Saturday, 7 December 2019

Building a 13 Course Lute, Start to Finish, 10: Making the Bridge

Hi there. Today, a small, crucial, highly-detailed and finicky item, the bridge.

My favourite material for making bridges is plum. I search high and low for it, and good quality stuff is perhaps surprisingly not easy to find. The tree grows everywhere, but most of what I see for sale has such twist in the log that it's impossible to use for anything but the smallest items--tuning pegs (if I'm lucky), or more likely end buttons. Most of what I've seen could, sadly, be best used as firewood.

Yet every once in a while a log or a board will come my way that has enough straight growth in it that I can use it for bridges. For a bridge blank, I need a width of about 20mm (that's the dimension that's quartered), a height or thickness of about 12mm, and a length of around 200mm (depending on the number of courses and length of the bridge.) If I find such a log or board, I covet it, and cut it very carefully to yield as many bridge blanks as possible. Only the best, straightest sections of it go for bridges; smaller pieces will be used for pegs. (For this purpose not only is it beautiful to the eye, it also turns well, and is stable over time and wears well in the peg box.)

The main reason I like it for bridges is that it is one of the most beautiful sounding woods I have ever heard. If you get the chance, tap a couple of pieces together, and listen. Plum makes a magical, bell-like sound that is both pure and rich, and I want that sound to be part of every lute I build. Every piece of wood in the instrument has its own beauty of tone to bring, but they are all different, like different beautiful voices in a choir. The best choirs blend those sounds into a single voice, and the result is greater than the sum of the parts. That's what I hope, and want, for my instruments.

And so to work.

(By the way, the model I'm using for this bridge is from the 13 course lute of 1727 by Sebastian Schelle, in the Cité de la Musique in Paris. You might see a photo of it in one or two of the photos in this blog post. This lute also serves as the model for many of the features of the peg box, as we shall eventually see.)


Here's my blank. I've flattened the bottom first, with my low-angle block plane. If you look carefully at the photo, you will see on the right end of the blank that I've drawn an arrow in pencil--this marks the best planing direction for this piece. As you'll see if you tune into the next episode, I will want to know that direction later, after the bridge is finished, and I'm ready to glue the bridge on the belly.

Here I'm shooting the front edge of the bridge. I want it straight and perpendicular to the bottom surface.


When I've got those two surfaces, I can begin marking out the length and width(s) of the bridge. Like most if not all lute bridges from the period, this one tapers from bass to treble end...


...as you see here. I've now got bottom surface (facing up), and front and back sides.

Now the top. I set the piece in this jig, which allows me to plane a slope into the top surface with my low-angle block plane. (Again, the back-to-front slope is characteristic of most authentic lute bridges.)

Here's a look at the jig. There's a slot or edge on each side of it, allowing me to plane the sloped top surface (on the left side) and the sloped back edge (on the right side, as you will see in a minute.)

Like the original, this bridge is going to have an ebony cap, that's slightly more than a millimetre thick. If you take a close look at the photo, you see a sloped, cork-lined caul on the top (the slope matches the slope I've planed into the top surface); below that, a piece of plexiglas as a non-stick liner for the caul; below that, the ebony veneer; below that, the bridge itself; below that, a caul with a gentle curve of about 1mm planed into its length. At each end there are two aluminum pins, which allow me to locate the plexiglas, veneer, and bridge.

It occurs to me that interested readers might have one or two questions about what I've just described. Please feel free to leave them in the comments section at the bottom. I will anticipate one here: why do I place a caul underneath everything that has a 1mm curve planed into its length? Answer: I want this bridge to have a slight curve or crown to it, and the best way to ensure that it does is to incorporate it at this stage, when I'm gluing the veneer to to the top. When I remove the bridge from the cauls and clamps, that curve will be there, built-in. As you'll see when I describe gluing the bridge to the belly in the next episode, I will incorporate the curve into that gluing procedure (and if you remember my description of gluing the braces on the belly in the 7th episode of this series, I did the same thing then.) All will become clear in the fullness of time, I pledge it!

