Official Luthiers Forum!

For those of you who are also milling necks on a CNC, has anyone come up with a cool hold down/registration scheme they would be willing to share?

It seems every time I mill a neck I vow to make a better jig. Currently, I use two:

For the headstock I screw the blank outside the perimeter of the headstock to a spoil board. The spoil board has an 18" long 15° wedge to support the shaft with a spline down the middle to register the truss rod slot. The shaft gets clamped to this wedge support.

For the shaft I have a 2-1/4" wide 2" tall cradle with a 15° ski ramp to support the headstock. The headstock is attached through the tuner holes/slots to the ski ramp, the shaft of the blank extends past the heel stack and is screwed to the cradle at that end. The cradle also registers from the truss rod.

What are my issues: mostly that using two jigs invites registration errors. If one isn't perfectly lined up along the X axis I get a subtle kink in the neck that is remarkably obvious. If they don't have them perfectly inline in the Y axis I get weird neck shaft to headstock transitions.

I dream of one pivoting jig, but so far it eludes my engineering prowess to visualize and build. So rather than struggle alone, I am hoping for a group think.

PS: these are acoustic guitar neck that have 3.5" heels from bottom of the fretboard to top of heel. The forces on the heel when milling the top create a lot of torque on the blank. The hold down for an electric neck is obviously much simpler.


Sent from my iPad using Tapatalk

I'm making thinlines, so my heel is not as long as yours and my methods change often, but this is how I'm doing it today:
I start with a squared up neck blank big enough for two. With simple templates I'll draw out basic side profiles and using the nut location will route for truss rod slots, anchors, adjustment cavity and index holes. I'll do both faces then bandsaw them apart.
BHGO6504 by llokluas, on Flickr
Truss rods get glued in and splines get planed or sanded flush:
IMG_3774 by llokluas, on Flickr
Neck shafts then go face down on a holding fixure, indexed with small nylon dowels and held in place by clamps. I limit what parts of the neck I'm machining to leave places to grip the workpieces:
IMG_3824 by llokluas, on Flickr
Next up I'll glue wings if necessary, and true up headstock face. Decorative veneer gets glued up. After nut shelf gets squared up, I'll use a template-at this point a proto template- to layout tuner/locate holes. The template has holes that fit into index holes in that I first put in the face, along with truss rod stuff. It's hinged at the nut area and lays down flat onto the headstock face no matter the angle. (I typically use a shallow 10 degree) The tuner locations are drilled with 1/8" drill using this jig. Not a very good pic, but maybe you can see what I'm doing.
IMG_3829 by llokluas, on Flickr
These holes are used to position headstock upside down located with 1/8" dowels. the headstock is then profiled. I have a couple different devices to hold the work, here's one:IMG_3838 by llokluas, on Flickr
(The tuner holes are left undrilled, just the 1/8" locating holes until I tuning machine brand and type is chosen, typically 21/64" or 10mm.)
Using the completed headstock profile itself as indexing device the neck goes face up into an elevated box with slight recess. There is a mechanical clamp located at one end and I also use vacuum at the headstock itself.
IMG_3913 by llokluas, on Flickr
In this position I can do anything the facecap needs, such as logo inlay, truss rod acess, etc. I could cut tuner holes too if needed, but typically leave them until later:
IMG_3912 by llokluas, on Flickr
I'm not sure if you can use any of this but there you go-

Thanks Saul. That is pretty close to what I do. Only difference is that I machine the headstock faceplate off the neck and glue it on later. I have a similar wedge as in your 5th picture for the headstock (which I mill first) and then a holdown similar to your second picture for the shaft. Again, only difference is that the neck sits elevated so that I don't cut into the jig.

No issues with registration between the three different jigs?

Not really. Everything relies on those first six locating holes that are on the face of the neck shaft. So if I've zeroed out everything,and the machine knows where those are, it's always going to cut in the right places.
I have some threaded holes in my cnc table. Most of my permanent jigs index off of those and are quickly swapped out accurately. (it's only the one off/temp jigs that might need a little finese)
At this point the hinged tuner hole jig is the main weak link. Now that I see that it works well, I'll make another from a more robust material and use drill guide bushings.

Anyone else? How do you do yours?

The system Saul laid out is very close to what I was using.

Your registration problems come from one of three places:
1- Your indexing scheme isn't indexing accurately (truss rod slots aren't a great index)
2- Your fixtures aren't mounted squarely or securely (so the found edge doesn't indicate actual location)
3- Your offsets for your fixtures aren't set properly (edges aren't found accurately)

The place to look first is definitely that you're not using pins for indexing. Pins, almost universally, are a superior indexing method. Two pins are sufficient, though using a few more like Saul will make the mounting a bit more stable. I used to use two on fretboards and four on necks and hold everything down with vacuum. I always used 1/4" or 3/16" steel dowel pins, but anything will do the trick. Drilling the second set of holes using a jig mounted in the first set is also the way to go.

If you're using a good blank indexing method, then the remaining issue must be your fixture. I always lined mine up using two square edges with an edge finder, and a top surface. I had dowel pin holes in my machines table, so they could be taken off and put back on in the same place. The one issue you can get (and why I checked alignment every time they went on) is that you can have dust under something, one side clamped too hard, fixture warp, etc etc...so it's worth doing a quick sweep with an indicator to make sure your fixture is square, flat, and where you think it should be.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Here's a method I developed in order to take the neck on and off the machine but still retain precise registration to the vacuum plate. Two pins laid down at 90 and two laid down at 180 locate the mounting plate down onto the vacuum plate perfectly.

Note also the method of securing the headstock to the plate. Those are just little o-rings stacked on little allen head screws. The o-rings fit just a bit snugly into the tuner holes when not squeezed...but when the nuts are tightened and the o-rings compress...not only do they hold with INCREDIBLE power...but the system draws the headstock FLAT down to the mounting plate as it increases holding power.

_________________
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

For what it's worth...those six o-ring posts are more than enough to hold the headstock down to the plate without the need to hold or secure the root of the neck in any way.

_________________
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

I'm interested in trying the headstock layout template, similar to what Saul has. A tapered section to index off the neck with a hinge connecting it accurately to the headstock pattern with drill bushings. This is to accommodate a 15 degree angle. A quality hinge would be necessary to remove any side to side movement. I've done some searching, but the only thing that looks close is a piano hinge. Anyone know of something better?

Bob