Tone Bar Fitting Jig - it's kind of insane.

[Marty Jacobson]

Hello everyone,

So I have fit many, many tone bars over the last week or so while doing some experiments with my new production mandolin design. It takes me a couple to three hours to get them cut, sanded, scraped, and shamed into joinable shape.

One of the problems is that the more angled the bars/bracing are, the more of the serpentine shape has to be truly transferred to the new bracing member. It can be really crazy geometry. A spindle sander does not really work except for very rough shaping.. you all know the drill.

I thought it would be nice to have a fixture which would automagically get the tone bars to match the geometry of the top.. at least within 5 or 10 thousandths of an inch or so. From there I can scrape it to final shape, no problem.

Here's what I came up with. It's just a mockup at this point, so don't be too hard on me.. the test is just in a piece of cheap plywood.



It's skeleton is a parallel torque reaction arm which I also use for binding. It keeps the cutter dead plumb but free to move.



It also counterbalances the router so it only "weighs" about 8 oz. Not enough to damage the spruce plate, but enough to keep the assembly grounded.



There is a Delrin probe which slides really smoothly but doesn't dent the spruce. The probe is coaxial with the cutter.

The workholding fixture above the unit holds tone bar stock.

Setup consists of aligning the tone bar's location with a line on the table, and clamping the top plate in its quick release rim fixture to the table with the blue hold-downs.

The torque reaction arm keeps the cutter from moving wildly. Because of the torque counter-action of the "elbow" on the unit, this router can be operated with one finger, it is very stable and does not "dance around" like a Dremel.



So far just proven the concept in plywood. It takes some thought to get a really smooth surface, but it is possible in probably 5-10 minutes per tone bar. Probably another 20 minutes left for final fitting, but I think it's a huge advantage. Works for X-bracing, too, just needs a different X-bracing holder above the unit.

It's basically just a 3D pin router. But I think it will work for me.

[amowry]

Nice jig! I'd encourage further experimentation with doing it by hand, though. I'd say you could get it down to about 25 minutes per tone bar, which is less than your overall time with the jig and final fitting. I scribe mine with dividers, cut them on the bandsaw, and get as close to the line as I can on the disc sander. After that it's only 15-20 minutes to chalk-fit, scraping with a razor blade.

You might also check out Tom Ribbecke's setup, though I've always been somewhat skeptical that one can get a really good fir with sandpaper: http://www.frets.com/FretsPages/Luth...onebarfit.html

[grandcanyonminstrel]

Very nice jig Martin, but I'm in agreement with Andrew. I spend under a half hour fitting the tonebars, using a very sharp handplane. Most of it just comes from making a lot of them. I believe this is a new model for you, no? Once you get used to that one, you'll be able to get things very close by eye and then it will be just a small amount of fine tuning. Since you've got a nice pimpin' cnc setup, you could also make an inside mould with the same profile as your inside arch and overlay it with sandpaper and fit right on that. A lot of guitar builders use this method.

Early on, I use to spend a fair amount of time like you are, then I started dong a lot of bass bars on double basses...suddenly I started looking forward to fitting them on mandolins!

j.
www.condino.com

[sunburst]

Another voice in the chorus. I by no means mean to discourage anyone from their chosen methods, I realize that some people love building and using jigs to mechanize processes and doing what we enjoy is at the root of all this anyway, but to me it's just "speeding up the fast part". I do as the others here have mentioned, mark a line, bandsaw close, refine with a few strokes of a block plane, and it's on to chalk fitting with finger planes followed by scrapers and sometime final fitting with an emery board. Of coarse my aversion to screaming routers and the dust they make is probably part of my choice of method.

[Marty Jacobson]

Thanks for the feedback, Andrew, James, and John...

I completely understand and know that I could build up the dexterity and knowledge necessary to do this completely by hand in under 30 minutes per instrument. I really respect you gentlemen that have a few hundred instruments in player's hands and have done the reps to burn this stuff into muscle memory.
But I'm trying to hit a really low price point with these things, and I have to cut down on the opportunities to obsess... or I would never get anything shipped.

CNC fitting is out because I tweak the grads on each plate from the inside, so they are all different. I could vary the plate graduations from the outside, but then things like neck angle would have to vary from instrument to instrument.

The goal of this is not to replace fitting by hand, just to make a no-brainer way of getting it close (like bandsawing to rough shape, but in three dimensions instead of 2).

My tests today have it taking more like 2 minutes to rough the brace, then it's ready to start chalk/carbon fitting. A little light scraping gets it there.

Not for everyone, I know... I'm right with you on the "screaming routers" thing, John.

[sunburst]

I don't have hundreds of instruments in players hands, probably less than 50, I spend more time on repairs than building. I tweek arches and grads from both sides of the plate each time, each session of fitting tone bars is a relatively unfamiliar experience because it only happens a few times a year. In other words, I can't do this from memory, and I'm really not very fast at it compared to someone who does it a lot, but even so, it doesn't take long at all to get to a state of "pretty close", then the fine chalk fitting begins. That's the part I wish I could get faster at!

[grandcanyonminstrel]

Martin:

How about we make a trade: you cnc the rock hard curly Michigan red maple backs that are averaging 14-16 hours each of hand carving for me and I'll do your tone bars!!! I've got a few Brazilian rosewood backs too....

j.

[Marty Jacobson]

How about we make a trade: you cnc the rock hard curly Michigan red maple backs that are averaging 14-16 hours each of hand carving for me....

Heh! Great idea... but this worked out pretty well. Got both tone bars fit and glued up in about 20 minutes. I should be able to fit 5 sets of tone bars in a morning.. but I'll report back after I try to do that next week.
Two curly maple, two mahogany, and one Bubinga mando on deck.

Are you including the sapwood on any of that Brazilian?

[Marty Jacobson]

And that bass top plate is gorgeous, man...
Next step is carving spruce kayaks...