Official Luthiers Forum!

So, in preparation for my impending CNC router purchase, I'm trying to draw things up in advance so that I can start carving things right away (purchase is still 3 to 6 months away). One of the things that would be very helpful for me would be to get necks with scrolls quite close to final. I'm not very good at carving so it usually takes me weeks to finish the scrolls and necks - most of which is procrastination. Yes, I do get some pleasure from carving a neck but not nearly the kind of pleasure I get stringing up a finished instrument! I am definitely a beginner at advanced surfacing with CAD but I do have experience with run of the mill rectilinear type geometries.

The scrolls I use on my banduras are kind of like a violin scroll but perhaps a little simpler. I also don't have two flutes like a violin has because that just looks a little too classical for me (though I've done it before). First couple pics are some scrolls I'm carving for some instruments for my kids. I've then included a side and back shot for reference.

Anyhow, at my disposal I have Rhino 3D (for OS-X which is free at the moment) and SolidWorks 99 that I have from an old job. I believe that Rhino is almost full featured but is missing certain things like full import/export like the windows version has. I've also found this mini tutorial for drawing a violin in Rhino: http://www.renderosity.com/mod/forumpro ... _id=600161 I've tried that technique in both SW and Rhino and can't get it to work so far.

Solidworks troubles:

First off, it looks like a helix can't be scaled. A scroll isn't circular but some other shape.
Second, It seems like SW99 won't allow the surface to self intersect during the loft and also won't allow zero thickness
Next I tried to do it like I carve - start out with a cylinder and use a cut loft with guide curves and got the last picture - not what I expected.

With Rhino:

Tried to do the tute and ended up with a scroll that seems to warp through itself or something (see pic). I tried adjusting settings for the sweep with two guide curves but most of them were grayed out.

Anyone here willing to provide some guidance or mentoring on how to do this. Perhaps via skype or something?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Andy,
That is a good one. I am fairly good with SW and tried a few things and was close but it wouldn't accept the inputs. I will try a few things and get back to you.

_________________
Brad
Avon, OH

Andy, I suppose one way of doing this would be to have one of your existing scrolls 3D digitized and create the 3D model from the point cloud. It would be nice if scrolled tool paths could be generated in CAM but I suspect that it's going to have to be conventional parallel tool paths. I'm not sure but looks like a very small ball nose cutter would/could be used. Then again, maybe this job is going to require a 4-axis setup. Don't know. Will be interesting what some of the guys have to say on this.
Nelson

I'm not planning on buying a 4 axis machine (though I might get a 4 axis controller if it's an option) but I fully understand that this will require machining on all 4 sides.

I've thought about the digitizing thing except that a) I'm not satisfied with how well I carve my scrolls today and b) I'd like to learn how to do it in cad so that I can make artistic modifications as things get boring.

Visual Mill is one of my front runners and once I get a scroll drawn up, I can try out some tool paths and see what it's got.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Andy,
Here is what I as able to come up with using SW2008. I do not know the differences between SW1999 and SW2008 but since you were using a helix/spiral command I think you may be able to come up with something similar. This picture is just a sample but I think it is conceptually a scroll.

Let me know how this looks and if it is good I can post the approach.

_________________
Brad
Avon, OH

Definitely a step in the right direction and much better than what I've come up with!

I did use the helix feature for a test but really scrolls use something in between a helix and a golden ratio curve (though not a true golden ratio). I think though that if I can get it working with helixes, it's just a mater of defining some 3D -helix-golden-esque curves to do the same thing.

I'd love to know how you did it!!!!

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Andy,
Here is how I created the scroll.

Step1: create a sketch on the front plane. The rectangle will be coincident to the origin. The height is .375 and the length is .500.


Step2: Create a circle on the top plane. The center will be coincident with the origin and the diameter will be coincident with the outside corner of the rectangle from step1. This will create a circle with a 1.00 diameter as shown.


Step3: Create a helix/spiral from the circle created in step2. For the pitch I used .375 which is the height of the rectangle. Select the Helix taper option. Under this option I selected taper outward and used 53 degress for my taper. This number was from trial and error but assuming the start of your scroll is 1.00 minor diameter (taken from the circle diameter in step2) and the scroll has an major diameter of 2.00 it can be calculated. To calculate use this formula. 90deg - arctan (pitch/((major dia. - minor dia.) / 2). Using the formula the caluclated taper would be 53.13.


Step4: Create a sweep using the sketch from step 1 as the shape and the helix as the path.


Step5: I then created a 1.000 diameter sketch on the top plane and extruded the diameter to create the inner diameter of the scroll.



Good luck and please post your results when you get it figured out.

_________________
Brad
Avon, OH

Outstanding! This I can work with. I've attached a pic doing exactly as you wrote. For some reason though, the 1" diameter circle in the center isn't displaying correctly - I tried both a boss-extrude of a circle and revolving a rectangle like you did. both display the ends but not the guts in the center.

Probably a bug in something or other as I'm running this version of SW in a virtual machine. At least I hope so - I just loaded it in eDrawings running natively in OS-X and it still displays incorrectly.

Thanks so much!!! I think this will work with any 3D curve. It seems that sweep works differently than loft in this situation in that it's much more forgiving.

I'll be out for a week but when I return I'll try and get it cooking!

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Andy,
Good work. The display issue that you have is something that I also had happen when I first started messing with this. When I was putting the tutorial together I tried a bunch of different settings and haven't had the problem since. My guess is that you have a condition when you create the inner diameter extrusion or revolve that is creating a geometry error that screws up the display. Play around with it a little and you may find you can get it to go away.

Good luck!

_________________
Brad
Avon, OH

hehe....when it comes to complex shapes with cad...

