Prototype Archtop Build

[Geoff St. Germaine]

I've started a prototype archtop guitar build. There's nothing revolutionary about it, it's simply a prototype because I've never built an archtop guitar before and there are a number of new skills that I'll need to develop. Therefore, I'll be developing them on a prototype and then moving on to more expensive woods. If anyone's looked into purchasing the traditional type of "tonewood" for an archtop guitar, then you'll see that it is at a considerable premium for price. The aim with this guitar is to use suitable but inexpensive woods so that if something unfortunate were to happen then the cost of redoing something like a top or side would be low.

I selected red cedar for the soundboard. Spruce is the most commonly used wood for archtop guitars, but in searching through the woodpiles at my local hardwoods shop, I was only able to find good quality quartersawn red cedar. I also looked at their sitka spruce and douglas fir, but I couldn't find anything decently quartersawn or without a lot of grain runout. The total cost for the board was about $15, far better than the $100-$300 I've seen for AA through master grade sitka spruce from specialty tonewood suppliers.

For the back and sides it was a bit of a toss up. The guitars I'll make after will be flamed maple, so plain maple would be one obvious choice, but I find it too boring. Walnut seemed to be a more aesthetically pleasing choice and I was able to get the wood for the back, sides and neck for about $40.

With that material collected I've started on this adventure. I'm following Benedetto's book pretty much exactly.

Here are the specs for the guitar:

17" x 3" cutaway body

Red cedar top

Black walnut back, sides and neck

Ebony fretboard

25" scale

22 frets

White MOP block inlays

Ebony headstock overlay

White binding

Ebony bridge, tailpiece, pickguard

Kent Armstrong suspended pickup

Chrome Gotoh open back tuners

[Geoff St. Germaine]

As following with Benedetto's book, I need some moulds and jigs and fixtures to get this done. Bending wood is completely new to me, so I've been doing a lot of reading. It seems that a side bending machine type of setup gets consistent and easy results. The downside to this compared to bending over an iron seems to be mostly the difficulty in setting up for many different shapes. As I haven't planned out making a bunch of different body shapes, I think the bending machine type setup will work well and will probably give me better results in less time.

I built up a set of side moulds using 3 layers of 3/4" MDF. I sealed the inside with polyurethane.

I made a template of the body, which was mostly to use to set up other jigs and fixtures, such as the body mould, the inside mould for the bending machine and some inside mould turnbuckle clamps.

The bending machine I've built is based around some designs I've found on the internet. It uses three 200 W bulbs wired through a 600W dimmer switch. The body inside moulds are built to be set on top of the base so that the base can be reused and if I did decide to make some different body shapes then it just needs a different set of moulds. The other option to me was to use silicon heating blankets. I still may invest in them, but at $200+ for a setup, I figure I'll try this manner first, which seems to have worked for a number of people.

Here's what I came up with, shown with the non-cutaway side mould.

The mould is 3/4" MDF sides with zinc plated flashing mounted on it. The idea as this is a sandwich of the body side wood and another layer of flashing to support the backside of the side during the bend, which is supposed to reduce the chance of cracking or a faceted look to the bend.

The eyelet bolts on the base of the bender take turnbuckles that connect to 3 clamps, a curved one matching the waist and two flat ones for the tail and neck ends.

I'll try to update some more later today. Thanks for looking!

Edited January 6, 2013 by Geoff St. Germaine

[Dean]

Very nice , you have skills ,looking forward to your build.

[Prostheta]

Have we all gone archtop crazy this month? Is this the way forward for 2013 then?

I like it.

[maull]

I would love to try my hand at one but think I need a few more builds under my belt first. Great idea on the home built bending forms.

[Geoff St. Germaine]

Hey guys, thanks!

Prostheta, yeah, it seems like lots of guys are doing archtop builds here right now. It is great. I'm a bass player, so I'm watching your thread closely.

Here's where things went with the carving.

