Gibson A4 Restoration

Here is a wonderful old mandolin that gave me a great deal of trouble recently. It’s a Gibson A4 from 1909, and it required extensive restoration work, along with an inventive solution to work around an original design flaw in the neck, to get it playing again.

It’s hard to see in this photo, but there’s a large split in the spruce top of the instrument. In fact, it appears that this mandolin has gotten so dried out that the center seam has completely come apart. I was able to fit a business card easily through this crack. This mandolin has a carved arched top, and that arch is what gives it its strength, allowing it to resist the downward pressure that the strings exert on the bridge. With a split this bad, running the entire length of the top, the structural integrity of this instrument is severely compromised.

On the peghead, there is some missing inlay that needs to be replaced. I found the missing letter 'b' from the script logo floating around inside the case, but the top section of the fleur de lis was nowhere to be found, and will need to be replaced entirely. The missing inlays, along with the crack that runs down the face of the peghead, is indicative of a far more serious problem.

 As you can see, there’s a matching crack on the back of the peghead, and in case you were wondering…

The split goes all the way through. In fact, this neck is split in half clear down to the dovetail. This is a problem that’s almost exclusive to early Gibson mandolins. More on this, and what causes it, later.

The fingerboard has sprung loose, a not uncommon problem on older instruments. I get the feeling that this mandolin spent awhile in someone’s hot dry attic, which likely caused the failing glue joints, as well as the center seam split. Had the fingerboard remained firmly glued, the split headstock likely wouldn’t have been able to continue splitting the rest of the way down the neck.

In addition to being loose, the fingerboard is badly split. When I ran my fingers along the edge of the fingerboard, I can feel bumps in the binding next to each fret. This is another sure sign that this instrument was dried out, which caused the fingerboard to shrink across the grain, leaving the fret ends to protrude and push the binding out slightly. In this photo, you can see that I’ve already removed the pickguard. Modern mandolins and archtop guitars have the pickguard attached with a wood screw to the edge of the fingerboard. However, on early Gibson mandolins, two skinny finishing nails were used instead. In order to  remove the pickguard, I carefully wedged a guitar pick between the edge of the guard and the fingerboard, and slowly levered away at it until I had pulled the guard out with the nails.

Here’s a better photo of the pickguard, so you can see what I’m talking about. This pickguard is made out of celluloid, a rather primitive, and incredibly unstable (not to mention flammable) plastic. It’s remarkable that this pickguard is intact. Celluloid has a tendency to shrink and degrade over the years. You’ll notice that I’ve prepared a piece of scrap wood in which to nail the mounting nails of the pickguard. I did this to prevent any further shrinking of the guard during the month that I’d have it off, so the nails would still line up with the mounting holes when the time came to put it back on.

I removed the tuners and tailpiece next. Again, I’m amazed that the beautiful inlayed celluloid buttons of the tuners are still intact.

Next, the fingerboard is removed. Normally, I apply heat to the board with a small tacking iron, while slowly working the fingerboard loose with a thin artist’s spatula. I had hoped to get a few photos of this process, but in this instance, the fingerboard literally fell off by itself before I could apply any heat. As you can see, the split in the fingerboard is quite bad, and it goes all the way through. I ended up gluing the board back together separately, and then filling the cracks with a mixture of wood glue and ebony dust.

Here’s a better shot of the split neck from the inside.

The split runs right along the center stringer, all the way to the dovetail. I initially thought that I’d have to reset the neck joint, but the joint seems to have held just fine.

 Here’s a better shot of how nasty the fingerboard split is.

Next, I set the instrument aside for about a week and a half to humidify. As I previously stated, most of the damage to this instrument was caused by dry air. In order to repair the cracked top, I first must introduce moisture to the wood through the use of a humidifier. At Five Star, we sell case humidifiers to help maintain the proper 45 – 50% relative humidity that your instrument needs to survive. However, this instrument has been so neglected, that a simple humidifier wouldn’t cut it. When I’m dealing with an instrument that’s this dry, I usually hang the instrument from a wall hanger, and then bag the body up in a trash bag with a moist sponge hanging in the bottom. I tie the bag tight at the neck, and leave it for a week. After humidifying, the crack in the top of this mando, which previously was large enough to fit a business card through, had swelled up tight. Now it’s time to glue.

I decided to tackle the split neck first. I thought that I’d leave the center seam glue up till later, since I expected it to give me more trouble than the seemingly straight forward split neck. I couldn’t have been more wrong.

The glue up of the neck actually went quite well initially. I used hot hide glue, which I try to use on all vintage instruments. Since the fingerboard was off, I was able to easily work glue into the crack from both sides. I clamped the neck with two clamps on the headstock, as shown in the photo, along with an additional two clamps on the neck, which aren’t shown above. Since there’s no glue squeeze out visible in this photo, this must have been a dry run. Hide glue is heated to 140 degrees, and applied hot, and the working time can be as short as fifteen to twenty seconds, so I often forget to take photos when I’m rushing to get everything clamped up in time.