Here's the result, with ebony edges trimmed close to the sides of the bridge.

Here you see clearly the taper from bass to treble end. You can also see where I've marked out the final width of the top of the bridge; I will now plane the back angle into the right side of the bridge.

The plane, in action. As when I planed the top slope, all I need do is hold the piece lightly against the jig, and let the pressure of the plane do the rest. 

My result: the top to its final width, the sloping back surface to its final dimension, and--if you look carefully--about 1mm along the bottom of the back side that I haven't planed the slope into. I don't want where the back slope meets the bottom to become a knife-edge, for two reasons: I don't think it will look good, and it won't be well supported when it's glued to the belly. 

Time now to drill the string holes. I have pretty specific sizes of bits that I use for each of the courses, and the strings within the courses (with 24 holes to drill altogether, I need to keep track of them carefully.) The bar of wood that the bridge is clamped to is sloped slightly--3 degrees, as you see written on it. I will drill through the front face of the bridge through to the back, which means that the hole will tilt slightly upward from front to back. (Again, this is a feature of the original bridge, and most originals from the period.)

The bass side of the course gets a larger diameter hole. In addition, the octave on all the bass courses is drilled slightly lower than the fundamental. (By the way, the big brown waxy-looking block you see in this photo and the one above is a block of lead, which I inherited when I moved into my first shop space about 10 years ago. I use it to weigh down the drilling jig to keep it steady on the drill press platform.)

Back to the vice for another jig. This one allows me to plane a slight slope (an undercut) into the front surface of the bridge.

Then I can begin to cut recesses into the front and back of the bridge. Here's the front side: I make a knife cut, with a straight edge, down the line of string holes. Then I cut into that line, across the grain, with a small chisel.

Here's what the first round of cutting with the chisel looks like--now I need to do the same from the opposite side.

There's the result, tidied up with knives, files and small scrapers.

Here's my sloping jig for cutting the recess into the back side.

I use the same procedure here: knife cut on the string hole line, then cut in toward it with the chisel.

Here's the rough result, before the final tidying up with files, knives, and scrapers.

I'm finished with the shaping of the piece--now I can carve the finials on the ends. I mark them out with a pattern on the bottom of the bridge, then cut them out on the band saw.

In this photo and the next three that show me carving the bridge ends, you can see that I'm working against a block of wood, which I sometimes cover with a soft piece of leather. If I have to make some chisel or knife cuts across the grain, as in the photo below, I might hold the piece against the bare wood to back up the cut--so I don't chip out material. If I'm making some free cuts with the knife along the grain, I will usually hold the piece against leather, so as not to mar it.

I need to be very conscious of grain direction when I'm carving these things. One wrong move, and a chunk of this bridge-end could break off--and I might be left with no choice but to start the whole process over, and make a new bridge from scratch.


As it is, things worked out quite well with these bridge ends, and with the bridge as a whole. Here's the bass end, which I finished up with knives, files and scrapers.


Here's the whole thing. I've left the bridge ends a little long--they will be cut back a little before gluing on the belly, and I'll fit them with ebony tips (be sure to come back for the next episode for all the exciting details.)
To finish the bridge, I give it a couple of coats of a drying oil (T&T Oil, to be specific), and a vigorous polish with a soft bristle brush. (It should go without saying that I'm very careful to not get any oil on the bottom of the bridge!) Here it is with the dyed-black bridges for the two other lutes I'm building, a 7 and 8 course. 
And that's my story for today. I wish you all a fine week, and I'll see you next time for gluing the bridge to the belly. Cheers!




Saturday, 30 November 2019

Building a 13 Course Lute, Start to Finish, 9: Fine-fitting the belly, Locating the Bridge

Hi there, friends. This time out, as the sign says, I'm going to talk about fine-fitting the belly into the waiting body of this 13 course lute, and then locating the bridge. I've already done a 'rough' fit of the belly, by doing a preliminary trimming of the bar ends; now I'm going to get down to the business of giving this lute a sweet shape that corresponds to the outline that I drew on the belly a couple of episodes back, before I glued on the braces.

