Friday, March 29, 2013

Heel Block - Body Joint Repaired

I glued some mahogany pieces (about 1 inch square, with an angle cut out to make room for the kerfed lining) to the side of the mahogany heel block and to the top. I was pretty sure that I hadn't been able to get enough glue into the space where the heel block had pulled free from the sides, so I thought this extra reinforcement might do the job.

And it worked! Several days later, the strings are up to tension (tuned down a note to D) and the repair is holding.

I will post some pictures of the interior bracing; the area in front of the sound hole is now much more heavily braced, while the top behind the sound hole is still the lightly braced Gibson design. I'll also post a recording so you can hear what it sounds like.

Monday, March 25, 2013

The Mostly Repaired Guitar

The mostly repaired guitar. If you look close at the left side of the sound hole you can see how the middle section of the top sticks out towards the bridge a little past the side section - look at the purfling stripes. (You can see it better in the bigger version of this picture at the top of the blog.)

Repairs

To summarize everything so far: by trying to fix a 12 string I am learning the great importance and specific function of the bracing and design of the guitar in front of the sound hole. I never understood why Martin puts that big fat brace under the fret board - now I know. The extra tension of the 12 strings makes structural problems very obvious.

I glued and cleated the cracks and glued the loose brace using hide glue - the hot kind that you mix up from flakes the day before. It is more trouble to use than yellow glue, epoxy, or super glue, but has the advantage of being totally reversible with heat and moisture, so that if I screw up it can easily be removed, or if the repair fails it can be re-glued with the existing glue just by heating with a moist cloth warmed on an iron and then letting it cool.

The next day I strung up the guitar as a 6 string tuned to concert pitch and within a few minutes the brace that had been loose broke in half (I guess it had a weak spot). It was about 10 pm so I just went to bed and glued it the next day, also sandwiching the joint between two new strips of wood the depth of the brace and about 3/32 inch thick (cut from a scrap from a guitar top left over). I let the glue cure overnight and this time strung it up with all 12 strings, tuned down 1 whole step to D. It lasted a few more minutes this time and then the fret board and the section of the top right below it shifted about 1/8th inch toward the bridge. I hadn't noticed, but I suppose in the break the previous day the 2 cross braces under the fret board had come loose at that location (under the fret board); at any rate they were loose now. I pulled the 18th fret and drilled 2 holes through the fret board and the wide cross brace beneath, shaped 2 ebony pegs, added hide glue and tapped the pegs in, then put the fret back and re-glued the cross braces to the top (at least around the edges) where I could get hide glue in.

After curing overnight, this repair failed just as before, with the fret board and the section right below it shifting toward the bridge (the piece of the top directly under the fretboard has cracked on both sides and can move independently from the rest of the top).

I was stumped so I went and asked a local luthier for advice. After bringing the guitar up to tension he immediately concluded that the heel block had become separated from the guitar body, and suggested removing the neck and rebuilding the block. I had not expected to have to do such a drastic repair, but figured it would be a good learning experience since I had read about the procedure but hadn't done a neck reset before. However, I kept wondering how the center section of the top was moving (while the outer part of the top wasn't) when it was now solidly attached to the wide cross brace that spanned the width of the whole top. I didn't see how the problem with the heel block could explain this.

I looked at the braces more closely and noticed that the wide cross brace had come loose from the outer section of the top on one side; I further noticed that only this side of the center section was moving, i.e. the center section was pivoting, held in place on the other side but moving on the side where the wide cross brace was detached. Aha! I thought. I wedged a small piece of wood under the loose part of the brace to hold it away from the top, and put as much hide glue into the opening as I could, then heated the area with a hot moist towel to make the glue runny again and let it cool while holding the joint in place.

The next day when I strung up the guitar to partial tension it still moved, but not as much. Even though it was not tuned up fully to D, I thought I was on the right track, and figured maybe I just didn't get enough glue in the joint. So I drilled a small hole in the top over the brace and put hide glue in the hole. And I glued a 3/32 inch piece behind the wide brace, butting the other side against the brace by the front of the sound hole, to act as a stop to keep the wide brace from moving. But the "sistered" pieces I had glued on to fix the broken sound hole brace protruded so I stopped the piece where these started, which happened to be just at the edge of the crack in the top (it would have been better to fit a piece and span the crack, to double the effect of the wide brace). So just in case, I glued a similar piece between the wide brace and sound hole brace, under the fretboard, hoping that it spanned to the other side of the crack enough to serve this function.

After the glue cured and I tuned up, the center of the top only moved about 1/16th of an inch and held for 6 hours, before I detuned the guitar. In this condition the guitar was quite playable, and sounded good. But I could still see that the center section was moving, changing the neck angle slightly for the worse, and buckling the far side of the sound hole some. I figured the luthier was right, the heel block was detached from the sides of the guitar. When I checked I found that this was true. Also I saw that even the far side was moving, but the heel block was not detached on that side so the effect was different. I'm thinking that this may be caused by the loose side and may go away if that side is fixed.

But I'm still hoping to avoid removing the neck (since the guitar does not need a neck reset, the neck angle is good if the joint holds). So I pulled back the binding at the top and heel joint and put hide glue in the small opening, then heated and let cool. I don't know if I got enough glue in the joint to do the job; maybe I will end up having to pull the neck to really fix it.

