Official Luthiers Forum!

Hey guys, I put a quick neck surface modeling tutorial in the CNC section. A few guys in one of the other threads were curious and I thought a few others may be as well.

Rhino Neck Modeling

_________________
http://www.PeakeGuitars.com

Thanks for doing that.
So you don't worry about surfacing a complete heel, I suppose that makes sense as you can just fair in this with some sand paper.
What's your take on Rhinos ability to heel with a wine glass/high heel profile at the body joint.
Thanks once again.

_________________
Jim Watts
http://jameswattsguitars.com

When I am designing a neck with Rhino I do a more complete model than just the carve surface..but I wanted to focus the tutorial on how to model the carve surface itself. I could model in the rest of the surfaces...but if you can model that carve surface then the rest of it is a piece of cake.

I have to admit that I'm not familiar with the wine glass shape that you're referring to, but I'm pretty confident that you would be able to model anything you may need to do with Rhino.

The tricky part of CNC is figuring out how to hold the part, and what steps to break the process into. Once you get that figured out (and make a few jigs) you can pretty much machine whatever shapes you need to.

_________________
http://www.PeakeGuitars.com

I'll second that one. After your CAD skills develop to an intermediate level, all your real challenges will be in processes and fixturing.

If something exists in three dimensions, then Rhino can accurately model it. The hard part is usually breaking an object down into the information you really need, like the profile curves needed to model a neck.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Jim, Rhino isn't the problem with that kind of contour, though it may not be the easiest thing to draw. The issue is whether you have to deal with undercuts.   Using a "lollypop" cutter allows for that kind of sweeping curve that some Spanish guitars show in the side view of the heel. These cutters have to be custom made.

We've recently gotten some amazing 3/4" three flute cutters from Onsrud.   One is a wavy fluted roughing cutter, and it cuts mahogany like there's nothing there. We cut roughing passes with it and then switch (automatic tool change) to a 3/4" finishing cutter that takes off the final 1/32" or so and leaves an incredibly smooth cut that's ready for 150 grit sanding. No chip outs, no blow outs, even in Western red cedar which has been difficult to cut clean on the pin router.

I was wondering if you were talking about spanish style necks..

I would carve them on their side. Carve the first side, and then flip it on the jig and carve the other. The low-tech fixturing solution would consist of locating pins and movable clamps...a more high tech (and expensive) would consist of a vacuum jig.

I use double stick tape and wooden dowels for a lot of my fixturing..but I wouldn't trust it to hold a neck in a cantilevered position like that.

This hits on a design for manufacture issue...I'll probably never make a neck like this because it poses a complication for the tools I have.

_________________
http://www.PeakeGuitars.com

Does anyone use an indexing fixture to hold the neck? Looks like tilting the neck 90 degrees each way would facilitate machining of the heel.
I do use an indexing vacumn fixture to CNC the inlay pockets in the fretboard tilting it a couple degrees each way just to let the split block inlays conform to the fretboard radius a little.

I can see where that could be useful, cool idea. That way you could potentially use thinner shell as well...

_________________
http://www.PeakeGuitars.com

I've been playing with necks some in VX. Here's a couple of screen shots of where I'm at now with it.





Getting the surfaces to follow the curves and be tangent everywhere has been somewhat of a challenge.

Nelson and Rick, I have been planning on cutting the neck on it's side as Nelson is suggesting. I'll locate off of the truss rod slot and a locating hole on the neck blank using vacuum clamping. After the first side is complete flip the neck and mirror the program. This will allow me to get the shape I desire in the heel area. Now, if I can only get the surfaces just the way I want them.

A lolly pop cutter would of course work but it seems to me it'd have to be pretty long, maybe not as bad as I think. Are those available with out going to custom cutter?

It maybe easier as Parser shows in his tutorial and just get the major surface and touch it up later.
Then again it really doesn't take all that long to hand shape a neck does it.

Any thoughts are welcome.

_________________
Jim Watts
http://jameswattsguitars.com

The fewer times you have to move a part, the better.

We have a 10 hp vacuum pump with our new machine, and it works into six vacuum zones on the table. We just drilled the table for 30 drill bushings which will allow pinning holding fixtures in various sizes to the table with known X, Y, and Z coordinates.   

We're very much in the early stages of all this, and there are a couple of guys helping with the programming and all, Matthew Tolley, who works here full time, and Richard Aras, a pal/consultant/partner in some misc. ventures.

We're about a month away from starting to make necks on the machine. We started with simple 2 1/2 D parts and then Matthew tooled up to make our Model 1 guitar bodies. They are the best we've ever made by far...

Rick--I totally agree with your comment on moving the part. Seems like indexing fixtures have a built-in nemesis and that is the spindle has be zeroed precisely on centerline of index. I recall back in the late 60's running a Devlieg Jigmill which was a precise horizontal mill with an indexing horizontal table. Some of the parts we were machining had opposing bearing bores that had to be concentric within .001" or so. Any error with the spindle not centered to the table index doubles when rotating the table.
Interesting discussion, guys. Jim, looks like you're nearly there on the solid model.

Just a couple comments with regard to vacuum fixturing...how are you going to run vacuum to a slotted neck?

Also, it is very possible to use a "vacuum generator" (google this for more info) as a MUCH cheaper source of vacuum than your typical high powered CNC router vacuum system which are made to suck down a whole sheet of plywood. The vacuum generators run off of a normal sized compressor.

I'm not familiar with VX, but in SolidWorks, Rhino, and all the other packages I've used getting tangent/curvature continuous surfaces is dependant on setting up tangent/curvature continuous guide curves.

That model that you have in your screenshots is very similar to the level that I complete mine to with Rhino.
Looks great!!

