Sprutcam and 4th axis

[justgary]

saabaero -

I put on my data mining helmet, clicked on the headlamp, and went prospecting out on the vast internet. I found just enough nuggets of information to piece together the motivation to keep trying. Sometimes all I had to go on was the trust that it could be done. Lucky for me the batteries in the headlamp didn't die while I was still mining!

SprutCAM made it pretty easy to tell when I had done something wrong, because I got it to lock up or instantly crash several times. I have learned over the years that once a crash is repeatable, you probably should try something else.

If you are talking about synchronized A axis as in helical milling, I have not tried that. It would not be too hard to code a helix by hand, though, since all of the axes run at constant rates, even on a tapered one. Sometimes you just have to take the bull by the horns and hand code to get what you want.

I'm confident that SprutCAM will eventually do it easily and correctly, but for now I'm glad it does at least some of it. Tool-end milling, as the program calls it, is just that: they simply rub the end of the tool all over the outside (or inside, if you skip a step) of your surface while they incrementally turn the A axis. Automatic toolpath generation is quite a trick in three axes, and I imagine that four or five axes just makes it exponentially harder to do. And then you'll probably want it somewhat optimized...

Right now I'm working on a 4th axis toolpath for a twelve-sided pyramid that tapers to an eccentric cylinder. Rather than try to explain it, I'll attach a picture. So far I'm not satisfied with the toolpath or the run time. I'm redrawing my part now that I know how SprutCAM works, and I'll draw what it needs to see so that it will cut what I want. It's not too hard to adjust lengths and slopes to account for the cutter since SprutCAM won't do it yet.

As for your last question, go get the latest postprocessor file. If you have the Tormach-only version, it is this file, or if you have the open version it is this one. As mentioned several times before, the new postprocessor file has the MULTIGOTO command capability, which SprutCAM uses to drive all four axes at the same time.

Regards,

- Just Gary

[saabaero]

justgary,

You certainly did some fine detective (mining) work in developing that procedure.

I thought that I had the latest post processor but I'll have to double check.

What I had in mind was 4 axis synchronized motion as in doing a turbine impeller blade or something that required varying "Z" height while the 4th axis was rotating.

In looking at the recent model you posted I'm thinking that may need synchronized "Z" and "A" axis motion!?

[justgary]

saabaero -

I have played with the part that I posted a picture of a few days ago. I now have a toolpath that I am satisfied with, although I have not yet tried to cut it. The best part is that I'm getting much more comfortable with the small details in SprutCAM, and I can get what I want a whole lot quicker than before.

The bottom line is that you have to play with the settings in SprutCAM, but it will definitely give you a coordinated X-Y-Z-A path. In simulating the part I have, I can clearly see all four axes working together in a coordinated fashion to put the flats where they belong. It's actually impressive to watch.

I did have a glitch in one path that I had to fix by hand, and could not get SprutCAM to correct it. I fixed it on the simulator page by editing the offending lines before posting the G-Code. You can do that as long as you don't rerun the toolpath (which would clobber the fix). I think I'll also remove a lot of the air cutting that I see before I cut the actual part. Right now the job is looking like it will take 100 minutes to cut, but I expect I can cut that in half or better by removing needless motion.

I think that as long as your tool will reach all parts of your turbine blade without crashing on another part of the blade, you should be able to get a useful toolpath out of SprutCAM. It might take a while to cut, but it should work.

If you have an IGES file of your blade, send it to me and I'll take a look to see what I would recommend. Also tell me what types of tools you have in case I get a toolpath for part of it. I might as well use the tools that you already have...

Regards,

- Just Gary

[saabaero]

I'd be really interested in seeing the simulation for your part. You can email the file to my yahoo address that you sent the other file to if you have time and don't mind.

I think you have the right idea in attacking such complicated problems in that you have so many more opportunities to learn all sorts of other things in the process and afterward the difficulty of other projects pale in comparison.

I was trying to think of an example of a part that would require 4 axis synchronized motion when I suggested the turbine example.I don't actually have an application for one at the moment (but you never know). The only problem is that in order to articulate the cutter between the fins or the turbine wheel I suspect that 5 axes may be necessary.

Actually, one very real application I do have for true 4 axis motion is in fluting rifle barrels. I have done straight flutes on tapered barrels but would like to do spiral flutes on tapered barrels.

I am in the process of installing and wiring the additional stepper drive and have a two different size rotary tables that I need to fabricate mounting brackets for before I can actually have 4 functional axes. However in the meantime, I really want to understand Sprutcam's abilities and limitations as much as I can.

I really appreciate you sharing your experiences with the group. With so little documentation available for the Sprutcam it really helps if we can learn from one another.

[justgary]

I'll see what I can do about creating a video of the simulation.

I took about ten minutes yesterday and drew the attached "turbine" just to see what SprutCAM would do. Please note that I just threw it together, and it is not supposed to represent any working turbine design.

I then took another ten minutes in SprutCAM to see what it would do, and in the immortal words of Scotty, "what came through, didn't live long." Maybe I'll give it another stab sometime just for the experience.

To flute a tapered rifle barrel, you only need to move X, Z, and A. Also, they all move at constant rates. In other words, you don't even need a program to do it. Suppose your flute starts at X0 Y0 Z0 A0 (and for this discussion, the zero Z value is already plunged into the work). Now suppose that the flute should curl 15 degrees and run 12" in X and -0.5" in Z to account for the taper. Further suppose that you had already set your speed and feed to your favorite numbers and started the spindle. The MDI command G1 X12 Z-0.5 A15 will do it.

Of course, you'll probably want to do it in Z steps, but if you write them on paper (or even in a text editor so you make a program that you just load and run), you can easily keep track of them. Rotate A to the new starting point and zero A, then run it again for the next flute.

Regards,

- Just Gary

[saabaero]

Hey, you may have just come with a new design for an advanced jet engine turbine wheel!

As far as the rifle barrel fluting is concerned... Wow, I never even thought of that. I always amaze myself at how complicated I can make the simplest problem. Your solution certainly saves a lot of steps in having to create a CAD model, import it into Sprutcam and generate the tool paths.

I always say that anyone can come up with a complicated solution to a simple problem but it takes a genius to come up with a simple solution to a complicated problem.

[justgary]

saabaero -

I over complicate things all the time. It's probably a function of having cool toys to use. Of course, a year later it dawns on you that you could have just tightened the screw instead of making a thicker replacement part. Oh, well, it's fun to think up solutions to problems.

Post a picture of your fluting job when you get done!

- Just Gary