As with the neck resetting portion of this article, I'm not going to present each and every step. For that kind of detail, you can turn to any of the several articles on refretting listed in the Big Index Page.
I will illustrate some of the job, though - mostly the stuff that's unique to this guitar.
I always use a fair amout of heat when removing frets. The heat of this 80-watt Weller soldering iron is usually sufficient to get the frets hot enough that they sear their way out of the slots without causing chips in the surface of the fingerboard. As each fret got hot enough, I pulled it out with my flush cutting end nipper:
The frets came out nicely, even the first six which had been set in with hide glue:
I'd say that in addition to the entire fingerboard having been refretted at one time, the first six frets were subsequently replaced by a different luthier.
My first job was to clear out all that glue, and clean up all the other fret slots - a fairly simple task requiring a nice sharp, think kerf saw - this is the one we use at Gryphon for only this job:
Now it seemed obvious that the person who replaced the first six frets did not use sufficient heat to remove the frets, because unlike all the other slots from which I pulled frets, these slots all had chips and filled divots:
A very light preliminary sanding outlined those difficulties nicely:
I set about gluing back the old chips and filling the divots, first scarring the shallow craters with a thin blade to improve adhesion:
Then a bit of cyanoacrylate and lampblack made a nice filler for the little chips:
While those old chips filled nicely, a bit of leveling with 220 grit sandpaper revealed that the fifth fret inlay had a problem - a little void started to appear at one point of the diamond:
Looks as though the inlay had a small defect in its bottom, and when it was leveled, the defect started to show through to the top.
I cut a new diamond from the same kind of pale abalone shell used in the original inlays. It's important to make the cuts with the same size blade used originally so the inlay would look as though it had not been replaced. Luckily, they were originally hand curt and a bit irregular, so it made sense to cut the new inlay the same way:
It took a stout blade to make those cuts, and, because I do this job from time to time, I have a little fixture I made up to hold the diamond and square pieces for Martin fingerboard inlays. It's a simple block of aluminum with little recesses that just fit the diamond or square shapes:
I used a small chisel to chop out the old inlay and prepare to insert the new one, which I pressed in place with more of the cyanoacrylate/lampblack mixure:
Now, that's more like it:
The inlay complete, I sanded a bit more with 220 grit and my long sanding block to level the entire fingerboard preparatory to installing new frets.
Without exception, I use this little plastic faced mallet to install frets:
It's the one that's typically sold with one brass head and one plastic. I find the brass adds too much weight, so I replaced it with another, identical plastic head. That gives me the same feel and balance I used to get with my old Stanley plastic hammer, now long out of production.
When I first signed on to do warranty service for Martin about 35 years ago, I spent quite a bit of time staring at the fret ends on new Martin guitars. I was looking at the file marks made when they rounded over the ends so that I could simulate the same treatment with my fret work. Times have changed of course, and the Factory now uses a different method. I continue trying to make them look like the vintage instruments, being consistent with the angles at which I hold my files:
After dozens of identical operations and a bit of fret leveling and polishing, the fret job was complete, and this old critter is starting to look like a guitar again:
Speaking of which, you may notice that the tuners are in place. Before refretting I installed a set of new Waverly 4060 tuners - the direct replica of the Grover "G-98" tuners that would have been original on this instrument. While the value of the guitar is such that it would be reasonable to hunt up and pay for an original set, these new Waverly gears work so much more smootly and positively that the owner of the D-28 chose to use them for the sake of playability.
I used the little press I'd made on Day 12 to insert the bushings:
The bushings are a reasonably tight press fit, and this little press makes it possible to have really good control to avoid injuring the finish by pressing too hard. The rest of the tuner installation was a simple matter of screwing the plates to the back of the peghead, so I didn't shoot any pictures of that.
Down to the final stages - time to finish off all the details that have to do with playability and action.