Just before I do, could I ask a small favour? I get a few comments and questions regarding these blog posts. Some--though very few--appear in the actual comments section of the blog; most have come to me either through social media, personal messaging, or email. The favour is this: if you have a question or comment, would you mind leaving it on the blog, in the comments section below? I'm happy to answer, but I would prefer to do it here, so that those questions and answers become part of the post, to clarify, correct, or otherwise comment on it--for the benefit of all readers.

(There's no reason to be embarrassed to ask a question, but if you'd rather not use your real name, please feel free to use a pirate name or your luchador alias. Your secret's safe with Google.)

On with the show.

As you might recall, when I initially cut off the bar ends I left them a bit long, on purpose. It's difficult to saw the bar off at exactly the right spot on the first go, so that the length of the bar and the angle of the cutoff are both correct. I leave them a bit long, so that I may trim them more carefully, using a small chisel rather than a saw.

To figure out exactly how much to trim them, I need to fit the body down over those bar ends. So, the first thing I need to do is remove the false belly, which I do simply by slipping a palette knife around the false belly edge, separating it from the rib where I originally spot-glued it with weak size.

I then set the edge rib carefully down against the ends of the bars. The photo below shows how I hold and view the work, though I haven't quite got the edge rib all the way down yet (that's one of the problems with using a camera with tripod and 8-second timer--I don't always get the event staged exactly as I want to.)

Imagine about one second into the future: the edge of the rib will be down against the belly, and the ends of the bars will be resting against the inside surface of the edge rib. I can then take a sharp, soft pencil, and make a mark on the belly, right against the rib, opposite each of the bar ends.

Please keep in mind that during this process, I'm not attempting to press the entire outline of the lute down over the bar ends (since the all the bars are over-long, it won't actually fit.) Instead, I do it in three sections. In the photo above, I'm working with the bass side. When I've made marks for all of the bar ends there, I'll turn everything around and mark the treble side. Finally, I'll work with the bottom of the body, and mark out the bar and tab ends there.

Here's a well-focussed photograph of my knee. Actually, what I meant to show here is how I hold the edge rib against one of the tabs on the bottom of the belly while I make my mark.

Here's what I get when I lift the body away--a pencil mark outside the drawn body outline that shows me exactly how much I need to trim this tab. 


And so I trim it, using a small chisel (1/4", with the handle cut off--perfect for small operations like this.) On the upper left corner of the photo, sitting on the edge of the bench you see a little jar of water and a brush: if I dab the end of the bar with a bit of water, the chisel cuts through the end-grain a lot more cleanly and easily.

When I get close to the outline, I start fitting the body down all around and looking carefully at the way the edge ribs curve. I want a smooth outline, so if I see a bar end that needs a little trimming, I'll take the time to do it. (Even a small adjustment can make a big difference to a graceful shape.) I also carefully flex the edge rib against each bar end, to feel whether the angle of the bar-end cutoff matches the true angle of the rib.

When the bar ends are trimmed and the outline looks good, I tape the belly securely into the body with masking tape. I make sure at this point that I've really located the belly carefully, by matching body and belly centrelines at the bottom, and neck edges (marked on the belly) at the top.
Now I'll move onto the next step in the process: locating the position of the bridge on the belly. I want to do this very accurately, because I'm finding and marking the exact spot where I want to glue the bridge.

Most of the time during the working process, I protect the sharp veneered edges of the neck by covering them with masking tape. However, right now I want to see the edge of the neck clearly, because that's one of the crucial references I use to locate the bridge. I've rolled back the masking tape, and there it is.
When I originally lay out the shape of the lute on my working drawing, I have a very clear idea how far I want the first string to lie from the edge of the neck (which, essentially, is also the edge of the fingerboard.) This distance is very important for the playability of the instrument--and anyone who has ever tried to play a lute with a first string that's too close to the edge of the fingerboard will know what I mean. The distance can vary a little bit, according to a few factors--the length of the neck, for instance, or the preferences of a client who's a very experienced player--but whatever those factors are, I want to have them worked out completely before I start building the lute.

