October 22, cont'd: While admiring the instrument, I noticed that there's an enormous amount of deflection being introduced to the soundboard at the treble bridge. Worse: the backboard is being deformed a similar amount, so the pressure of the strings is pushing straight through. I can't tell if the amount is increasing or if I have just noticed it. I am thinking I need to install a girder of some sort between the top and bottom rails near the treble bridge to limit this deflection. Let the soundboard come down far enough to put pressure on the treble bridge brace, but not so far or so hard as to distort seriously the back- and soundboard.

First, think it through. Second, take RD's advice to dowel the rails to the ends of the pin blocks because after this mod they'll be involved in resisting the vertical component of the strings' pressure. I don't think there's any hurry to do this, but it needs to be done during some round of improvements (along with raising the bass bridge slightly). Next month? Sooner if I notice a change for the worse or in case something dramatic happens.

Ah, hell. See the problem, fix the problem... I notice a pattern here: dulcimer design problems nag on me until I can think of nothing else and then I have to fix them in order to get on with other business.

I detuned the treble strings to relieve the deflection. Drilled two half inch oak dowels into each end of the long rail and one half inch dowel into each end of the short rail. Then measured half a dozen times and drilled two 3/4-inch holes to receive a brass rod that reaches from the long to short rails adjacent the treble bridge brace. (Have I mentioned that I am building this to be a lightweight instrument? I didn't think so. I'm not, and it isn't.) While I was at it, I detensioned the bass strings to raise the bass bridge. First, I raised it by 1/4 inch, which was too far. The bass strings then interfered with the treble bridge. So I dropped it back to 3/16's or a whisker less which seems good all around (made these adjustments by ripping strips of cherry to fit beneath the bass bridge).

I managed to damage the short rail while driving the brass rod into position, but it's nothing some glue and a pipe clamp couldn't fix. I sanded the rails again to bring the dowels flush. And after dinner I spent another hour re-tuning. The strings are getting tired of this: I popped three tonight. I left the dulcimer in a poor state of tune. Figured I might as well let it reach some state of equilibrium overnight before bringing it back on pitch tomorrow. And now that I look over at it, you know what? The soundboard and the backboard are still distorted right where they were to begin with.

The soundboard sinks about 3/16 inch under the high pitched strings. The back sinks just as far as ever, too. At least now there is a solid brass girder in place to hold shims if this continues to be a problem. . .

Of course it continues to be a problem! How could it not? Every time I looked at the dulcimer, it reminded me of the Tacoma Narrows Bridge just before The Collapse. But never fear: the shower think-tank worked again.

October 23. I stepped out of the shower and sketched the solution late last night, then got up this morning and spent half an hour in the shop. I drilled four 3/4-inch holes in a handy oak stick exactly 3/16-inch from the edge; then ripped the board so that its thickness was the width of the opening between the short rail and the sound board (that's so I can place the spacers without removing the soundboard again); ripped the board through the diameter of the holes (so the oak bits would sit securely on the brass rod -- see picture); and finally I used the miter saw to turn the series of arches in one piece of oak into four seperate saddles (two practice saddles are shown with the leftover brass bar). I detensioned the high treble strings (again!) and slipped four of these saddles one after another onto the brass rod, pushing each back under the soundboard. When I brought the strings back up to pitch, voila! No deformation and now all the strings pass through the bridges with plenty of room to spare. The tone at both ends, especially the high end, is cleaner, purer, louder. The soundboard may be more restrained than RD's design would have it, but at least mine is no longer deformed and inappropriately stressed. Other drawings of floating soundboard designs suggest mine is still less bound up than many. What contact there is between strings and soundboard, between the soundboard and backboard, is improved.

Now, can I please pretend to be a musician for a while and stop pretending to be an engineer?

While I sat here writing these notes, the dulcimer popped and pinged quietly at odd intervals like a hot stove cooling. The strings exert 1,500 to 3,000 pounds of tension between the pin blocks and it takes some time for the wood to reach equilibrium.

Twelve hours later, the instrument was silent and the tuning seemed reasonably stable. At 24 hours, things were still pretty much tuned. At 36, several courses were a quarter to a half step flat. I replaced the broken strings and retuned. Over the next week (it's October 30), I've needed to tweak the tuning every 24-48 hours, but few courses have gone more than a quarter step off, and only a third or at most half the instrument's courses have drifted that far.

 

---

Let me steal a few moments here to send "thank you's" to Don Hurless, James Jones, Ardie Davis, and Jerry Reed Smith.

Things would not be happening on these pages without you. The wierd stuff is all mine (a brass rod through the middle of it?) but the basic plan and inspirations are all yours.

 

Throwing in the dowel: two of six oak dowels through
the long and short rails into the pin blocks.

 

A 23 x 3/4-inch brass rod runs from one end rail to
the other and lends strength to the backboard. This keeps
the backboard from yielding to the downward force of
the strings, which keeps the bridge brace from sinking,
which keeps the soundboard level, which holds the bridge...

 

Oak saddles provide adjustable shims between the
soundboard and the brass bar. Photographed
on the porch rail just to show how they fit
over the brass girder.

 

You're on page 18

Building: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Refinements, etc: 16 17 18 19