Sunday, March 17, 2013

Initial Clean Up

I cleaned the entire guitar by wiping with a damp rag and drying, and wiped down the fretboard and wooden saddle and adjustment screws with alcohol (ethanol). To remove the white glue from the cracks I dabbed the glue line with distilled white vinegar using a small piece of cloth and tried to keep the vinegar off the surrounding area by wiping the excess with a wet cloth and drying. This helped some, but I had to apply vinegar many times and it left an etching or slight discoloration on the finish. Eventually, after maybe 20 minutes the glue softened a bit and using a 1/4 inch chisel I carefully chipped at it until it was gone. Then I cleaned the vinegar off with a damp cloth (water) and dried.

I only chipped the Sunburst finish in a couple of places, should touch up pretty easy!

I french polished the top. This got rid of the discoloration and new scratch marks around the cracks, as well as making the many other cracks in the finish less noticeable. I had earlier tested a small area and discovered that it was soluble in alcohol; maybe Gibson used shellac over the Sunburst, or even if they didn't, I figured french polish would work because of the solubility.

** A NOTE OF WARNING HERE ** I have sort of taught myself how to french polish over the last 8 years, but it's only lately that I've felt like I could get predictable results. It's very tricky. I don't use oil (maybe that's the problem!) - I know there are many that polish without oil lubrication and I figured that would be one less thing I could screw up. But with no oil it's very easy for the pad to dig in and scuff up the existing finish. You need just enough polish (liquid) to lubricate and polish at the same time, but if you have too much it will start removing the existing finish.

Just to give you an idea of what can go wrong: I had tried using mineral spirits, and then rubbing alcohol to remove the adhesive left by the missing pick guard. Neither worked, so I used alcohol (ethanol) to clean off the rubbing alcohol. A little of this got outside the pick guard area and removed some of the clear finish, so I tried to touch up by french polishing. I had too much liquid or something and it kept removing the finish and also left a white discoloration over the whole area around the pick guard. I thought maybe this would come out if I waited until the next day, so I went on to french polish around the cracks. But again, I had too much liquid and it just left a smeary trail everywhere I polished. I finally gave up and went to bed.

The next morning the white discoloration was still there, I had another try at french polishing that area and was able to remove it. I went over the whole top and improved it, but it was still dull and showed marks from the pad. That evening I sanded the entire top with 800 grit and then 1500 grit sandpaper using an eraser for a sanding block. (I first did the pick guard area with 150 grit). Then I french polished again and finally got pretty satisfactory results. I still have some more work to do on the finish, but it's getting close.

The Guitar Arrived!

Cracks in the top patched from above with a fat line of white glue

cracks above soundhole and missing purfling

Overall, the guitar was in much better condition than I expected! The problems were limited to the cracks on the top and a loose brace under the end of the fingerboard. Obviously this was catastrophic, the neck was free to rotate towards the sound hole due to the cracks on either side of the fingerboard and the detached brace; but it looks like it will be pretty easy to fix by gluing and cleating. What had looked like a repaired crack in the back of the peg head was just an impression in the finish from the tuning machine. The neck is one of the wide, fast ones: 2 inches wide and 3/4 inch deep at the nut, 7/8 inches deep at the 5th fret. I also play classical guitar, so I like the wider neck. And the top gives a satisfying clear tone when tapped, even with the cracks and loose brace!

The question is, why did the brace come off and the top crack? It won't do much good to repair everything just to have it break again. This guitar is very lightly built: the instrument doesn't weigh very much and the braces are all thin and not very tall. Perhaps that's what makes the great Gibson sound of this era. Gibson discovered that these guitars couldn't be safely tuned to concert pitch, maybe the original owner didn't know this. I plan to tune it to D, this worked fine for Leo Kottke's Gibson B-45.

And the neck looks straight, and the neck set seems about right!

Tuesday, March 12, 2013

Broken Instrument Purchased

Saturday I bought a 60's (maybe 1968?) trapeze tailpiece Gibson B-45 in very rough condition (the price was right). The top has large visible cracks crudely patched with heavy-beaded glue lines (maybe Elmer's?), it looks like the peg head split off a piece that has been glued back, it is missing the tailpiece, tuning machines, pickguard (but it did have one, you can see the outline and glue/residue), and truss rod adjustment cover. Additionally the seller described cracks in the neck and heel, although I couldn't see these in the photos. I managed to find all the missing parts online in the past several days, so I am just waiting for the instrument and parts to arrive.

My initial thought was to convert it to a 6 string, since it was already so far gone and I don't really play 12 string. But a little research showed me that this instrument is valued by 12 string players and ought to be saved - and even though I haven't personally inspected it yet, I believe it can be. I have built and repaired several guitars and will be doing the work myself.

Further investigation led me to some of Leo Kottke's recordings posted on You Tube. Of course as a guitarist, I knew about Kottke, and had heard some of his stuff over the years and was definitely impressed. But I guess I found something that grabbed me a bit more, or maybe my tastes have changed so I better appreciate what he's doing. Anyway, I want to play that stuff, and if I can get this instrument back in shape then I've got the guitar to do it.