Guitar Build #2 - Dreadnought

As long as the first one worked out, I was always thinking that the second one would be for my son. He’s interested in a Dreadnought so once again we head to the Martin archives.

This is the iconic design that people typically associate with acoustic guitars. Strangely enough, Martin originally developed it in 1916 as a private label design for the Ditson company. 33 were made for Ditson between 1916 and 1930. When Ditson was sold in 1930, the orders dried up until former Ditson dealers started asking Martin about the design. So in 1931 Martin started to build it sparingly under the Martin name, but they didn’t think it was worthy of their catalog until 1934…..and the rest is history.

I still haven’t quite figured out Martin’s naming conventions, but this will be modeled after a pre-war “D” (for Dreadnought). The numbers following the D seem to indicate the details of the build, as in materials and decorations. This will be closer to the D-18 than the D-28, with the primary difference being the back and sides of east Indian rosewood instead of mahogany. Bracing will be pre-war as well, actually pre-1938, with some fun details and Martin lore we’ll get into as we go.

The one at the right is available for $50,000. Mine will be a little more affordable and have a nicer finish, it’s a win-win.

I’ll be using materials in this one similar to my OM: Sitka top, east Indian rosewood back and sides, mahogany neck, blocks and kerfing, ebony fretboard and binding, and not sure on the peg head yet. The top has bear claw figure, which is hard to see in the photo below but will show up later.

Plans & Molds

John Hall at Blues Creek Guitars says that the Stew-Mac “Herringbone Guitar” plan is accurate, so after that arrives I start in on the body forms. It’s a bit easier this time around, I have a better idea of what’s around the corner.

Sides

Now that the molds are done the next step is bending the sides. The first stage of this process is getting them down to bending thickness, which I do with my handy-dandy garage sale thickness sander. After this we dust off the Bend-O-Matic, review the process and prepare to bend.

Martin Lore, Part D-One

In which we interupt the regarlarly scheduled programming to go down a rabbit hole

One of the facets about this process that I find fascinating is the history and the reverence for the craftsmanship of the past. As you get into the pre-war Martin mystique there are lots of rabbit holes to go down, and each one has its own blend of hard evidence, clues, and red herrings, so it really does make for a fun investigation.

It helps to remember that Martin was a small family business trying to survive in the world of ever-changing musical tastes. They were evolving on the fly so model changes and methods of construction seem to have “morphed” in a lot of cases versus the distinct model changes that we’re accustomed to now. This was also a world of individual craftspeople and not computerized production lines, so there are interesting variations due to human elements and the economic climate which sometimes influenced material availability.

As I was researching this build, one of the things I tripped over was a discussion of reinforcing strips; these are strips glued across the inside of the sides to discourage cracking. The plans on my OM called for cloth or thin wood strips, I opted for thin wood strips because I liked the look and thought they would have better strength in tension if ever called into action. And I glued them between the kerfing strips, not under them…because the plan didn’t indicate that they went under the kerfing, nor have any of the videos I’ve watched. And the Dreadnought plan is the same way. So when I came across a discussion about running them under the kerfing it caught my attention, the argument being that if you use a butt joint between the reinforcing strip and the kerfing, it allows for a “stress riser”, i.e. a spot that would be more likely to crack. This jibed with a remark from Benedetto’s archtop book in which he feels that in order to reinforce the whole side as intended, the strips need to run the entire width of the side. Sounds logical. Probably totally inconsequential in any detectable way but logical.

But the more probable explanation is that it was more efficient for Martin to do it this way.

The other aspect of this discussion was the material used for these strips. It seems that there was some inconsistency over the years, but that generally linen was used in the early years, giving way to other available ribbon materials such as grosgrain and polyester as time went on.

I liked the linen idea so off I went in search of 5/8″ wide linen strips, ribbon, whatever they might be called. Things on Amazon tended toward craft applications so they were colorful with frayed edges. It was also proving difficult to find material that was 100% linen, there was usually a small percentage of polyester. After more digging I was able to find what I was looking for at a bookbinding supply house, in the guise of bookbinding tapes: 5/8″ wide, nicely bound, natural color and 100% linen. Inexpensive and quick delivery, was very excited when I got them, they have a great vibe.