The next day, I went and tackled the center seam split. Cracks in spruce instrument tops are most frequently caused by drying out, so they can usually be closed up with humidification. However, some cracks that have been open and dried out for a long time will never close back up. When that is the case, a spruce splint is shaped to fill the crack, and a finish touch up is performed to make it disappear. I lucked out in this case, as the crack closed up on its own, but there could still be other complications. The longer a crack is open, the more sweat and grime and contaminants are allowed to work their way in there, making a proper fix even more difficult.

I used hide glue again, to rejoin the center seam. I used the lower jaw of the smaller clamp in the photo above (the one in the soundhole) to push up from the inside, spreading the gap open, while I worked glue into the joint. I then lightly clamped with two C clamps, one through the sound hole, and one over the top of the body. Normally, you should never clamp an instrument over the top of the body as I am above. If one were to over tighten that clamp, they could easily damage the top or back very badly. I’m not using these two clamps to force the crack together, however. I’ve only tightened the clamps up enough to ensure that the two mating surfaces stay lined up. No clamping pressure was required to hold the joint together while it glued; humidification caused it to tighten up quite a bit, and the added moisture from the glue caused it to completely swell shut.

I previously mentioned the short working time of hide glue.  Above, you can see the space heater that I’m using to keep the area around my bench hot, in order to give myself a little extra time to get the clamps set. In some cases I’ll even heat up the surfaces that are to be glued, either with a heat gun, or, in the case of an acoustic guitar bridge, a quick fifteen seconds in the microwave.

Next up, I reglued the fingerboard. In this case, I clamped the board up dry first, using the four clamps shown above. This way, I could be sure that everything was perfectly lined up. Then, I removed the two black clamps towards the top of the neck, worked some glue under the board as far down as I could, and clamped. I waited a few hours, and then removed the third and fourth clamps, and glued the second half of the board.

Normally, when I have to remove a fingerboard, I pull the first and tenth frets from the board first. Then, I drill a tiny hole through each fret slot into the neck, and drive a skinny ½ x 20 wire gauge nail through each hole and into the neck. I then clip the heads off of the nails, leaving them sticking up about 1/16^th – 1/8^th of an inch above the board. I then remove the fingerboard, leaving the nails behind as indexing pins. This helps me line the fingerboard up perfectly when the time comes to glue it back on. It also prevents the board from slipping when I clamp. When I’m done, there’s enough nail sticking up for me to pull it out with a pair of needle nose pliers, and the two holes are easily covered up by the bead of the fret wire. Well, usually. In this case, the fret wire used on this mandolin is nowhere near large enough for me to get away with this approach, so I was forced instead to use the two step process described above.

Here’s a close up showing just how small the fret wire on this mandolin actually is. The second fingerboard that you can see in the background belongs to a ukulele I’m currently building. I used the smallest fretwire that LMI offers on that uke, but the vintage Gibson wire is far smaller. As far as I’m aware, no one is manufacturing fret wire in the sizes that Gibson used on prewar mandolins and banjos. I’m hoping a refret won’t be necessary.

With the structural repairs complete, I decide to reattach the hardware, and string the instrument to tension, to make sure everything holds up, before proceeding with inlay and cosmetic touch ups. It’s a good thing that I did.

 Here’s the neck on the night that I strung it up. The damage to the ebony peghead back strap is still visible, as I haven’t done any touch up work, but everything is holding up fine.

Here’s a photo I took the next morning. As you can faintly see, another crack has formed. I removed the strings, glued it up again, and left it for a few days. Then, I strung it back up to see if it would hold. This time, it lasted about four days. I noticed that it wasn’t my glue joints that were failing. Rather, the wood itself in the middle of the neck seemed to be rotting and coming apart. I did some research, and discovered that this is a not uncommon problem.