I drilled a number of holes to establish the depth of carving for the top and back. Unfortunately, my drill press depth stop gave out and I punched through the top near the end of the drilling. On top of that, the depth stop had been slowly allowing some slippage before it broke completely. The result was that when the top was carved it was too thin at the edges. This required adding some material to the edges. Fortunately I had enough of the quartersawn red cedar left to fix it, but still it's far from ideal. Here are some photos of how things went.

The top and back drilled:

My carving cradle. The back is a mirror of the front.

Carving away. A lot of work and a lot of shavings. Seriously, between the inner and outer carves of the two plates I filled an entire garbage can. Curled shavings take up a lot of space.

The top is getting close here. I used a combination of chisels, gouges, a palm plane and a spoke shave to get this done.

You can see around the rim on the last picture that the entire way around the drill starting going too deep. A poor craftsman blames his tools, but it is what it is. I blended it down and had to add 1/8" of material to the back as I was down to 1/8" on the sides, which is too thin, especially for cedar rather than spruce. Based on some reading I'd done on cedar vs spruce soundboard wrt stiffness, the red cedar top is 25% thicker than recommended in Benedetto's book, which is based on a spruce soundboard.

Edited January 7, 2013 by Geoff St. Germaine

[Geoff St. Germaine]

Next it was on to the inside carving. I used a 5/8" forstner bit with the point ground off to rough out the inside. I had done it like this before on a carved maple top for an electric hollowbody with very good results, so it seemed like a timesaver in this application. I set the drill to leave 5/16" all around to allow for carving the back down to a uniform 1/4" and the top to a graduation from about 7/32" to 5/32". Once this was done it was cleaned up, mostly with an Ibex palm plane that had a convex sole. I used my home made caliper to measure the top, made from the plan in Benedetto's book.

Here's the walnut back starting to be roughed out with the forstner bit.

The inside of the top. You can make out where the lamination went on with the ring around the rim. By the end of sanding this was quite a narrow line, extending only very slightly beyond the flat gluing surface around the rim.

The completed back, clamping to the carving cradle.

Since I'd punched through the top very near the rim the traditional f holes I was planning wouldn't work. Off to google images and I found a nice solution from Bill Moll's work that looks somewhat traditional but had the holes near enough to the edge to encompass the error.

Here they are cut out and roughly cleaned up.

Now all that is left for the plates is to clean up the f holes, bind them and install the parallel braces. I've starting bending the sides, and I'll try to get that updated soon.

[Geoff St. Germaine]

So I moved on to the sides so I could get them in the mould and get some concurrent activity going between the top and the sides. I planed the walnut down to about 0.120", which was as thin as I could get them with my planer. I then set up my oscillating spindle sander to thickness sand them further. After doing a fair amount of reading, it seemed that 0.090-0.100" seems common, so I took them down to, as best I could measure, 0.095". I further thinned the cutaway section of the side to about 0.088". Then I went ahead with bending them.

My process was soaking the sides in hot water for about 1/2 hour before attempting to bend them. I had read about using a thermometer or also that water sprayed on the forms to just start to sizzle and pop when they're hot enough. I went with the qualitative measurement and went about bending.

Here's the machine, heated up and ready to go.

I clamped the waist into the mould first, as this seems to be a pretty common procedure and logically made the most sense. It only took about a minute to get it clamped all the way to the mould. I listened closely for any cracking, but I didn't hear any.

Next I clamped down the tail end before moving on to the neck end. Overall things went very well. There's not a lot of tight curving as this is a 17" body. I hoped this would be easy, as the cutaway would undoubtedly be more challenging. Fortunately it was.

I got the non cutaway side clamped into the mould. It was time to do the cutaway... a more daunting task.

Machine heated up:

Side clamped in:

Overall the cutaway took more time and I was very patient pressing it in place. I used my hands to get it to start conforming to the mould before I used the cutaway shaped caul to pull it in tight to the machine.

