September 18th, 2010
I haven't updated my notes in awhile, in part because it has been a somewhat repetitive stretch, working on the sound post patch and on the back.
I try and do some most days, and, of course, shop cat helps, here clamping the sound post patch block after it has been glued. She has a gentler touch than me, which is important when using C-clamps.
Carving in progress...
... and done.
I wish it was really that fast.
It's not immediately clear from the image, but I've made the area of the patch where the sound post is likely to connect about one millimetre higher than the top itself for extra strength.
I posted a picture to the thread I started on fiddlehangout.com, and Michael Richwine (fiddle hangout handle of woodwiz) there expressed some concern about the angle:
Most people just angle the patch 2 or 3 grain lines. This reduced the chance of further splitting.
Rotating the patch more presents two problems. First, it stiffens the area substantially. This could be good or bad, but if you are trying to keep the original sound, it probably won't help.
Second problem is that wood expands and contracts approximately ten times as much across the grain as with it. This means that the two pieces of wood with crossed grain will tend to fight each other as the humidity changes. The more the angle, the more tension is created. That's why, in conventional woodworking its axiomatic that cross grain joints always fail, eventually. The movement shears the glue line.
I had thought the conventional approach was to line up the grain and that I was being naughty by angling it at all. Good to know for future reference.
But as I responded to woodwiz in the thread, I'm a little paranoid about this thing having self destructive tendencies, and I'm not sure I would feel comfortable with a wee two or three grain lines.
I considered redoing it, and would if I were less concerned about strength, but on this one if I redid it I would be inclined only to lesson the angle say 50%, and I don't know how much difference that would actually make, that is to say, is it worth it.
And I haven't been neglecting the back.
This approach worked well until the last couple of cleats, which I removed. The seam was insufficiently tight and depressed. So I decided to do a variation on the sand-bag thing, and lay down a strip of damp J-cloth, pressed down upon by a bar of maple with rounded edge, essentially applying force more specifically on the edges of the seam than did the sand-bag.
You may note that the spacing of the cleats is wider in this pic than in the previous. I'm removing every second one once it has performed its duty in the zip up process. I don't think they all need to be there permanently.
I left it for four days. On removing it, the seam was just about perfect, just a little light showing through at the bottom. For about five minutes. Then there was a loud snap that startled the bejezuz out of me as the seam failed dramatically. Sigh.
On the up-side, it did make further progress in flattening the edges of the seam, though not as perfect as it was for those wonderful five minutes.
I considered doing it again, perhaps leaving it for a week, but decided to return to the zipper strategy. If it wants to be ornery then let it. Five minutes is actually better than five weeks or however long it takes before the top goes on -- this would be a very annoying thing to have happen after the top was glued in place.
As I've written before, what's really required is disassembly of the body then reassembly with fixes. The plate halves should be made to fit together perfectly without application of any force. But that will be a project for an unspecified future. For this project I'm focussing on the top, and work on the back is essentially incidental. I'm already giving it more attention than I originally intended when I thought I would just fill the seam with hide glue to make it air-tight.
October 10th, 2010
I'm getting quite slack indeed with keeping up with my notes. I've been plugging along, but progress was slowed down by a trip out of town.
I've basically been working on the M1 crack complex, from bottom to top, adding new cleats, and replacing old ones.
Of interest is the use for the first time of linen 'cleats', one at the very bottom of M1 where the surface is curvy and linen easily conforms to the curve, and one just above the sound post patch.
The linen patches were given two doses of hide glue, one to initially saturate and stick, and another to reinforce after the initial had fully hardened and shrunk.
The back seam is now fully cleated, though haven't gotten around to trimming. I've also decided to more fully address the ribs, especially the bottom ones, unglueing them in order to better be able to repair the crack in the bottom block, and then gluing them back in place from the corner blocks down. If there's excess at the end, as I've read there sometimes is in old fiddles, it will be trimmed. I'm still thinking I won't give the top ribs similar treatment, but who knows. I'll do the bottom bout ribs first and see how that works out.
I wonder if it will be finished by this time next year.
That is all for today. The story will continue
October 25th, 2010 (p10, broken bits),
Back to home,
page 1 (intro),
page 2 (mapping cracks),
page 3 (shaving bass bar)
page 4 (crack repair),
page 5 (crack repair),
page 6 (back repair),
page 7 (patching and back repair),
page 8 (patching and back repair)