I started with the bridge. Once the neck was in place, I calculated and measured the saddle location for best intonation, and set up my bridge routing mill to cut the saddle slot:
Here's a better view of that rig:
This machine runs a Bosch laminate trimmer on friction-free linear bearings in all dimensions. The fore-and-aft slide is locked for saddle use, of course. I recently modified the vertical slide mounting so it can tilt backward at any angle from zero to 20 degrees to accommodate the growing use of tilted saddle slots. Even though I think the backward tilted saddle is a more structurally sound design, I still follow the maker's original style, particularly when working on vintage instruments.
I reamed the holes to fit the original bridge pins, and, using this high-tech device, I countersunk them with a nice chamfer:
The original bridge pins are solid, with no slots, so I slotted the bridge to accommodate the strings in the holes, notching for each string separately by trial fitting as I went. My tool of choice is a jigsaw blade I mounted in an old Shaker peg handle:
Over at the other end of the neck, I started to clean out the recess where the nut fits, only to find it had way more stuff glued in there than I first noticed:
It was only after I dug out a small mountain of paper and glue that I noticed that the nut slot had been cut about 1/16" below the level of the fingerboard. Rather than making a new nut that would look oversize and call attention to this unfortunate modification, I decided to make a mahogany filler so the nut could bed down normally:
As you can see, I ran the grain direction the same as the neck, so the filler would present well when viewed from the edge. I used a guitar capo to clamp the shim with a cork pad on top to distribute the pressure evenly. Enough for today - gotta wait for glue to dry.
Maybe I should tell Rick Shubb to think about another potential market for his capos. . .
Day 20: Setup and Final Touches
Rummaging around my collection of ancient ivory scraps, I found one that would do nicely for the new nut. First I ftted the bottom of the piece to the recess at the end of the fingerboard:
Then I roughed out the shape to nearly the final dimension and contour, and glued it lightly in place:
I then slotted for the strings and adjusted the action at the nut before removing the nut to polish it nicely before reinstalling it. Most all of the final setup was pretty routine, trial fitting, setting action, and all that, and I didn't photograph that stuff in detail.
Here's some interesting stuff about setting up the nut I thought was worth some discussion. It's a system for porportional spacing of the strings I learned from that eminent guitar maker, Julius Borges.
The idea is to have equal spaces between the strings rather than spacing them evenly, on centers.
First, I set the outer two strings in position, notched the nut and set the action for both of them. Once the first and sixth string positions were fixed, I measured the distance between them:
A simple bit of calculation gave me the spacing I'd need to establish between each of the other strings:
Taking the total of the diameters of the inner four strings and subtracting it from the space between the outer strings, I found the amount of space I'd need to divide up between them all. Five spaces between six strings gave me a figure of .2674. Naturally, the final digit is way beyond any level of accuracy needed here, but I thought I'd include it just for fun. My caliper only resolves to the nearest .0005, and I never count on it for greater than .001 accuracy.
This is a set of gage blocks ( or "Jo-blocks," named after C.E.Johansson, a swedish engineer who devised this method of precision measurement about 100 years ago) and it contains just the right assortment of precision spacers necessary to generate stacks for measuring from about a tenth of an inch to several inches with accuracy to one ten-thousandth of an inch:
So, I got out the appropriate ones, and made up the necessary stack:
Now, with the first string in place, I installed the second string, spacing it over by simply placing the stack of blocks between the two:
I marked the both edges of the string with a sharp blade, and then rubbed a bit of pencil graphite into the scratches:
That gave me the perfect location for the second string. Once it was in place, I located the third the same way, and finished up the job by repeating the process starting from the sixth string and working inward.
Here's the final nut, before I removed and polished it:
Back at the bridge, I added a little tradiditional detail to the tops of the string notches:
And, I made up a new saddle, trila fitting it as I went. Here, I'm using a little hand-crank drum sander to shape the curved ends of the saddle:
Oh, and here's another little old time traditional touch:
When I installed the new Waverly tuners, I lined up all the mounting screws - just for looks.
Well, that's about it - the guitar is back to its old playability, and ready for some tunes.
Next up - a 1940 D-28 with "issues". . .