For both sides I let them cool after the initial bending before heating them back up and cooking them for about 10 minutes. I had read that this helps "set" the bend. Whether this works or not, I had very little spring back and the sides went into the body mould easily.

[Geoff St. Germaine]

Next up was fitting the neckblock and tailblock. I made both from vertical grain cedar. Vertical grain spruce is recommended by Benedetto, but cedar made sense as it is what the top is made of and I had plenty on hand without dipping into the spruce I have for the other two I'll be building after this guitar.

The dimensions of the neck block were as discussed in Benedetto's book. I really love that man's guitars and his book is excellent. It's a great resource, especially for the work on the body.

Here are the sides with the tailblock and neckblock clamped in and drying.

That catches things up to where they are now, so I imagine it will be a little more slower going on the updates.

[Prostheta]

Prostheta, yeah, it seems like lots of guys are doing archtop builds here right now. It is great. I'm a bass player, so I'm watching your thread closely.

Totally. I love seeing people bringing fresh directions to the table for PG such as Chris' OMarlin! I have literally no direct experience of building acoustic boxes so I am going into this armed purely with information and a rough guiding line. Threads like this will no doubt influence my working methods so I should be thanking you rather than the opposite way around!

I guess my decision to go archtop rather than acoustic is on the basis that I think the bracing methodology for an acoustic bass would be far more involved with a more delicate balance to get good results. An archtop is great for working on all of the aspects of an acoustic without having to worry about producing a house of cards box or one that is so stiff it might as well be a thinline Tele. :-D

Your bending jigs are awfully well planned. My initial idea was to glue laminate a load of 0,5mm birch up with a facing veneer however now I might heat bend them first to truly lock in that shape. I need a metric tonload of turnbuckles methinks.

[Geoff St. Germaine]

I can see the bracing and tuning of a flat top acoustic being more difficult on a one off or limited production of instruments. All flat top acoustic basses I've played have suffered from the same low volume and lack of bass compared to what you would want or expect it to have compared to an acoustic guitar. I think the archtop may be more forgiving as you can adjust the response of the body via the recurve after string the instrument but before finishing if you wanted. I imagine a large part of the difference is going to come down to body size. Look at the size of a double bass... unfortunately that would make it too large to play they way we play bass, so perhaps the loss of volume and bass is the compromise we accept from an acoustic type bass.

I spent a great deal of time making sure the bending machine would work properly. Bending the sides is the task that I am most uncomfortable with, especially when I'll be working with the master grade flamed maple I have for the next two. I won't say I can't afford to screw those up, but I really don't want to have to replace any of the sides as they are quite expensive.

[Prostheta]

Exactly. I don't think any of the sizes we normally see in acoustics are appropriate for bass, even dreadnought. Obviously there is better and worse however ideal is way out there from my mental calculations. The analogy of a double bass (as you suggested) to a cello is in the right magnitude of scale I think.

To me, an acoustic bass would be a personal space instrument rather than one meant to project or to fill a room. A sound port in the side would likely be on the cards in that respect. For the moment I would like to take the important learning steps of side bending and producing a box in. Maybe after that is in the bag I could move onto a full acoustic.

[Geoff St. Germaine]

I think as a personal space instrument the idea works. I've been toying around with building something to travel with. I deploy on navy ships as an aviator and space is limited. I've been thinking of either a small acoustic or a bass with a built in headphone amp. I think in the end the headphone amp idea is going to win that one.

Here's the progress I've made.

I got the sides braces and linings installed. It's a bit of a tedious job. I cut the linings too thick. Benedetto's book suggested 3/32" thick spruce. Well I read 3/16" and as I was using cedar I bumped it to 1/4". In the end it won't be a big deal. It saves me having to add bits of kerfed mahogany to the braces. I'll just carve them a bit in the center to remove a bit of material.