This mandolin has a three piece neck, made up of two halfs of mahogany joined with a strip of darker wood in the middle, commonly referred to as a stringer. This stringer looks like ebony, but is actually a strip of pear wood that has been ebonized through some archaic staining process that is thankfully lost to history. This staining process significantly weakens the wood. Gibson was aware of this problem pretty early on, and they soon switched to using solid mahogany necks with a decorative inlayed strip of pearwood, rather than using one that ran through and through. Upon further inspection, the center strip in the peghead of this mandolin seems to have turned to mulch in some areas, so conventional gluing alone in the manner that I’d been attempting wouldn’t cut it. Hide glue makes for very strong joints, but only when the mating surfaces fit tightly, as it has terrible gap filling properties on its own. Aliphatic and polyvinyl acetate wood glues aren’t much better. Epoxy is the best bet for gap filling, but I hated the idea of using an epoxy on a vintage instrument like this. Besides, unless I was completely impregnating the center strip with epoxy, I knew this still wouldn’t cut it. And replacing the pearwood strip completely would be entirely out of the question; such a job would be so invasive and noticeable that it would severely devalue the instrument. I started brain storming. I could dowel the head stock, from the bass side to the treble side, but that would look hideous. It would be impossible to touch up the exposed end grain of the dowel well enough; it would stick out like a sore thumb. I had earlier considered inlaying a reinforcing spline in the neck under the fingerboard, but I decided that any strength that this would provide would be negligible, as the fingerboard glue joint itself provides a great deal of strength. This was apparent when the head stock split the second and third time;  since the fingerboard was now glued firmly in place, unlike before, the pearwood stringer was only able to split down to the fingerboard.

I also considered inlaying a spline in the back of the peghead, but decided against it. Installing a spline underneath the missing piece of inlay that I'd eventually have to replace sounded like a pretty cool idea, but I knew it wouldn't be wide enough to make a difference. Removing the ebony peghead overlay or back strap to inlay a spline underneath was out of the question; both the overlay and back strap were so thin that I’d likely have destroyed them in an attempt to remove them. I ended up getting a clever idea for a relatively seamless, but somewhat invasive fix, from luthier Frank Ford.

First, I carefully examined the wood grain to determine in which direction the grain ran out of the edges of the peghead.

Then, I removed the upper A string tuner ferrule, inserted a tapered violin peg, and carefully tapped it into the hole with a mallet.

I used the tapered peg like a wedge to split a piece of the headstock cleanly off.

Now, I can drill from the exposed tuner hole across to the opposite D string hole, and dowel through the weakened pear wood, without anyone being able to see it. I’m confident that I’ll be able to glue the broken piece of the peghead back on, and make it disappear with a simple finish touch up.

Here, I’ve got the mandolin clamped up in my neck vise, ready for drilling. I have a Jorgensen cabinetmakers clamp with padded jaws mounted to the side of my bench for use as a neck rest; it has the advantage that I can clamp instruments up on their sides like this, if need be.

First, I drill my pilot hole. I think I’m using a 3/32” bit here. I drill slowly and carefully to ensure that I stay centered.

Here’s the finished hole, with a guitar string stuck through it, to show where the bit came out. Not bad.

Next, I followed the pilot hole with a ¼” bit. I determined that I had about 1/16^” of wiggle room on either side. If I fell outside of that narrow window, I’d end up poking out the Gibson inlay from the inside, or drilling clear through the back of the headstock.

Made it through cleanly without any trouble.

Next, I prepared a length of ¼” dowel.

And here it is, glued in place. I waited fifteen minutes, and then removed the squeeze out. Here I’m using Titebond instead of hide glue.

I trimmed the excess with a knife. I know this isn’t the best photo, but take a look at how thin the ebony overlay and back strap are. Before splitting the headstock with the violin peg, I was a little worried that the ebony would not split cleanly with the mahogany. However, I had already seen this peghead split three times down the middle now, so I knew with a great deal of confidence that the ebony was thin and brittle enough to split wherever the mahogany did.

 Next up, I reglued the severed piece of peghead. I’m using the blue Irwin clamp to apply pressure, while the cast iron C clamp with the wax paper padded cauls keeps the two pieces lined up.

 Lastly, a piece of pearl is selected to match the yellowed patina of the original inlay. I then cut the pearl to fit the existing cavity. Normally, when doing inlay on a new instrument, I’ll glue the pearl in, and then sand or scrape it level, however, this piece of pearl is only yellowed on the very surface; if I were to sand at all, it would end up a bright pearly blue color, nowhere near how the rest of the pearl looks. I ended up sanding the back of the inlay until it dropped into place flush. Then, I glued it in with cyanoacrylate, and filled the gaps with black stick shellac.

  Finally, I used a combination of stick shellac and French polish to touch up the damage to the headstock where the pearwood split tore into the ebony. I also touched up and polished the edges of the peghead where the split that I made was most visible. I made sure to do just enough touch up to hide the repairs, while not overdoing it, as a perfectly glossy smooth peghead on a 105 year old instrument would look out of place.

Finally, I strung the instrument back up. Surprisingly, I didn’t really have to do much in the way of setup work to make this instrument play well. Despite the condition of the frets, this mandolin plays fine all up and down the neck, with no warping, despite the lack of a truss rod, and no buzzing in the higher registers. 

The crack from where I split the headstock is visible, if you know to look for it, but it doesn't stand out, and it’s far less noticeable than a spline or dowel would be.