Installation was the next step to figure out. These are glued on, so they would need to be soaked in glue and then put in place while minimizing the mess. Thinking this over I decided to try masking off the areas, soaked a strip in hot hide glue and put in place. I was able to place it and then kind of rub it into place, squeezing the glue through the linen and into the joint, and this worked much better than I had expected.

Kerfing

Installing the kerfing should not be a big deal, but hot hide glue makes it exciting and I have another idea that may make it an adventure.

The linen strips are about 1/64″ thick, so if I glue the kerfing over it I’m going to have a small gap on either side of the linen unless I create a dado in the kerfing. Most of this discussion around the reinforcing strips, and all guitar construction actually, is about joint integrity. The idea being that solid joints doing a better job at transmitting sound than joints with gaps. Again, sounds logical.

So to eliminate this miniscule gap, I need to create four 1/64″ x 5/8″ channels in each kerfing strip. Which doesn’t sound so bad until I try to do it. To start with, the kerfing strips are fairly fragile due to all the slots cut in them. Not so bad, but then they get wetted down and pre-bent to conform to the shape of the body. And the only way to mark where the dadoes need to go is if I can fit them to the body. So I tried one and thought, “this is ridiculous”.

But here at the Long Spoon Guitar Company we relish a challenge, however ridiculous it might be. So after a few minutes I had the idea of using the bending mold as a support, and then realized that with some maneuvering I could also get a clamp on it, so now I had a fighting chance. A little Zona saw plus sharp chisel and voila! Reinforcing strip dado. C’est magnifique!

Fitting these wasn’t actually that bad after I got the system down, but I did sacrifice one to the cause.

So after cutting 16 of these notches its time to put them in. Heat the glue and get the clothespins!

Okay, so this wasn’t a totally smooth process. One or two broken kerfing pieces, one aborted installation attempt, and few other minor mishaps. I will make some adjustments to this process moving forward, including an upgrade to the ‘clothespin with rubber band’ clamps. But eventually I got the kerfing in…

The back

First step on the back is to get the pieces down to around 3mm thick. They start around 4.5 mm and the sander moves pretty slowly and doesn’t take big cuts, so this takes a few sessions. But we get there.

Now I need to get rid of that gap between the two pieces so we have a good joint for the glue-up. And how do we make that joint? Glad you asked….

Tool Story - Stanley #6

Another tool rescue! This one popped up locally and I was lucky enough to be the first one to see it. It’s a Stanley-Bailey #6, a bit of a tweener plane but it has the low front knob that indicates an early date, and was $15 so I just grabbed it without asking a lot of questions. The #5 is sort of the all-purpose plane, and the #7 is the standard jointer plane, so the #6 is not as well known, but I was looking for something a little longer for shooting the back and top joints so thought I would give this one a shot. It turns out to be a Type 11 so dates between 1910 and 1918, pretty cool.

These are fun to clean up. There was pretty good surface rust on everything, but fortunately nothing serious on the body or the blade.

Some time with the wire wheel and things start to look a lot better.

Now we get to the tote (rear handle). I don’t remember the one on the #5 being a big deal but this one was a bit of a bear, because I couldn’t find the router bit I had used. The trick to these is that you need a 1/2″ cove flush cut router bit without the bearing. Not sure that they make them that way any more but I happen to have one. But before that we need a blank.

There was so much of the japanning missing on the body that I toyed with the idea of doing a full bore restoration on this one – meaning new blade and cap iron, stripping and painting the body and frog, etc….but after cleaning the body I really liked the look of the bare cast iron, so I decided to hit it with paste wax and I liked that even more. And since the objective was a tool to shoot edges with and not a museum piece, I waxed practical moved forward.

Following the waxing and shellacking I turned to sharpening the iron. There’s not a lot of blade left and as I started to flatten the back it became apparent that someone had used it as a pry bar at some point. I was able to get some of the lumps out and then get most of the back flat via diamond stones. As much as people like newer blades I’ve always found that these original Stanley blades take a nice edge.

After sharpening the blade I ran the sole and sides over some abrasive plates and got things looking fairly flat and shiny. Then assembly and we’re done!