I used mahogany reverse kerfing. The claim is that reverse kerfing makes for stiffer sides. I'm not sure if that's true or not. For other two I'll be building I'm going to use regular spruce kerfing. I laid out the braces roughly as per Benedetto's book. I used about 4" spacing around the lower bout and 6" spacing around the waist. The upper bout was again 4" spacing and the cutaway has no bracing as it is so much stiffer anyway.

Here's getting the linings and braces installed:

The final bit around the cutaway... a bit of a pain to install.

I used a flat piece of scrap plywood with a piece of an old sanding belt glued on (80 grit) to level the linings and braces to the sides. It worked like a charm.

Here it is levelled. All that is left is to carve the braces a little.

I've always thought that a bent body rim with installed braces and linings looks fantastic. This is the first one I've done and it looks even better in person. I also used to think that it took a master builder to get a body rim like that... apparently not.

An interesting thing that I'd read about was colour changes in the sides due to the bending. Apparently this is caused by having the wood against the heated metal in a humid environment. I'd read of koa turning green and apparently walnut does the same.

It's a little hard to tell, but there's a green-grey hue to the sides. I spot testing sanding it and it appears to be a very thin layer of wood on the surface that has changed colour. I'll try to avoid that next time and from what I've read, wrapping the sides in craft paper while bending will prevent at least most of the colour change.

Next it will be getting the f holes finished up and the top braces installed. I expect both of those jobs to be fairly tedious.

[Prostheta]

Hmmm. I am wondering if the change is due to the Tannic acid content reacting with the water and the steel. Certainly, iron tools can quickly discolour woods with high content like Oak and Chestnut. I wonder if that is what is at work here?

Yeah, sharing a bunk with another human plus a huge double bass might not work....and I thought they smelled bad on the outside....

What was the objective and methods with the sanding stick? Obviously to provide a flat surface for glueing, however were you working to a specific outline already defined by the sides (dropping the linings flat) or measuring specific curvature? I guess that not having done this, I feel I would want to factor flex and/or twist in the rim out so dialling it out of one side doesn't make it harder on the other. I presume the rim is quite stiff by this point?

[Geoff St. Germaine]

Yeah, that might be the issue. The heat and moisture would surely speed up any chemical reaction that might otherwise take weeks or months to show up.

The objective with the sanding stick was to drop the neckblock, tailblock, linings and braces to the sides while keeping them planar to the top surface. The sides were already at the correct height. The rim is very stiff and the sides were locked into the body mould with about 1/2" sticking up above the mould, so any flex should be negligible.

[Geoff St. Germaine]

This evening I got the braces cut out and the f hole binding finished up.

I cut the braces from quartersawn red cedar. The blanks were 5/16" x 7/8" x 22". I marked out where they needed to be placed as parallel braces from the Benedetto book and then transferred the top plate's curve to each of the brace blanks. To transfer it over I could have used a compass, but a piece of scrap dowel and scrap wood that got the pencil lead up to the bottom of the brace in the deepest part of the carve worked just fine. I traced the curve on and then cut them out with the band saw before cleaning them up with my spindle sander.

Tracing the curve onto the braces:

After that it was the tedious job of fitting the braces. The braces need to sit perfectly flush with the top plate so that the top plate is not pulled into a different shape when glued to the brace. The brace is much stiffer that the top so a misshaped brace would deform the top plate. Apparently instruments in the cello family are intentionally built with braces with more curvature than the top to tension the top, but over time this tension drops with a noticeable change in the voice of the instrument requiring the bass bar to occassionally be replaced. Since we don't want to open up the guitar when the tone changes, having the brace exactly match the top curve is essential.

Since the fit was so tight clamps were only necessary to get glue squeeze out and with the top plate being quite delicate, little pressure was used.

Once the glue was dried, the braces were planed and trimmed. The height of the brace at either end by the end of the planing is 3/16" and the height in the middle is 1/2". The top plate is very noticeably stiffer along the longitudinal axis.

Here's how the completed braces look. You can see that 5/16" difference in height between the ends and center of the braces is still more than made up by the curvature of the top such that the braces have a concave profile.

