The Luthier: Part XI- Fretboard

Before attaching the neck to the body of the mandola, I needed to lay the frets. There are a few distinctions here I want to make before getting into the process. First, the slots for the frets were already cut in the ebony, as I do not know enough about guitar harmonics to do this accurately myself. Also, I do not have an accurate means of cutting those slots to the necessary width to accept the fret wire. Second, I left the face flat instead of crowning it slightly like is often done on fretboards. Were I to do this again, I would make the surface convex, as it makes for an easier time playing the instrument by putting more apparent distance between the strings when pressing on the strings between frets.

In order to fit the neck and fretboard to the body, the top needs first to be attached. Like with the back, I used the flat surface of my old garage to sand the sides plane.

Then, same as before, I clamped and weighted the top for the glue to dry, careful of avoiding the mistakes of the previous evolution that resulted in the crushed fibres that needed to be repaired. Distribution of pressure along the entire parameter is especially important, so I adjusted the weights and placement until there was even squeeze out along the entire joint.

After the glue dried, I used the coping saw to trim away the excess. This is where I began to encounter problems with the cedar. It being a particularly splintery wood, there was a considerable amount of tearout on the back side, which, because of the configuration needed to see where I was cutting, was the top face. Had I been thinking more clearly, I would have left a significant margin around the edge instead of following so closely to the sides. Once I noticed the problem, I corrected for it, but there was a bit of damage that needed to be carefully sanded away. Another problem with the cedar was its tendency to break on the corners. Specifically, on the lower point it broke off, which I patched, only to have it break again. The glue is far stronger then the natural wood fibres, so it was difficult to cut cleanly. In the end, there was a few small chips that I could not fix, but overall taught me a valuable lesson.

Now then, I am able to fit the fretboard to the body. The fretboard is longer by a fair bit than it needs to be, allowing for my shaping to whichever form I desire.

The first step in fitting it to the body is cutting the slot in the top to accept the neck. Most of the surrounding wood will later be routed away to make space for the fretboard, but I wanted to cut as close to the existing slot as possible to avoid complications.

Next, a slot needs to be cut for the truss rod. Underneath the top, the neck block extends nearly the exact same length of the overhanging truss rod, which is a happy coincidence. Without having it in the early stages of design, it was impossible to measure. Had the truss rod been longer than the neck block, there would be problems with tension being put in dangerous places, likely resulting in some sort of delamination or other catastrophic failure.

In order to align the necessary hole in the cedar, I used the other scavenged truss rod from a few months prior. Realistically, I needed only to find a single line that lying entirely within the existing slot in the neck block, then widening it with chisels to the previously defined shoulders.

Which is, more or less, precisely what I did.

With that channel for the truss rod, the neck slides cleanly into place. As shown above, the fretboard is considerably longer than it needs to be.

After a bit of design and deliberation, I decided on a basic compound curve. It is important to note that this absolutely needs to cover the end of the truss rod in its entirety, as there is a clear and obvious slot cut in the top that would otherwise sit exposed without the fretboard to cover it.

A bit of shaping later, and the fretboard can then be traced onto the body. Looking back, I probably would not do it this way again without some other design changes. For one, I had to remove a small bit of material from the rosette, which in a humidity swing caused it to crack. Second, getting the fretboard to inlet in the body perfectly was more trouble than it was worth. And finally, the most significant of the three, it lowered the fretboard closer to the body. I did not account enough for a downward angle of the neck, so (looking several months into the future from when I did this) the bridge causes the strings to sit a bit too high above the body end of the fretboard. You may be thinking that I could simply lower the hight of the bridge, but that is not the case. The way I designed for the tailpiece, it sits on the heel of the body rather than being anchored in the centre of the top. That causes the strings to be anchored higher than the surface of the top, so a certain height of the bridge is needed for sufficient tension across it to prevent the strings from rattling.

Now then, between a handful of chisels, the slot is routed. This would have been an excellent application for the router plane, but by this time it was already packed for the 3000 mile relocation west.

Returning to the beginning of this post, the neck is finally ready for its frets. Fret wire is exactly as it sounds, a wire in the shape of a fret. Due to its coming in a coil, it would have been a bit easier to lay on a slightly rounded neck, but as previously addressed, I left it flat.