So, I have the edge of the neck, and I have the position of the chanterelle in relation to it. This means I should be able to run a straightedge down that line, and somewhere along it will be the location of the first course on the bridge. But where? And what will be the angle of the bridge in relation to the first course?

The clearest and most foolproof way for me to find this out is just to set my layout drawing down on top of the lute, and see how things match up.

The drawing is done with 4H pencil on drafting mylar, and shows the positions of the neck edges, the first course, the bridge, the rose, and the body outline.
If I've done my work carefully so far, things should line up pretty well. The neck angle is good, the location of the rose is very close, the belly outline is fine... I should be able to just go ahead and mark. (If on the other hand some crucial piece of alignment is off--for instance, the neck angle's not quite where I want it to be--then I need to move the drawing around a bit, to compensate, making sure that the crucial relations of the neck edge and chanterelle are maintained.)

The rose centre and the bottom of the belly are two reference points I use to locate the layout drawing.

Using the working drawing to mark the position of the bridge allows me to get the correct angle of the bridge, which is tilted slightly upward on the bass side. 


Here's how I mark: a pin prick right through the mylar, at exactly the point where the chanterelle meets the glued front edge of the bridge.
I make another careful pin mark on the bass side of the 13th course, at the bridge's glued front edge.

I tattoo the pin pricks lightly, with a sharp, soft pencil.

And there it is--one of my two bridge location marks.

Now all I need is a bridge to glue there--and making that will be the subject of my next post.






Sunday, 17 November 2019

Building a 13 Course Lute, Start to Finish, 8: Trimming the Braces, Fitting the Belly

Hello, good morning, welcome back to the shop. This time out, I'll be trimming the braces I've just glued onto the belly, adding a few more small braces, and preparing to fit the belly into the body--one of the turning points of lute making, since we're nearing the point when the sound box is closed, and the lute finally starts to look like a lute.

You might recall that I ended last week's post with an ode to randomness in placing and gluing the braces on the belly. I would like to assure you that that idea did not originate with me. Like all (or nearly all) my good ideas about lute making, it came from Grant Tomlinson, and he has told me that the observation originally came from Stephen Gottlieb, whom Grant studied with in London for a year in 1986. I worked with Grant for a year in 2009, and I remember him recalling to me, fondly, Gottlieb's advice to him: that one should put the braces on "higgledy-piggledy."

Words of the masters, passed down through the ages....

Here's my higgledy-piggledy result.

The very first thing I do is trim the edge of the belly on the band saw, taking it to about 5mm of my marked body outline. (It looks much more orderly already.)

All of the braces are glued on taller than their final height, since the go-bars bite into the tops a little bit. Now I use my low-angle block plane to trim them carefully to height. (Note that I'm supporting the ends of the curved braces with the same cauls that I used for gluing them.)

 I then chamfer the top edges of the braces. For this job I use a little Veritas miniature bull-nose plane, with a scrap of wood clipped to one side as a fence. (It works very well for keeping the plane at a 45 degree angle.)
There's my high-tech setup.

In preparation for scalloping the ends of the braces, I mark a height of 5mm with a wooden spacer.

 I then scallop the ends of the all the braces with a skew chisel.

I look for a graceful and fairly uniform shape on the ends of all the transverse braces.
When I have a good shape on all the braces, I need to shape and glue down a few specialty pieces. I put spruce tabs in a few spots around the outline of the lute, where there are no bar ends to support the edge rib. The number and position of tabs varies with different lutes and barring patterns, but on this one I'll stick down a couple between the first and second transverse bars, and a couple on the bottom of the belly, below the j-brace.

Here's what the tab looks like--just a short length of bar material with a slipper-shape carved into one end.

