MIMF

New to the board, but not to luthierie (I made a violin in 2007-2009). Around 2009, I bought wood for making a guitar, but for various reasons, I’m only really getting started now. In ‘09, I got the plates joined and thicknessed, and started making the neck. I’m following Cumpiano’s book. First mistake I made was installing the truss rod upside down, and I’m taking steps to fix that now.

My main question is, how are the sides radiused? Cumpiano explains how to make a set of radius templates for the braces, but from what I can see, doesn’t say anything about the side profile/radius. I’ve seen videos on YouTube showing radiusing the sides on a radius dish (and I’d like to make one, but can’t find information on how to achieve the correct radius). I’m assuming the sides should be thicknessed (and bent/assembled?) before radiusing. I understand that the sides should be about 2mm thick.

First guitar build. Hope I haven’t asked too many questions at once…

Welcome! I hope you get the advice you need. I am reluctant to provide an answer since it has been more than a decade since I read that book. I don't use the same method he does so I don't want to confuse the issue by recommending something that might not be compatible with his method.

Are you making the steel string or the classical guitar in that boo? I know the classical is made on a workboard. I think I remember that the steel string is as well but I might have that wrong. I use radius dishes to radius my rims but if you are building on a workboard, you neck angle and top geometry are built into that system.

I mostly am posting to make sure you are aware that he no longer recommends the taper pin neck attachment. If you google the book and find his site, he has posted several updates to the book. The most important one is a more reliably executed neck attachment for the steel string guitar.

Thanks! Yes, I know about his improved bolt-on neck design. I’m building the steel-string guitar. My violin came out pretty good, except for being overly large (I made the top and back too big, then had to modify the side garland to fit, but that’s another story, lol!

I don't radius the rims - I just leave them flat and clamp the top and back to them as it looks like Cumpiano does. The difference in radiused and flat is about 0.010" at the inside of the lining for 15' radiused back, and 0 at the outside, which is a thin enough glue line for me. The 25' top will be about half of that.

But that's just me.

Here are his updates if you need the link.
https://www.cumpiano.com/gtt-updates--commentary

Usually the back will have enough radius to require the sides to be radiused quite a bit. Radius dishes make it easy but there are other ways to do it.

On my first couple of guitars I made a radiused sanding board with 24 and 16 foot radii (which I think is what Cumpiano shows) and attempted to sand the little angle into the rim and kerfing. The problem is that the back slopes at the head block and the waists need to be taller than the bouts - the only good way to do that is by sanding the rims against a radius dish while still in the mold.

Making radius dishes is a hassle but the usual way is to make a router sled that spans the circular piece of MDF and just keep going around and around. I found some pretty reasonably priced at one of the lutherie supply houses and those became my chosen top and back domes.

I have also seen a couple of sets of plans that have the actual side profile which will get you close but still require some sanding.

Ideally this would be done in the mold. Notice the gap at the waists - that needs to be sanded into the dome shape

There are a couple of ways to dress the edges of the rim (usually called 'profiling') to fit the radii of the top and back. A dish makes it easier, but first you have to have one...

One way is to make a stick somewhat longer than the length of the body with one edge dressed to the proper inside radius. Since these radii are usually about 25' for the top and 15' for the back it's something of a challenge to make one. There is a formula that gives the radius if you know the length of the stick and the rise, it's
((A/2)^2+B^2)/2*b = R when
A is the length of the stick
B is the rise
You have to keep the units consistent, of course; if you're working in inches then 14'= 168"
I'd make a long 'brace' with the right outside radius, and use that to mark the inside radius on another piece. You can stick some sandpaper to the 'brace' and use that to even off the curve of the 'inside' stick.

Once you've got that you can start with the two end blocks at a bit more than the right height. Cut the sides to full height from the tail to the waist, and taper straight down to the right height at the neck end (plus a little for insurance all along), and then bend them. Once the sides are glued to to the blocks you can lay the radius sticks along from end to end and side to side, and work out the profiles that way.

I use a shim trick to get outside radii. If you stand on the beach on a flat calm day and look out to sea the horizon is about five miles out: that's where your line of sight intersects the curve of the earth. A rise of about 6' over five miles defines that radius. Suppose you stick a shim onto the sole of your plane behind the cutter with CA glue. The sole of the plane is the line of sight and the shim thickness is your height. Given the height of the shim and a base line from the sole of the plane just in front of the cutter you can find the radius that the plane would make. You do have to take the thickness of the cut into account; the less the cut the the closer the contact point will be to the front of the slot, and the tighter the radius. I use this to radius my braces, and you can use it to make the 'outside' radius to dress off your stick.

One you've got that radius you can also make two concave rails. Attach those to the sides of a box about 2' wide. Mount a lazy Susan on the box and screw a 2' diameter piece of MDF to it. Make a sled with rails that your router can run between that will ride on the radius rails. Set the router up to take about 1/8" off the middle of the MDF disk, and run it across. Rotate the disk and repeat. Increase the depth of cut until you've got a radiused surface all the way across. Do this outdoors if possible: it makes LOTS of dust. I find it helps to glue a piece of 1/8" or 1/4" Masonite to the bottom of the disk first: there seems to be a fair amount of built-in stress in MDF, and it tends to pull up around the edges as you remove material from the upper surface. The Masonite helps keep it flat.

Thanks, everyone. Lots of good information here.

So the question of profiling the sides/lining is being discussed. I've never done it, as I've followed Cumpiano's construction methods for the most part. He leaves them flat, and glues the top and back, radiused by the braces, to them with no profiling since he uses a flat sanding board to prepare them. I do profile the neck and tail blocks.

Using a radiused sanding board is also problematic at the waist areas, as Freeman points out.