Without any information or guidance on how to do this, the best method I arrived at was to cut the wire to length, press it in as firmly as I could with my fingers, and hammer down on a block of wood to set it in place (having ample support directly underneath the neck while doing this). A few things I learned in this process- One, the frets very easily deflect to one side if not careful to use direct downward blows from the hammer and alignment of the wooden block. Once it begins to twist or deflect, there is not much that can be done to fix it without taking it out and sometimes scrapping that bit of wire altogether. Also, ebony splinters very easily. In many places the vertical edge of the wire began to lift splinters as it was driven in. Fortunately, 99% of this was held down again by the inner corners of the fret as it fully seated.

After a few hours of cutting and hammering, the last of the frets were in place and the neck was ready for the upper nut. I could have put this in at any time I suppose. This nut, which holds the strings near the tuning pegs, is historically made of bone or other natural material like antler or horn, but mine is some synthetic material that cuts like a hard wood and carves like stone.

Grossly oversized, the nut protrudes a significant amount above the fretboard, which is fine until I am able to string it. I would much rather have to work it down a bit than realize too late that it is too short.

Before setting the neck in the body for the final time, I took a moment to file down the edges of the frets. The burrs from cutting and sanding were sharp in places, and the hard corners needed to have a nice chamfer.

Then, a little more sanding in those hard to reach places before the opportunity passes.

And finally, a hole needs to be drilled through the brace on the top so the truss rod can actually be adjusted. With the truss rod channel as a guide and a long brace auger, the task was fairly straightforward.

And there it is, all assembled. You can see why that hole is necessary with the wrench in place. By twisting and untwisting, the truss rod adjusts tension and subsequently its pressure on the neck, allowing it to bend up or down slightly to compensate changes in humidity and string tension. I believe that the neck is glued in place here, but I did not take any pictures of that. With the joint already so tight, it was simple and unexciting.

Next time, the body receives perfling and begins the finishing process.

The Luthier: Part X- Headboard, Fretboard, and Rosette

With the damage of the previous mishap fixed, it is time to begin the final stages of assembly. Before attaching the front, there are a few items that need attention. Namely, laying the fretboard and cutting the sound hole. At this time, I also drilled and cut the holes to accept the tuning pegs in the headboard.

To hold the tuning peg assemblies, I needed to drill a series of three holes and route a slot on the perpendicular face. After some deliberation, I chose a bit that was fairly close, but the cylinders of the tuning assembly are a strange size that perfectly matches any of my bits, whether for the brace, forstner, spade, or conventional bit sets.

Careful measuring of the spacing and alignment made me realize that there is virtually no margin for error in this operation. At the time when I ordered the hardware for this project, I tried to find pegs that were a top-down alignment instead of a side insertion like I have, but the options were limited and significantly more expensive. That look, in my opinion, is better suited to the style of the body, but I had to work with what I had available.

In a scrap 2x4, I made a series of test holes to gauge my layout and accuracy with the brace. With the angle gauge set perfectly vertical to the plane of the board, I followed it on two axes for whether I was drilling square.

Fortunately, the pegs fit tightly and flat against the edge. Because the holes are a hair larger than the cylinders, they do not need to be absolutely perfectly parallel to the wide face of the board.

Onto the real thing. This was largely stressful because the need to follow the line of the headboard is absolute. Less than a degree and a half tilted forward or backward and the hole would rip through the offending side.

In the end, the first side was a success. Shortly after, the second was finished to the same end. All along the headboard, I marked the relevant dimensions of the pegs. Namely, the start and end of the edge, the limits of the cylinders, and where the hole for the string is located against which I can centre the routed slot to be cut.

Due to the depth of the pegs, it was not possible to angle the headboard much beyond its rectangular form, another reason I regret being unable to find pegs of the other orientation.

A little shaping later, and the headboard is about as developed as it can be. During the drilling, I realized that the two sets were mirrored, and can only fit in the headboard in one orientation (based on which end is longer due to the worm gear placement). This was an important distinction to make, as it dictates how far down I could safely cut the top scallops. More than anything, being consistent is all that matters.