Shooting the back Joint

Now we can finally get to joining the back, sheesh!

So, before the plane interlude we were looking at closing the gap between the two plates, below:

This is the basic “shooting plane” setup. The idea is to plane the edges perfectly straight so that you get an invisible joint.

I was able to get very close with the plane but still had some convexity, so I rigged up a long, square sanding block and in a few passes was able to get it.

The test is holding it up to the light, looks good so far.

And now for something Completely different...

At this juncture, in a fit of enthusiasm, it seemed like the right time to run power to the shop area of the basement. I did the original framing and wiring in the basement so in my head this was about a 15 minute job…naturally it consumed 3 weekends and many weeknights. But I was tired of running (and tripping over) extension cords so, no time like the present! Pro Tip: If you ever finish a basement, don’t finish the ceiling. Lots of goodies up there that you might need to find later.

Okay, we’re up and running. Now back to our regularly scheduled program….

Shooting the back Joint...Again

You might remember that I had fitted the back joint and was ready to glue it up…but when I checked the joint it was no longer seamless… So apparently 4 weeks is a long time for two pieces of wood to remain in perfect harmony. Not sure what happened, they seemed to be such a good match! Maybe the mahogany started looking good, maybe they just got bored, who knows, but they moved apart! And so, I had to sit them down and remind them that they need to stay together, think of the sides, think of the top – they need stability!

So I re-fitted the joint. Nothing wrong with some extra practice.

Gluing The Back

Now that we’ve got these two crazy kids back together its time to seal the deal! I’m using the tape technique again, this time with measurements and duct tape which is a pretty significant process improvement.

So there we have it, about 2.5 mm give or take

At this point you need to figure out the orientation of the back, usually pretty self-evident due to the grain pattern.

Also inside vs. outside. The “O” (for outside) in the corners reminds me what’s what.

One of the challenges moving forward from this point is locating the joint, which is the center line of the back. If you’ve done a good job it will be very difficult to find, and you need to know where this is so you can line things up for assembly. Before cutting the shape out, the slight misalignment of the edges of the boards shows you where the seam is, but once it’s cut out that reference is gone. The trusty pencil line is not the greatest solution because it gets covered by the center reinforcing strip on the inside and sanded on the outside, so I was always fighting to find it or re-establish it on my OM. After some cogitation I decided to drill a small hole through the joint just outside the pattern on both ends, which is proving to be a good process improvement.

The reinforcement strip is next. Cross grain sitka cut from the waste from top wood. It starts out about 3/16″ thick and is planed down after gluing to about 1/16″ and then rounded over by taking the edges down to about 1/32″.

Below you can see the wonderful variety of nature. This brace material is hand split so that it follows the natural grain direction of the wood. Even so, on the left you can see how its not quite straight, and on the right how its not quite perpendicular (on the quarter). So I adjust, if possible, where I can.

The next step is sanding the braces to the radius of the back on the radius dish. So I mark off the approximate brace locations on the disc and whack away. The lower braces are tricky to hold while sanding, so after I sand a few fingertips off I come up with a better idea and attach them to a holder with double-sided tape. In addition to saving my nails, this provides the added benefits of even pressure across the whole brace and gives you a way to keep the brace perpendicular while sanding, both of which should help the fit.

Now we get to shaping the braces. At this point they’re to thickness but oversized in height.

Interesting to be doing this a second time. There is a lot of mystery and lore around this process, more so for the top than the back, but similar things are happening. As you shape the braces you hold and tap the back in order to “listen” to it and get an idea of how it sounds. On my OM the back was pretty stiff at this stage, meaning it wasn’t very resonant; as you remove wood this improves and then you have to decide where to stop. This back is sounding better at this stage than the OM did…of course the Dred is larger but not by very much so I’m finding this interesting.