Here's how the inside of the top and back now look with the sides:

I've heard it said before, but it's a shame that most of the inside work will be completely invisible. While completely done for utility, I think the inside of the box has a very attractive "organic engineering" look to it. Oh well, the point at the end is to make nice music.

Next up will be getting the top and back plates glued to the sides. I'll be sealing the inside with a couple of coats of sealer to prevent or at least very greatly slow any moisture absorption by the inside of the body.

After that it may be on to the neck, or I may build up the other two bodies. I'm not totally sure. The only thing new about the necks are the fretboard extension and dovetail joints... all of the other work I've done before. I guess we'll see what mood I'm in when I get the body finished up with binding.

[Geoff St. Germaine]

Today I got the endpin jack hole drilled and the back glued on and sanded flush to the sides.

Here's the back and sides with all of my homemade spool clamps.

I got the back glued on. The spool clamps needed very little pressure to get a nice tight fit.

Here's the back and sides, now as one piece.

The inside of the body.

Here's the endpin jack. I had to bore a 3/4" hole about 1/2" deep into the tailblock in order to get the endpin jack to sit at the correct depth. The other concentric hole connecting to the outside is 15/32", which matches the LR Baggs or Fishman endpin jacks (I have some of each).

I'll have to get the inside of the body and backside of the top sealed before I glue that up. That probably won't be for a few days.

Edited January 16, 2013 by Geoff St. Germaine

[Hitone]

Amazing work. And well documented.

[Geoff St. Germaine]

Thanks a lot Hitone!

I usually don't put as much detail, but since this is something a little more unusual I figured I'd put more detail in to get more critique and help others out if they're thinking of doing one with what did and what did not work. Once this one is done I plan to do a post with just lessons learned and what I'll be doing differently on the subsequent guitars.

[Geoff St. Germaine]

I've got a bunch of things done in the past couple of days, to the point that the body is essentially complete.

I got some nitro on the inside of the body to seal it. I sprayed 3 coats, which was enough to seal and start building a gloss coat. At that point I considered it to be sufficient to do what I wanted.

I decided that it would probably be easier to route the back binding before gluing on the top as the sides still defined a plane that could be easily routed from. It went very smooth, but as I learned from routing the top, it really doesn't make a difference when it's done with my jig.

I really like this picture:

I glued up the top using my spool clamps. Again, no problems here.

I routed the top for the binding, but didn't really take any photos of that.

The next task was to route the mortise for the neck. It is a dovetail joint. In accordance with Benedetto's book, the mortise in the body is tapered 1/16" from top to bottom while the neck tenon is the full width so that it "locks in" when the neck is pushed to the bottom. With walnut into cedar I imagine it should be not a lot of trouble to get the neck to squeeze in, but we'll see. I adapted my carving cradle to accept the mortise template.

I had to break out some geometry and diagrams for making up the template. I'm using a 3/4" dovetail bit and a 1" brass template mounted in the router so that there's a 1/8" offset between where the template runs and the cutter. The Benedetto book doesn't give specific direction for making the templates to route the dovetail joint, it only describes the process, what to use (3/4" dovetail bit and 1" template in the router) and the dimensions of the mortise and tenon. Coming up with the dimensions of the templates to route against is up to the reader. In the end it worked out fine and I have a mortise that runs 2.5" and tapers from 1" on the outside at the top of the mortise (top plate) to 15/16" at the bottom.

My template and the finished mortise:

The router with the 1" brass template installed:

A couple of looks at the completed mortise:

I had a very small amount of tearout at the very top of the joint, but that will be under the neck anyway.

Thanks for looking!

[Geoff St. Germaine]

I got back from visiting my sister last night, so I was able to get back to work on this project.

The body is pretty much done at this stage. I have to install the binding and finish sand the body and that's about it. That meant it was on to neck work.

