Harpsichord Project Part 20B - Installing the Soundboard

Status
Not open for further replies.

ErnieM

Ernie
Corporate Member
Hi all,

As is the case with most woodworking projects, preparation for gluing takes longer than the gluing itself. This is certainly the case with a soundboard. Before any glue can be used, a couple of things need to get done. The first of these is the installation of what I call the crown dowel. A little explanation is in order.

A harpsichord soundboard is a flat, 1/8” thick piece of sitka spruce. It’s front edge is glued to the top of the Upper Belly Rail. All of the other edges are glued to the top of the cheek, bentside, tail, and spine liners (see Part 7). Since all of these structures are exactly the same height from the top of the case, the soundboard will be flat when it’s glued to them, even if it’s not quite flat before gluing. The fact that it’s so well secured to the case is a problem, although a necessary one. As I write this, it’s January 21, and it’s quite dry in my shop. What’s going to happen to this glued-in soundboard in July when the humidity is quite high? The spruce will expand across the grain – or it would if it weren’t glued down on all sides. If this were a thicker piece of wood, it would probably crack, but this thin piece of wood has another option. Since it can’t expand normally, the only thing it can do is to bulge at the center. This bulging is called crown. What we don’t know is the direction of the crown. Will the soundboard crown downwards (sink) or will it crown upwards (rise)? Either way, a well made harpsichord will still function but it’s always better if it crowns upwards. The crown dowel helps to persuade the soundboard to crown upwards. Here’s an exaggerated illustration of how it works.

install-14.jpg


A dowel is glued to the case bottom with a drop or two of CA glue. The height of the dowel is 1/16” longer that the height of the soundboard liners. When the soundboard is placed in the instrument it will not lie flat – instead it will rock because the dowel in the center is keeping its edges from touching the soundboard liners. Of course, we’re only talking about 1/16” and it’s still easy to clamp the soundboard’s edges to the liners with the padded nails we will use. This causes the soundboard to be glued into the case with a slight upwards crown. When the humidity arrives, the soundboard is already crowning upwards so it’s more than likely it will continue to do so as it tries to expand. I have no way of proving that this works scientifically, but I do know that I have never had a soundboard crown downwards, and neither has my teacher, Paul Kennedy, who taught me this procedure. Between the two of us, we have built more than 50 instruments, all crowning upwards. Here’s the crowning dowel in position in the case.

install-15.JPG


Notice that the dowel has a string attached to it. Once the soundboard is glued in, we need some way to get the dowel out. The other end of the string comes out of the instrument between the lower and upper belly rail (I told you the space under the soundboard wasn’t air-tight) and lies in the space where the keyboard will sit. When we want to remove the dowel, we just find the end of the string and give it a tug. The CA glue joint between the dowel and the case bottom is easily broken and the dowel is pulled out by pulling on the string. We’ll leave the dowel in place until the strings have been installed in the instrument as it gives support to the thin soundboard when we push the bridge pins into the bridges.

Okay, we’re ready for glue . . . almost. Since we’re going to be driving some 70 brads through the soundboard only ¼” from the edge, we’ll drill holes in the soundboard to keep the thin spruce from splitting. This only takes a few minutes and could avoid some splitting that might show in the finished product.

install-16.JPG


Now we’re ready to ...... nope, not quite. Let’s take a little closer look at the padded nails we’re going to use for clamping.

install-17.JPG


While it’s pretty easy to tap these padded nails in during the glue-up, it isn’t always easy to remove them after the glue has dried. Glue tends to wick up through the hole in the soundboard and glue the pad to it – don’t ask me how I know this. So we’ll take a few minutes and brush some shellac over both sides of each of the spruce pads. This keeps the glue from sticking to the pad (or at least not sticking well) and makes removal much easier. Believe me, it’s better to take some time to do this than to spend even more time chiseling glued pads off the soundboard.

And now, we’re ready for glue-up. This is a race against time – it takes a while to set the clamps on the belly rail and to tap in 70 padded nails. We have 20 minutes or so to get the job done so everything we need is prepared ahead of time. A bead of yellow glue is put on top of the Upper Belly Rail and all of the soundboard liners. We don’t take the time to spread the glue – the soundboard will do that for us when the nails go in. No glue is placed on the edges of the soundboard. We do the clamping at the Upper Belly Rail first and then go around the perimeter of the case tapping a padded nail into every other hole.
When that’s done, we go back and tap the remaining nails into the holes we skipped.
This glue-up took 23 minutes and here’s what it looks like.

install-18.JPG


After an overnight dry, it’s time to remove the padded nails. The hitchpin rails will be glued on top of these nail holes so they won’t show, but if we’re not careful we can damage the soundboard in areas that won’t be hidden. The best and safest method of removal is to split the little spruce pads in half with a chisel. Then the two halves of each pad are carefully pried off leaving the nail heads sticking up 1/8” above the soundboard – just enough to grab the nail head with a pair of pliers and pull them up.

install-19.JPG



After all the padded nails are pulled it’s time to install the hitchpin rails. The hitchpins for the 8’ strings will be driven through the soundboard and into the soundboard liners. The hitchpin rails that will be installed on the cheek and spine edges of the soundboard will have no hitchpins driven into them, and serve to cover up the holes in the soundboard that we made during the soundboard glue-up. The bentside and tail hitchpin rails, however, will have hitchpins driven into them and they must be firmly glued to both the soundboard and case sides in order to resist the pull of the strings. Clamping these rails is very difficult as conventional clamps cannot be used. Instead, we’ll use some shop-made cauls to apply clamping pressure. Here are the cauls.