And here it is, stuck down (with hide glue, of course.)
I also shape and glue a small "chanterelle bar" directly under the position of the first course. This bar helps to support the sound of the top course on the two highest frets. 
Now I need to take you on a small, rather boring detour (just to remind you, again, that lute making is not all glory!) My next step is to glue some small support bars across the rose, and normally I'd just go ahead and do that by reaching for my stock of ready-made rose bars, but... sadly, my stock is depleted. Oh well--nothing to do but take a couple of hours and make a hundred or so, so I don't have to do it again for a while.

I thickness some sheets of spruce brace material to 2.5mm, then shoot one edge clean.

I use my shop-made purfling cutter to cut off a strip about 3mm wide.

There's one--now I'll shoot the edge anew, and cut off another strip, and so on, until I have a whole bunch.

I stack them side by side in my planing box, rough side up, and....

Plane them flat, and a uniform 2.5mm thick.

I then chamfer the top edges with a ledge jig and my low angle block plane. 
That's it for the detour--now I can get back to the task at hand, which is gluing the rose bars on the rose.

Having cut the bars to length and blackened the bottoms, I carve a little scallop in the tip of the bar. Why the scallop? Because that makes it easier...
To scorch the tip!  I brush glue on the bar, stick it down in place, and then bring in my wood burning knife. Scorching the tip has the immediate effect of crystallizing the hide glue, sticking the tip of the bar fast to the belly. When I've scorched both ends, I don't need to use any other clamp to hold this bar down--it's secure. I can move on to gluing the next.

There they are, reasonably neat and orderly--and an illustration of yet another of the many miraculous properties of hide glue.


I'll let the glue dry thoroughly before working any further with the belly. In the meantime, I have to do a few things to prepare the body for fitting up the belly.


I remove the screw I used to attach the neck (a #8 deck screw)...
Drill out the hole, and swap it for something a little beefier (a #14).

I carve away the rough edges of the counter cap (which I glued in long ago.)

I don't want any edges inside the body to be too abrupt or too sharp. I have the notion (maybe it's a superstition) that they might impede the efficient production of sound.

Now I need to go back to the neck, and do a little more planing.

I want to make sure the surface is quite flat in long section (it's been a few weeks since I worked with the neck and it may have changed shape slightly since then.)
But that's not all I'm doing here. To tell you the complete story, I'm actually planing a twist in the upper surface of this neck. Specifically, I'm planing a slight downward slope across the neck, so that, at the nut end of the neck, the treble side will be somewhat lower (maybe about 1mm or so) than the bass side. (Even though it is twisted, however, the surface of the neck will continue to be flat in long section.)

This has the overall effect of lowering the action--that is, the eventual height of the strings above the fingerboard--of the treble strings in relation to the bass.

This operation is all carefully controlled and accounted for in the planning and building of the instrument, throughout even the earliest stages of construction. For instance, I know from the outset that I'll be planing this twist in the neck, and therefore I know that I will be reducing the neck's thickness by a certain amount (a bit less than a millimetre.) I have to account for that when I'm laying out and shaping the neck. Similarly, I know that if I plane away some material on the treble side of the neck, the neck will end up a little bit narrower--and so I have to plan for that too.

Just a few of the complications that make building a 13 course lute such special and challenging project!

All right--the body is pretty much ready to accept the belly. So now, I can get back to the belly, and begin fitting it up.

I use templates to mark the cut-off angles for the various bar ends (I've taken these angles from the mold.)

I use a razor saw (with a lot of wax on the blade) to make the cut. I make sure not to cut too far inside the line (the body outline, marked on the belly)--I want to leave myself some extra length, so that I can later trim the bar end to the exact length.

A small chisel removes most of the excess material, but I don't go all the way to the belly surface (because I don't want to mar it.)

Instead, I leave a bit of bar material right next to the belly.

To remove it, I use my wood burning knife and a strip of cotton soaked in water. I apply a little steam...

And the glue releases. I can remove the bit of bar with a sharpened stick, and the belly is left unblemished.

Last step for now is to trim the belly edge to within a couple of millimetres of my outline. I'm pretty much ready to start the fine fitting process, which will be the subject of my next post.