The neck blank was a large rectangular laminate of walnut, pau ferro and maple. I cut the headstock angle with my table saw and then cut the rest of the blank out with the bandsaw.

Here's the neck blanks as I'm setting it up for the headstock cut.

I routed the 1/4" truss rod slot using two bits. I first cut the slot to depth using a 1/4" round nose bit and then cut the two areas for the square parts of the LMII truss rod to fit down into the slot.

Next up in the process is cutting the dovetail to fit into the body. This was a little scary as I've never cut one before. Benedetto's book has a fairly elaborate jig to cut the joint. I used a router table and fence. While it worked, I found it to be less stable and I can see the method I used being more prone to an accident where the dovetail doesn't come out perfectly. For this reason I'll be building Benedetto's jig for the next instruments I make. Overall the joint came out very well. It's a very tight fit. I wanted to check the fit and since it's a tapered cut in the body and a straight dovetail on the neck, the joint locks in. The test fit proved very difficult to take apart. I had to clamp a piece of plywood to the body and use another clamp to push the neck out of the joint. I won't be test fitting like that any more, just a simple check at the top part of the joint will be sufficient.

Here it is pressed almost all the way down into the joint. The pencil lines across the top of the neck mark the dovetail depth and the line further back is for cutting the rabbet for the fretboard extension.

A side view of the joint. The neck is sticks out slightly into the cutaway, but not as much as I would like as it will require sanding the side down slightly to get the neck to be continuous into the side once it's tapered. This will need to be adjusted for on my later archtops.

A shot of the dovetail. Overall it's pretty good and will make a quality join with the body, but for later builds I want to have it a bit cleaner so I think a better jig is going to be the way to go.

That's it for now. Next up will be cutting the rabbet for the fretboard extension and gluing on the headstock ears and thinning the back of the headstock with my router-planer set up.

[Prostheta]

I haven't seen the Benedetto jig myself, however I would like to. I have a lot of respect for people who design jigs to perform complex operations cleanly and repeatably. The results look like magic was performed in many cases. I love it when a plan comes together.

Not gotten around to buying his book yet and I really should. I guess that looking at the operation I would be more likely to cut the shoulders as a matter of priority - given that it is end grain - concentrating on the consistency of the angle and therefore ensuring the cosmetic quality of seam. I'd then cut the dovetail by hand using a Japanese saw and clean up using chisels. The face of the heel is a butt joint and therefore not worth anything for glueup stability. It's all about the dovetail, baby. ;-D

I'd see this as a challenge for hand tool work, although more as one for workholding given the awkward angles!

Certainly, that shot of your mating surfaces looks fantastic. I would be very proud of that myself.

[Geoff St. Germaine]

It's a really good book. For some of the stuff it suggests a couple of ways of doing things. For instance it describes both a hand tool and router method for doing the dovetail. I figured I could do a more accurate job with the router than with saws and chisels, so I went that way. I spent quite a bit of time setting up and measuring to make sure it would work. I was still a little concerned right before I did the test fit with the neck, but it turned out that all the time spent setting up and measuring paid off. Who would have thought?

[Prostheta]

I hate feeling that doubt. It's a real killer. I have a lot of trouble reconciling the patience required to measure and re-measure continually vs. my inner need for tangible results. Screwing up through impatience soon makes that reconciliation happen though. :-D

[Juntunen Guitars]

I'm curious about spraying the inside and how that will turn out. I just got done reading online about it a little and it seems to be hit and miss on it working or making things worse. I've always thought that the finished interiors didn't sound that great but my experience is only one of the Alverez guitars that had a finished inside so I wonder how it affects the sound of the instrument over time. I don't really see a need for it because as long as you keep the instrument properly humidified it will be just fine. Acoustically though it would be fun to play around with, I'd like to try spraying just the back and sides and not the inside of the top becase spraying the back and sides would either dampen the sound or send the sound waves bouncing back and forth more depending on the finish used. Three coats seems like a lot though, what's the mill thickness?

This archtop is looking good by the way.