_________________
http://www.PeakeGuitars.com

[QUOTE=Parser] Just a couple comments with regard to vacuum fixturing...how are you going to run vacuum to a slotted neck?
[/QUOTE]

I was thinking I could use two vacuum pockets, one on each side of the slot.

I'd also have both sides on the same fixture so I would only have to pick up one set point. I agree moving something too many times is asking for trouble.

_________________
Jim Watts
http://jameswattsguitars.com

Just plug each end of the slot with some rubber gasket material. Also use the slot to align with pins. No big deal.   Actually, with 10 hp of vacuum sucking power, we have a big advantage in CFM. We can actually "afford" some leaks and still hold parts quite securely. 10 hp isn't considered large by major sheet goods manufacturing standards where they don't even make specific and gasketed holding fixtures, they just suck the plywood or whatever down through the pores in a sacrificial spoil board made out of low density particle board. When you've got 20 to 30 hp worth of vacuum pump, the router bit kerfs going through to the vacuum chuck (the entire surface of the particle board) doesn't matter.

Yeah, with 10 hp some leakage is ok .
I was planning on using the slot as one of my location features along with a pin to locate it along the slot.
It's you basic/classic 3 points of location.
I hadn't thought of just stuffing some gasket material in there but it would probably work.
Thanks for the suggestion Rick.

BTW, Rick if you setting up large vacuum chucks (spoil boards) there's a really good phenolic material made specifically for this known as Richlite. Google ought to turn it for you if you aren't already aware of it.

_________________
Jim Watts
http://jameswattsguitars.com

Man, you guys with 10 HP of pucker power are really spoiled!   

With the smaller vacuum generators you have to be a lot more meticulous about vacuum leaks. As you mentioned...not enough CFM's to keep enough vacuum pressure.

_________________
http://www.PeakeGuitars.com

I only have a 3cfm pump so I have to do a pretty good job of sealing.

_________________
Jim Watts
http://jameswattsguitars.com

To see what can be done with vacuum, just look at the setup at Taylor. They carve the backs of their necks after they're fretted. They also have the tables setup so they can remove fixtures and replace them at the exact same positions quickly. It can be a bit difficult to design the fixtures at first, but once you 'get it' it becomes second nature and they save you a lot of time. I'm not sure if they do anymore, but they used to just use venturi vacuum pumps for their hold down on the machines, as well. They have some sort of centralized system now.

I plan on using three fixtures for all the machining on a neck, and no hand work will be needed on them besides sanding when they come off. It could be done with two fixtures, but my total process time was shorter with the three.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Our Techno Isel machine has an effective working envelope of 49" x 98" with six long vacuum zones.   We drilled the table with a matrix of 30 holes and put in drill bushings that accept hardened jig locator pins that then match up to jig pin sockets we install in the bottoms of the individual vacuum fixtures.   The location of the pins is all in software, and so we can put a number of dedicated vacuum fixtures of predetermined sizes onto the table, and the machine knows exactly where the fixtures are located.   It only takes two pins to accurately locate a fixture.   I got tapered silicone rubber plugs to go into the drill bushings and plug the vacuum for when we are not using a particular fixture.

Bob, do you have a video or anything that shows how Taylor does their necks? I've seen the factory friday vids, but don't remember seeing that.

I know that martin actually presses frets in the fingerboards before they are on the neck. I want to say that they carve the neck before gluing up to the f'board tho.

The main thing with designing a vacuum jig is vacuum surface area. Also keep in mind that in addition to vacuum, you should also have locating pins or dowels...this makes vacuum fixturing much more effective since the only way for the part to come off would be straight up.

I have to admit that my home shop setup is pretty basic. I use double-stick tape and dowels in conjunction with some basic locating jigs. I'm keeping it low tech and low-cost until if and when I decide I really want to get into production on something. I'll post some pics at some point...but prepare to be underwhelmed!

_________________
http://www.PeakeGuitars.com

Here's a sample of a couple of fixtures on my machine (which is not very nice but it works, sort of).
I drill locating pin holes in the table along with some taped holes to hold the fixture down, much like Rick is doing. I write down the coordinates of my set point relative to the home position on the fixture. This way I just put the fixture in the same pin location every time , move the machine to the coordinates, g54 it and I'm ready to go.
If I had a Fadal type machine I'd mount and Aluminum tooling plate to the bed, cut it flat and place my bushings in the plate and just let live there. That is if it was a pretty dedicated machine. I believe this is waht Taylor does, only with out the aluminum tooling plate.

The fixture in the front holds the head stock down to cut my peg head shape. The neck registers in the truss rod sloton a 1/4 spline on the angled riser.

_________________
Jim Watts
http://jameswattsguitars.com

Taylor drills holes in their tables and inserts bushings in them for precision-ground pins that they use to locate all their fixtures. Same deal on my machine. It's a good system.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

[QUOTE=Parser] Bob, do you have a video or anything that shows how Taylor does their necks? I've seen the factory friday vids, but don't remember seeing that.[/QUOTE]

Can you be a bit more specific, I don't know exactly what you're referring to not having seen.

They take their blanks, mill for the neck joint, and put them together in a gluing jig.

From there, these blanks are fed through the rotoshaper to put a clean back profile on them and also to hog off some of the waste on the back of the neck shaft(cuts it to a trapezoid cross-section)

From there, as I remember it, they have the headplates, truss rods, and fingerboards installed. It was a year ago I was there, so I'm slightly foggy on the exact order. I'm pretty sure fretting is done with the fingerboard on the neck. They cut their own inlay pockets, but they don't cut their own pearl.

Then they flip 'em over and machine the back profiles on the Fadal with a 1" ball. They're then flipped over, headplate up and horizontal, and the top of the headstock and headplate inlay pockets are machined.

It's been almost a year since I was there, and I have a poor memory, so the details might be slightly off, but this is pretty close to their process.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services