install-20.JPG


For tonal reasons, the hitchpin rails are made in three different heights – ¼” high on the front 2/3’s of the bentside, ½” high on the back third of the bentside, ¾” high on the tail, and back to ¼” high on the spine. The cauls are made in three different sizes to accommodate these differences. Each caul will be clamped using two clamps – one to push the rail into the case side, and the other to push down on the rail and clamping it securely to the soundboard. Here are the cauls in action.

install-21.JPG


Once all of the hitchpin rails have been installed, all that’s left to do is to glue down the text band around the rose. That done, the next photo shows the final result.

install-22.JPG


Except for the lid, which we’ll build towards the end of the project to save shop space, all of the woodworking for this instrument is complete. What remains to be done? Painting, marbling, papering, soundboard decoration, stringing, and installing and regulating the jacks and plectra. I’ve decided to paint the harpsichord off white (almond) with marbled side panels, and block-print papers (a rather fancy Flemish wallpaper). The marbling and papering is usually reserved for larger (more expensive) instruments, but it does cut down on the painting and will be more interesting for you to endure. Unless you’d rather see 50 photos of paint drying! So next time, we’ll start the decorating process.

Till next time,
Ernie
 

DIYGUY

New User
Mark
Ernie,
I do not build musical instruments and have nowhere near the talent you do, but I can share one small technique with you that could make one task easier.

Whenever I have to pull nails out in tight places (as you have with your padded nails and pliers), I use my Vise-grips. They won't let go of the nail head and the top of the tool is nicely rounded, just right for becoming a fulcrum. I put a small scrap of wood next to the nail to protect the surface from the grips and then use the rounded side as a lever to cam the nail out. No muss and no fuss. The result is a curved nail and a slightly elongated nail hole but it almost always works very well.

Thanks very much for the truly mesmerizing essays on your craft ...
 

ErnieM

Ernie
Corporate Member
Whenever I have to pull nails out in tight places (as you have with your padded nails and pliers), I use my Vise-grips. They won't let go of the nail head and the top of the tool is nicely rounded, just right for becoming a fulcrum. I put a small scrap of wood next to the nail to protect the surface from the grips and then use the rounded side as a lever to cam the nail out. No muss and no fuss. The result is a curved nail and a slightly elongated nail hole but it almost always works very well.
Mark,

Thanks for your reply. Actually, I do use your method when I get a stubborn nail. The leverage is much better. I try not to use Vise-grips in this case because the spruce is so soft that the little scrap of wood placed next to the nail to protect the surface makes its own indentation in the soundboard. I use the smallest nail I can get away with - most of them pull straight out without a problem. If they don't - out comes the Vise-grips.

Ernie
 

mlzettl

Matt
Corporate Member
Ernie,

I thought that some of my glue ups complicated, but they're nothing compared to this. Which brings up my question. Today we have the luxury of many different types of adhesives that vary according to their strength, working times, durability, etc. When they were making harpsichords a couple of centuries ago, and only had animal glues like hide glue, how did they do these glue ups? Does the type of glue that you use on the soundboard have any effect on the instrument's sound? Original Titebond has a little flexibility, does it not? Most epoxies are fairly brittle once cured, and I wonder if this makes any difference in choosing one over the other. Don't violin makers still use hide glue? Sorry for all the questions, but the whole process just amazes me and arouses my curiousity.

Matt
 

ErnieM

Ernie
Corporate Member
I thought that some of my glue ups complicated, but they're nothing compared to this. Which brings up my question. Today we have the luxury of many different types of adhesives that vary according to their strength, working times, durability, etc. When they were making harpsichords a couple of centuries ago, and only had animal glues like hide glue, how did they do these glue ups? Does the type of glue that you use on the soundboard have any effect on the instrument's sound? Original Titebond has a little flexibility, does it not? Most epoxies are fairly brittle once cured, and I wonder if this makes any difference in choosing one over the other. Don't violin makers still use hide glue? Sorry for all the questions, but the whole process just amazes me and arouses my curiousity.
Matt,

I'm not certain how it was done but my guess is extensive use of go-bars and manpower. Many builders still use hide glue today, but I don't think I can work fast enough to use it on something like a soundboard glue-up.
Most pictures I've seen of soundboards being installed with hide glue show go-bars and two or three people frantically putting them in.

I've read where some folks believe yellow glue should not be used in lamination( like in laminating a bentside) because of glue creep. I think there must be some truth to this but I've never had the problem with yellow glue. If it ain't broke - don't fix it.

I only use epoxy in places where clamping is difficult and I need extra time to clamp. Otherwise, I stick to yellow glue. Why Original Titebond instead of the new versions? It's worked well for me and I'm too old and set in my ways to change. It's difficult, (for me, at least) to know how a particular glue will affect the tone of an instrument. Even if two instruments are built identically, except for the type of glue used, you couldn't conclude that the glue is what changed the tone (if there is a difference). There are too many variables - quality of the wood, construction methods, etc. to isolate the glue as the cause.

Still, the type of glue used is a 'sticky' subject for a lot of builders - I'm not one of them.

Ernie
 
Status
Not open for further replies.

Premier Sponsor

Our Sponsors

LATEST FOR SALE LISTINGS

Top