Doing final smoothing on braces is tricky because you want to avoid damaging anything in the vicinity. As I was navigating this I decided to makes some simple sanding forms and was very happy to have their assistance. Below is the assortment for the top radius of braces 3 and 4 and the end tapers.

to the Top

With the back and rims under control, we move our attention to the top. For this build I’m using a Sitka Spruce top with “bearclaw” figure. Bearclaw spruce is somewhat controversial in that it was originally rejected for sound boards because the unusual grain pattern was perceived to be a defect. Over time some people now believe that the figure adds to the stiffness of the board, making it superior to straight grain. Another wonderful debate with no definitive answer! But it IS visually striking which I happen to really like It will be tricky to see the figure before it gets finished but I’ll try to show it in the pics.

The pieces are book matched at the mill, they use the red crayon “V” to keep things organized. I’m still fairly new to this so I still get blown away by the general appearance of the wood, but after that passes you have to take a really close look at things so that you can spot any issues that you might have to deal with. Blemishes, pitch pockets or just odd grain are things to look for and will help determine which side is out and which is in.

After the basic orientation has been determined and I’m feeling good about the top its time to join the two pieces. This step is about getting the best joint that you can, and you know its good when it disappears. This is a hand operation with a shooting board and a long plane. It can take some patience but its not bad, and a little fine tuning with a long sanding block built for the purpose can work wonders.

After I’ve got a good joint its time to glue the two pieces. I tried the tape method but wasn’t happy with the result. As it turned out I tried this 3 or 4 times because I wasn’t liking the joint, but in the end figured out that what I thought was a gap was actually an optical illusion provided by the grain. The challenge with the tape method is that it relies on the spring in the tape to provide the clamping pressure, which works but feels a little odd when you’ve been applying serious clamping pressure to glue joints all your life, so I used I different method where I could bring some wedges into play. This worked well but I’ll eventually create a jig for this, its kind of ridiculous to have to fiddle around with it so much!

Now that the top is together it’s time to figure out the inside and outside, top and bottom. Using a clear pattern showing the location of the soundhole and bridge is very helpful at this point. As you can see below this top has some great figure and also a “wave” that runs across it. After playing around with the location of the pattern I decided to place the bridge over it, partially to hide it and partially because something tells me this might be a good spot for the bridge because the grain will be tight and should provide good energy transfer…maybe…? Also it kind of looks like a sound wave so what better spot for the bridge? Case closed.

With the layout set, it’s time to do the rosette

On the OM I learned some things about routing the rosette channels, one of which was that my used Bosch Colt needed new bearings. I was also not very excited about the fine depth adjustment on it, so I asked my Luthier friend Diane about it and she recommended the Ridgid palm router. She’s one of those people that keeps trying things until they find what they’re after, so I take her recommendations seriously, and Santa did too.

A new router means a new adapter plate, but fortunately the bolt pattern on this one fits inside the outlines of the jig, so I just had to drill new holes for it instead of making a separate plate.

The layout on this one was a little tricky. The sound hole is 4″ in diameter vs. 3-7/8″ on the OM, but the shell is the same inside diameter of 125mm and I’m using 5mm width instead of the 3mm on the OM. So there is less room between the shell and the sound hole which makes things a bit tighter, and then of course you need to be sure that the fretboard covers the joint of the inside ring. Below you can see my practice cuts.

Okay. After several deep breaths and centering exercises and we’re ready to go.

Here’s the setup. The jig rotates on the 3/16″ guide pin, so we drill a 3/16″ hole in the top and also the support board beneath, with a drill press so everything is perpindicular. The router is oriented so that in this position I can see the cutter – this makes adjusting much easier, lesson learned from last time. Its sitting in the area of the rosette that will be covered by the end of the fret board, so this is where you start the cuts and check things before taking the full plunge.

There is a brief interlude here in which I miscalculate the amount of purfling needed for this step and need to re-order. This is a rookie mistake – the issue being that I’ve cut these channels for the stock that I have, and its highly likely that the new stock I get will be slightly different in size…and it is. Not by much, and fortunately its a bit large, but it still requires some finessing to get the new stuff to fit.

As I start to fit the shell I wonder if there are patterns or groupings that might help in the process. This shell is multi colored and while each piece is different, there are some patterns and colors that emerge. I decide to put the more blue pieces at the bottom of the circle and graduate towards green and then yellow-green at the fret board. Not sure if it looks better but it seems worthwhile to put some additional thought into how to use this amazing material.