So the glue thickness is greater at the inside edge of a flat lining than is optimal, more so for the back than the top. How much of a problem might this be in the long run? I've had no failures, but it's only been a few years down the line.

Any opinions?

The couple of guitars that I built using a radiused sanding stick required considerable clamping force to close the gaps all the way around the back. The top isn't quit as bad. When I switched to using a radius dish the back (which was also braced in the dish) just seems to drop on and not require a lot of pressure.

If you leave the top of the kerfing flat as Peter points out you will have a little gap on the inside edge. You are going to route the outside off when you bind it, I like the idea that the top of the kerfing is the same angle as the dome of the top.

I have built according to the Cumpiano book and I seem to remember that the sides do get profiled for the back. Clamping to flat rims is a really risky idea, in my opinion.

Barry Daniels wrote: ↑Wed Aug 10, 2022 5:41 pm I have built according to the Cumpiano book and I seem to remember that the sides do get profiled for the back.

Well, someone will have to point out to me where in the book that is covered - I have never seen it and can't find it. He does profile the sides longitudinally along the length of the guitar, but not across the width of the lining (that I can find).

A traditional Spanish method of building on a solera is to use a flat sanding board to taper the sides on the back edge. The cross braces are then glued into notches in the liner with their tops proud of the rim. They are radiused by eye to heights that produce a uniform arch along the length of the back on the center line, and the edges of the rim are slightly beveled to match the angle the back makes with respect to the sides. The back is then glued onto the rim using string over the solera to clamp it down. According to Romanillos, Torres used to apply hide glue to the cross braces, which would gel as the back was being clamped down. When everything was in place he'd paint a line of alcohol across the back were each cross brace was, and light it on fire. The heat would melt the glue and stick the back to the braces. Evidently he sometimes used too much alcohol, and was unable to scrape off the scorch marks. Unlike the use of a sanding stick or dish form the radius of the back is not uniform everywhere. It's difficult to make this come out right and avoid a 'dimple' in the waist on wither side.

Some folks have made dish forms by putting a circle of shims around the edge of a thick disk of MDF or flake board, and then screwing a piece of 1/4" Masonite down in the center. The resulting curve is sharper in the center than it is around the edge, so uniformly arched braces don't exactly match it everywhere. Ideally you'd sand them to shape on the dish where they are to go. The difference is small, though. I know at least one very good local maker who uses this method and has had not problems.

He made a jig to radius braces on similar principles. It's got a slot that a brace can fit in that has a shim on either end to match the desired rise in the center. The brace is put into the slot, pushed down in the middle, and clamped in. The protruding material ate either end is taken off with a router bit in a shaper table that guides off the straight upper edge of the jig. Again, this results in a a brace that is more curved in the center than a true radius.

Colleen McTigue wrote: ↑Mon Aug 08, 2022 2:44 pm
First mistake I made was installing the truss rod upside down, and I’m taking steps to fix that now.

I did the same thing with the truss rod on my first build but didn't realize it until after it was strung up.

Several things to consider when putting a dome or contour into the sides with a hand plane:

The waiste will be the high point. Exact numbers regarding the arch aren't as important as a good fit.

The inings will be angled down and outward to accommodate the plates. Remember the bottom of the heel block also.

I use a dished sanding plate for the backs which usually works for a good fit. But I dome the tops of my guitars in an irregular manner.
The 3 essential considerations for the top are the height of the saddle in relation to the nut and the resulting action height at the 12th fret.
In that relationship is the angle (backward or forward) of the neck (nut) in relation to saddle height. For a scale of 650 mm being off 1/2 of a degree will alter the saddle height by 3mm. I shoot for 1/4 degree accuracy with a target saddle height of 12.5 mm. That is a tough goal and more than once have had to make another neck to acheive it.

For that reason too I never install the truss rod until after the neck angle is right where I want it. It gets in the way. The other reason being that I use the truss rod channel for a guide in aligning the neck to body center line.

I like to put a slight back angle in the neck and start with a 90 degree neck to top body joint. The doming of the top begins about the midle of the soundhole and is concentrated in an area encompassing that point to the back side of the bridge. I start with oversize sides and plane the contour into the sides using guide marks to keep track of my progress.

And no, Cumpiano and Natelson don't cover this well. At least in the 20 + year old edition I have.

I finally found where in the book Cumpiano covers “radiusing” the sides… it’s on page 213-214. He shows planing the upper bouts from the waist to the neck block, then using a (flat) sanding board for the final contour. He sands the neck block and upper bouts, sands the “peak” (created by the plane), and a bit off the tail block to create a smoothly curved profile. This is on the side where the back will be glued. No arching on the top side, from what I can see. It took me so long to find this because I was looking for it in the “sides” section, and it’s in the “assembly” section. Starting to make sense now.

Yes, that's how he profiles the back longitudinally, but the transverse arch/radius across the lining (which is what I think we're talking about here ) remains flat.

Anyway, I think that having the appropriate radius on the lining makes for a better joint - but out of ignorance I've just never done it that way, and hope that gluing the radiused back and top to the flat lining has given adequate strength. (In my endeavors I try for superior or good, but am satisfied with adequate )

I use these instructions to build my radius dishes. I have checked them carefully and they are correct. Mine are 24" in diameter.

https://forum.ukuleleunderground.com/th ... ish.30203/

-Eric

I'm less experienced than everyone else.

What I did was just set the braced top/back onto the rims...and use a small hand plane to remove side wood until the surface is flush/passive fit.
Then, I glue the surface on.

Maybe I'm doing something wrong?

Backs usually only have cross braces, and they can straighten out, so the lengthwise arch is less than it 'should' be.

This should work if the sides were presented from a template. This is how I am doing my current build.