Next, I measured an approximate width for the slot and took an appropriate bit. Centred on the centreline for each peg, I drilled down entirely through the headboard.

As a good practise and to prevent splintering the very splintery sapele, when the threads of the bit poked through, I stopped, turned the neck over, and finished the hole from the other side. This way, the wings of the cutting face scribe the fibres and leave a clean hole on the back side.

With all six holes drilled, it is time to remove the web between them.

Via a coping saw, I cut away the web and created a rough rectangular/ovular hole that ran the length of the pegs on either side.

Rather than risk cutting in a place I should not have cut, I left a bit of material in the thickness and use a chisel to flatten the surface. Had I been paying closer attention, I would have realized that I came dangerously close on one side to cutting the outside edge too thin, and as a result it sits slightly asymmetric.

For now, the headboard is finished, and the peg assemblies will be returned to their box until final assembly much later on.

The next order of business is to attach the fretboard to the neck. This will ultimately enclose the truss rod and set the stage for being able to close the body with the top, although those two are for the time being, independent.

Removing the waste on the top end was simple enough. Here, it is important that the cut face is perfectly perpendicular to the face of the fretboard, as the upper nut will sit directly against it.

The initial placement shows that there is a gap between the nut and the headboard. Looking back, this is an obvious result of how the rosewood was attached to the headboard.

With the use of a rabbeting plane, the corner is made square and the gap closed.

In preparation for gluing on the fretboard, I placed a layer of packing tape over the top of the truss rod. This temporarily prevents glue from leaking into the channel and forming a layer between the metal and the wood.

A healthy layer of glue applied to the neck, and the tape comes off.

And then the fretboard is glued in place. The upper nut, although not currently glued in place, is used as a spacing guide for the fretboard placement. Using as many clamps as I could fit, and support on the underside of the neck to prevent it from bending, shows a thin line of squeeze out of glue all along the perimeter. Unlike most of the others, this joint is not merely aesthetic. A poorly mated surface here could mean failure of the neck under tension of the strings and the truss rod.

While the neck dries, I turned my attention to the last piece of this post, the sound hole and the rosette. When I began laying out the braces, I marked the centre of the sound hole. Because the cedar is so prone to splintering, I need to transcribe it to the front, from which side I will do the cutting.

After a bit of measuring, I have that same intersection marked on the front side. Since I will be inlaying the rosette, and I do not want to have the cut positioned such that it hits the top brace, I need to be certain that it is actually marked in the right place.

From the back, I drilled through the centre of the sound hole.

 Which aligned very closely with what I marked on the front. Only about ,5mm difference to the right and top, which is close enough for me.

On the back, I marked on the rosette where the four 'corners' intersected the drawn lines. From that, I was able to use the same positioning (mirrored to the other side of the rosette in case it was not perfectly round) to mark where the rosette should be placed on the top.

With two small drops of glue, I temporarily attach the rosette after the same fashion as when laying out the braces for the top and back internal skeleton.

Then, when the rosette would not move, I scribed the outside with a marking knife. It was incredibly important to do this cleanly,  as it sets up how clean the inlay will be. Deviating from the edge of the rosette means that mark will remain after the waste is cut away.

None of my saws have a deep enough throat to make this cut, so it needs to be done the long way. All around the interior perimeter I drilled closely spaced holes, then carved away the webbing.

Several hours later, I have a sound hole. In the beginning I was largely sceptical that I would be able to make a hole perfectly circular enough for the rosette to fit without a gap, and fully expected to leave it as shown above.

After a bit more patient fitting and adjustment, the rosette fit!

In a few places, there were small gaps. Small enough that they were not really a problem, but large enough that I could not ignore them. To rectify this, I took the ebony offcut from the fretboard and sanded it down, collecting the dust.

Mixed with the glue, it became the same colour as the rosette, allowing me to use it in a very minor space filling capacity.

The only problem with this was that cleanup was a problem, as it dries black. In the end, it worked out as well as can be expected.

Finally, the last brace needs to be cut. With the lower edge of the sound hole as a reference, I was able to cut the bit to length, ends angled to fit the cross brace. Using the same curve template as before, I marked and cut the ends, and glued this final internal piece in place.

That's it, for now. Next time will be joining the top with the body and installing the neck.