...tenacity is capacity!

Nice work, guys!

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I used the mirror all instead of mirror feature and it worked great. Getting closer!

I'm still having display problems though. I exported the scroll as an IGES file and it imported into Alibre just fine. Hmmmm

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

As I'm refining the shape, I'm having more rendering issues. The center cylinder I originally did by extruding a circle. That stopped working giving me a failure to merge surfaces or something and I had to replace it with a revolved rectangle.

Again after importing to Alibre, everything looked ok. Any ideas?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I don't get it. For giggles, instead of filling in the center part of the scroll in the actual part, I made an assembly and mated the exact same section in the assembly and it works perfectly. I still have to do a little refinement at the interface of the two but even this would probably look pretty darn good in a real piece of wood.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I think how you pushed through to get that result is commendable, Andy. That result looks great! Looks like you pretty much controlled every aspect that you wanted.

One question now is how much of that tight definition will you lose when it's cut? You'll have to hand work it with a blade to get that crisp detail back in the crevices and inside edges after milling it. Another option would be to program a pass using an engaving bit...but your cam software has to agree with your vision on this.

The main thing I'm wondering about are the negative drafts and how this might be produced. The way it's drawn, with the center post tapering inward, it can't be produced on a three axis machine. Maybe a four-axis could do it but that's a fair programming challenge...assuming your software can do it.

Making that centering post a simple extrusion inward without the taper would enable this to be made on a three axis machine but there would be up to five setups to do it.

1. The right scroll side
2. The left scroll side
3. The top half
4. The bottom half
5. The trickiest - an angular attack at the inside corners where the scroll meets the main body of the headstock. You might be able to skip this fifth op and get it on the side ops but getting a tiny tool way down in there might not be possible from the side.

Note that there are two options to doing this. You can try to redraw it so it looks right ....or, approaching the scroll from the side (op 1 and 2) you can just cut the geometry as drawn with a surfacing operation and the centering post will be cut without the taper. Mastercam would ignore the inward tapering of the post and just roll the tool off the small circular top surface straight down to the main surface.

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

Thanks Stuart,

For giggles I've run some simulations running VisualMill (probably what I'll end up buying) and as you noted, the negative drafts are lost but, no damage is being done. I'm not all that worried about it as those can be done with hand tools pretty quickly.

My current thoughts are to have a 4 sided setup with probably around 3 or 4 tools with each setup. Doing that should get a lot of the negative drafting completed with just the inside corners being missed. I supposed for giggles I could come up with an 8 sided setup as you suggest with the additional 4 set ups being at 45 degrees to the first 4 and that should be able to get in there and do the whole shmeal. hmm....maybe a 4 axis setup might be a good idea after all!

The more I'm thinking about it, if I separate out the shaping of the neck/scroll in to two sections - one for the headstock and one for the rest, I could probably do 8 angles for the scroll and 4 for the majority of the neck even having some duplication there too. If I use square stock I could make a cradle that would hold the workpiece in either the "flat" orientation or the 45 degree orientation without swapping stuff out. I could then turn 45 degrees and machine each side of the scroll with progressively smaller tools. That would save a lot of tool changes.

It would still require quite a bit of baby sitting to rotate after each operation to rotate but at least I wouldn't be changing tools so often. Once the scroll is done, the rest of the neck could be done with one or two tools.

Must.......resist........4th.....axis......SWMBO will kill.......or will she?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Great work Andy! Looks like you have come a long way on your scroll!

_________________
Brad
Avon, OH

Thanks Brad. Couldn't have done it without your help. I think some of those blends near the base of the headstock are going to drive me mad though.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

The progress continues but man does my brain hurt. I'm kind of at a loss as to how I'm going to blend the bottom part of the back of the scroll - I'd be happy for any suggestions.

The mate between the scroll and the rest of the head stock isn't perfect but I think it's probably good enough. i.e. nothing a few seconds with sandpaper can't fix.

What do you think?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I thought for a bit that I should post this in the "what are you building thread" but anyway made a bunch of progress today. Two little fills left and then some tweaking. I can do the fills using lofts but I don't like the transitions/blends from curves to straights - something much easier to do by hand in wood.

I might need to think on this a bit or <gasp> RTFM or something.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I haven't been around for a couple of days and this shows up. Really amazing (at least to me) work you've done here Andy. Also a glimmer into why the 4th axis option might be added to a CNC purchase for luthier purposes.

Thanks for documenting the design process here.

_________________
Allen R. McFarlen
Barron River Guitars & Ukuleles
Facebook
Cairns, Australia

Thanks Allen.

A 4th axis would make this easier to machine but then you've got a new problem with holding the darn thing. That scroll is on the end of a neck that's about 16" long. I would think you'd need some sort of tail stock to keep it from flexing that far away from the drive end of the axis. Tailstock would make machining the top of the scroll very difficult.

The other thought I had if I ever do add a 4th axis would be to make some sort of giant bearing type of thing that the blank would fit in that would actually be holding the piece while machining so that the 4th axis is just rotating the blank. Once the scroll is done, I'd remove it from that fixture for a two sided setup to finish the remaining part of the front and back.

For now though, I don't want to mess with a 4th axis - 3 is already making my head spin. I'll be doing this with a 4 sided setup and clean it up with a gouge and chisels where I have to. If that's not good enough, I'll go to an 8 sided setup. The big plus of the 4th would be that I wouldn't have to be baby sitting the pice to rotate after each operation.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

The Blurry Customs guys have a trunnion table that mounts in their 4th axis that I was wondering about, and now you've presented one possible use for it.

_________________
Allen R. McFarlen
Barron River Guitars & Ukuleles
Facebook
Cairns, Australia