Toolpaths for indexer

[denver]

I'm in the middle of setting up an indexer, so I'm also investigating methods of generating the necessary toolpaths. I've had good luck so far making cylindrical things using Vectric's Cut3D and carving two sides. That works as long as the object is properly symmetrical, but I've recently gotten some requests for things that can't easily be done that way. I've included an example of one below.

I've noticed that if I use similar settings for the indexer as I used for Cut3D, the resulting indexer files sometimes take a lot longer to cut than the two sided ones. For example, the two sided version for the finial takes a little over an hour for each side. Using code produced by DeskProto, it takes nearly 5 hours.

On the other hand, the bead and reel molding that I've made in the past takes about the same time either way.

Is there a different relationship between the bit diameter and the stepover when using an indexer? For relatively small or detailed objects, I generally use a 1/8" ball nose. If I want a very smooth finish, I use a stepover of .01". If it doesn't need to be real smooth I'll go up to .015". Seldom as high as .02".

Thanks,

Denver


3096

[denver]

I should add that the finial is about 14" tall. The bead & reel molding I've made in the past was about 1" in diameter and 24" long.

Thanks,

Denver

[myxpykalix]

It would seem to me that you could use a more liberal stepover because most of it is symetrical and you could simply sand out any rough marks by spinning it and using sandpaper or putting it on your lathe to do that (if you have one).
As far as your reeding goes, i have been trying to make a toolpath using a simple profile of the reeded part to use a side reeding bit follow the path, move back, rotate (X) degrees then do the next, without much luck.
This would not work with a reeding bit keeping your part as a whole like you show, but would have to make it in 2 pieces like i illustrate here.
To make it like you illustrate i would think you would need a tall roundover bit and even then i can't see how you would make this as one piece, but if you can i'd like to know how also.

3097

[myxpykalix]

also, If you need any spiral toolpaths to try on your indexer or help setting it up, let me know.

[cmagro]

Denver,

What's the diameter at the base?

[denver]

Thanks Jack. Your approach is interesting, but the radius of the flutes, or reeds, gets smaller as they get closer to the center. It would be difficult to duplicate that with a fixed radius roundover bit.

I'll keep the offer for spiral toolpaths in mind. I enjoyed reading about your adventures along those lines and seeing your photos.

Christian, the base is about 4" in diameter in this model. I don't have exact dimensions of the originals yet. I just made that model guessing from a long distance photograph. I'll be making these as part of a restoration project, if it gets approved.

Thanks,

Denver

[cmagro]

Thanks for the info Denver....I'm a little ahead of schedule with stuff and wouldn't mind taking some time to make a couple of finials. I'll post step by step if I do get a chance.

Christian

[denver]

Hi Christian, that sounds good. Do you want a copy of the model?


Thanks,

Denver

[cmagro]

Already modeled it but go ahead and send what you've got.

[cutitout]

how many reads are on the turning? can you program one or two reeds then rotate/index a division of deg to reeds. if 36 reeds index 10deg

[denver]

Hi Eric,

There are 24 of them, so that would be 15 degrees per. If they were straight across with a constant radius it would be easier. These not only follow the contour, they also get narrower as they approach the center. I suppose I could take measurements across that 15 degrees and along the length and manually create a file that would end up working. Sounds pretty time consuming, though.

Thanks,

Denver

[rnels]

Granted this is rough it is out of a semi-green birch log but the ideas are there. Did not want to spend a lot of time getting the flat areas on the model so just used ball nose bit. Would guess one could add more material to the model and use an endmill to flatten out certain areas. The ball at the top is more of an oval and would need to be played with more to get right. I used ArtCAM to produce the model but maybe the new Aspire has these features now too.
Here (http://www.nelsonsdesignssp.com/artcam_denver_hull_finial_creati.htm) is how I went about it.


3098

[myxpykalix]

Very cool Randy. How long did it take for this to cut? What is its size?

[cutitout]

Thanks for the link -- great to see how other shopbotters do what they do.

[bill_l]

Randy,

Thanks for posting the "how its done". I took the time to review it ... always wanting to learn new tricks. I was wondering however, why you did not post the tool path direct from ArtCam? There is a post processor for both X and Y parallel for the indexer. I must warn you that the one for the Y parallel in ArtCam V9 (to be used if your indexer is mounted at the end of your table) needs to have a line fixed to make it work properly.

We have had our indexer running about a week now and I have taken an instruction paper that was provided at one of the ArtCam user groups and refined it. I'll make it available if anyone is interested.

Attached is one of the test files we created and ran and it turned out absolutely perfect. I'll eventually get a real photo and post it.

Bill


3099

[rnels]

Thanks for the replies!

Jack, It is the same size as above 14 inches long approximately 4 inch diameter. Took about an hour and a half to two hours.

Bill, Back in the day there was no post processor so I made the program. Still using it today as it works for me for what I need. I believe I tried the X indexer post processor quite awhile ago and it did not give the same results as my program. Glad it works for you it makes it easier.

[jdervin]

Bill, in your 11-20-2008 post, you said "I have taken an instruction paper that was provided at one of the ArtCam user groups and refined it. I'll make it available if anyone is interested." Consider me interested.

[donclifton]

I would be interested also

[zeykr]

I'd like to read it also.
Thanks

[hh_woodworking]

Please include me also
Thanks
Ed Harrall

[denver]

Thanks Randy. Sorry it took so long to get back to this. I really appreciate the time you took on that, and the detailed instructions you provided.

And I also would like a copy of the instruction paper you mentioned, Bill.

Thanks again everyone, this has all been very helpful.

Denver

[khaos]

Bill, add me to the "Me Too" list. Is the document small enough to be added here as an attachment as a zip perhaps?

My humble thanks in advance,
Joe