for tools imho

Hah! Clever.

I'm not overly concerned with the finish on the motor mount but I am still thinking of making a new one using my new capabilities anyway. One because it'll be shinier, two because it'll give me a chance to learn more about what my new setup is capable of.

What software are you using? I've been doing everything in the freebie version of fusion since I am just a hobbyist but someday I may step up to something more capable.

[mountaindew]

Hah! Clever.

I'm not overly concerned with the finish on the motor mount but I am still thinking of making a new one using my new capabilities anyway. One because it'll be shinier, two because it'll give me a chance to learn more about what my new setup is capable of.

What software are you using? I've been doing everything in the freebie version of fusion since I am just a hobbyist but someday I may step up to something more capable.

I started using sprutcam from days long before fusion. It can be a difficult program to learn. I get good results with somewhat ease mostly because of years of use. And the one thing that annoys me about 4th axis projects is the sometimes long rewind to zero if you dont use a homing sensor or other strategy too return to 0 more efficiently.

Anyway good reading and nice useful tool for more complex projects
search grabcad for interlocking puzzle. This is a simple 4th axis project you might enjoy.

[soofle616]

I started using sprutcam from days long before fusion. It can be a difficult program to learn. I get good results with somewhat ease mostly because of years of use. And the one thing that annoys me about 4th axis projects is the sometimes long rewind to zero if you dont use a homing sensor or other strategy too return to 0 more efficiently.

Anyway good reading and nice useful tool for more complex projects
search grabcad for interlocking puzzle. This is a simple 4th axis project you might enjoy.

The unwind thing definitely does annoy me. I haven't spent much time playing with my new toy since getting it going so I don't have any kind of "intelligent" solution yet but I do want to figure out a way to get the machine to reset itself to 0° rather than unwinding all the way from 360 or more.

[kstrauss]

Stop unwinding!
G92 A0 ( reset rotary)

[mountaindew]

Stop unwinding!
G92 A0 ( reset rotary)

Cant get my cam software to spit out that g-code.
Will do some testing.

[kstrauss]

You can always enter it via the MDI.

[TurboStep]

Stop unwinding!
G92 A0 ( reset rotary)

G92 will simply add an offset to the current position in order to make DRO read zero (or some other value). That might be fine if the previous op ended at a integral number of full rotations, if not you'll lose track of the zero reference position. The offsets are also lost after a G92.1.

To "unwind" the actual work offset and then return to zero (optional) you'd probably need something more like this:

#<new_A> = [#<_a> MOD 360]

o<unfold> if [#<new_A> GT 180]
  #<new_A> = [#<new_A> - 360]
o<unfold> endif 

G10 L20 P0 A[#<new_A>]

G0 A0 (optional return to zero)

Unfortunately I haven't used Sprut for quite a few years so can't help with manual code insertion.
Step

[kstrauss]

Yes, if you want to maintain the original setting you need to do something like you suggested. I mostly go full revolutions when making a part so resetting to zero works well for me.

[soofle616]

Update:

Finished building an installing a subplate for my a-axis yesterday to hopefully make it easy to drop in and run. Location is based on two countersunk bolts and two SMW diamond pins all locating on the fixture plate. Spent a lot of time dialing things in last night and overall I'm quite pleased with where I left it. 0.0003" over 2.5 inches in Y, tilt angle is good to .0004" over 6.75" along X and even the face of the cheap chinese chuck is flat to within about 0.0002-0.0004" based on a 360° spin. I do want to recheck my tilt angle though. I used a 1/2" aluminum rail as my test bar. Supposed to be straight to within .001" over 12" but since the end was unsupported, I'm curious if gravity and/or downward pressure from my haimer may have caused some deflection. I'm not sure that it matters that much as I doubt the deflection was dramatic and the greatest liklihood is that long pieces like my test bar will have a tailstock support at the end anyway so if the misalignment is slight, the tailstock will push it back in to place. The only thing I'm not thrilled with is the .003" TIR of the test bar measured about 1" from the chuck. For most of what I'll be doing that's not going to cause a problem but I would have liked to see it closer to .001. At some point I may look into a 4 jaw replacement but by the time I get that far I may have learned to accept that what I've built will never be a microarc and I'll just buy one of those. Next up is a matching subplate for the tailstock, to be put together over the long weekend, I hope. When all that is done I may have another look at my post processor to see if I can incorporate the code posted above to avoid unwinding.