'A Modest Gluing Surface,' or, Attaching a 12 course Ladder Extension

Today I want to talk about my method for fitting and gluing a ladder extension on a 12 course "double headed" lute. I've made three of these lutes now, the oldest one being I think almost five years old; the extensions are still firmly attached to all three, so I'm fairly certain that my method is a good one, and I can confidently share it with the wide world of the lute.

For those of you who don't know what a 12 course lute with a ladder extension looks like, here are a couple of photos of the finished product. These are shots of my latest one, a 12 course bass-baroque lute for Evan Plommer, completed in late 2015.

So what's the big deal with this joint? Well, the main problem is that the extension attaches to a very tiny corner of the bent back peg box, and that's almost all the support it gets. The extension, as you can see, carries a fair load of tension from the four bass courses strung to it, which pull it both forward and to the bass side. So the extension has to be well made, and the joint has to be well fitted and well glued; there is very little room for error here.

Let's have a look at the neck-pegbox area before the joint is fitted to see what the issues are.

Here's what the area looks like, once the bent-back peg box has been fitted and glued into place. See that little pink corner of the peg box? That's it.

As you can see, a small corner of the neck and fingerboard have been cut away on the bass side to accommodate the "root" of the extension. Just a quick word about layout: the neck is designed with the width of a 10 course lute, and the edge of the cutout is just a couple of millimeters to the bass side of the fundamental of the eight course. (By the way, the width of the bent-back peg box is determined by the position of the 8th course fundamental too, since that string must thread inside the peg box. Therefore, the edge of the cutout, and the inside edge of the peg box cheek, are virtually aligned.)

Confused yet? Try building one of these things. I remember making the first one of them and being just barely able to comprehend how all of this would align and fit together. I've found that the only solution when you feel this way is make accurate drawings, keep careful notes, and proceed slowly, one step at a time.

Here is a photo that shows some of the layout work I did on the fingerboard before making the cutout.

 And here's a shot of me holding my breath, and plunging in, saw-first.

This was the result--an accurate cutout on the corner of the neck and fingerboard. The back of the cutaway follows the plane of the back of the rebate for the bent-back peg box.

Once that was done, I fit and glued the bent-back peg box--and there I was, looking at that tiny pink pegbox corner, wondering how on earth I could get an extension to stick on that.

Last year, when Evan's bass lute was just finished, I posted some of these pictures to Facebook. One of our eminent lute makers of an earlier generation commented, upon looking at this photo: "A modest gluing surface." He then went on to tell a rather gut-wrenching tale (no pun intended) of having one of these extensions shear off, just as he was delivering it to his client.

I had visions of that very thing happening to me when I was making and fitting this extension for the first time (really, it haunted me in my sleep), and for a long time I racked my brain to find a solution, some way to lock the thing in place. I could just fit the extension and glue it and then... what? reinforce it with a wood screw? drill a hole and insert a dowel? wrap it in duct tape? None of these ideas seemed effective or plausible, let alone elegant. I was stumped.

So I decided to do what I normally do in situations like these, which is just go ahead and work one step at a time, and see what inspiration might come along during the process.

The first step seemed to be just to fit the extension--cut a matching notch in the end of it to fit onto the tiny corner of the pegbox.

There it is marked out. I cut the notch with chisels, and checked the goodness of the fit often. I checked the orientation of the extension by holding it in place with clamps and stretching fishing line from the bridge out to the end of the extension. I had worked out (on my layout drawing) the spaces between all the strings, going both to the main nut and to the extension: this gave me the horizontal orientation of the extension. I also had a clear idea how far forward I wanted the extension to tilt, and I used the same fishing lines to orient it in that axis too.

Here's a look at my clamping setup--I used it to check the fit and orientation of the extension, and later on, to glue the extension in place. This photo shows how I achieved clamping pressure across (note the shaped caul I've placed on the treble side of the neck and peg box).

And this photo shows how I achieved clamping pressure downward. Since there is only a small surface on top of the pegbox to apply pressure, I devised a kind of screw clamp (made with an old piece of threaded rod), with a very small foot below it, to get pressure on that very specific point. This whole rig needed to be clamped securely to the top of the fingerboard--that's why you see the wooden cam clamps in the background. It may look complicated, but in fact it's quite a simple and effective jig for holding and gluing the extension.

So the extension was fitted--now what?

Well, it seemed to me that there were still two problems with this joint. First, there just wasn't a very big gluing surface. Second, there was no real integration--no real link--between the extension and the peg box. I felt that, once string tension was applied, there was too great a chance that the extension would just shear off, and I'd be back at square one.

So, here's the solution I came up with: an internal tenon.

I think that's what it's called, anyway. I cut a small mortise in the inside of the rebate, parallel with its top surface; I fitted and glued a small tenon of pear wood into that mortise; and then I trimmed the tenon so that about 3mm or so of it stuck out.

Then I cut a corresponding slot for this tenon in the side of the peg box.

Once the slot was cut, I could slip the extension into place easily.

After clamping and checking the alignment one more time, I went ahead and glued the extension.

And that was it. Next morning, when the glue was dry, I unclamped and had a look. The fit was tight everywhere, and the alignment still looked good; but more important than that, the joint felt strong to me. I took the extension in hand and carefully flexed it, testing it for strength, listening with my ears and my hands for any tiny fractures, any weaknesses, and I detected none. I was pretty sure I had found a solution to the problems with the joint--the internal tenon gave an increased gluing surface, but more importantly it linked and locked the extension to the peg box.

So, I carried on and finished the instrument, strung it up, brought the strings to tension, and... the extension held. Five years later, it's still holding strong. And as I said, I've used it twice since then, and those lutes are holding strong too--so I'm satisfied that I've found a reliable solution to the problem of attaching a ladder extension to the teeny corner of the peg box on a 12 course lute.

Of course, the joint is still very vulnerable to damage--not from string tension, I think, but from impact. (I would be very careful when walking through doorways carrying a 12 course lute!) But I feel that it would take quite a substantial impact to separate the extension from the peg box. And if that did happen, the extension would not simply shear off--instead, the extension might break, or the corner of the peg box would. That's how well integrated the two pieces are on these lutes.

So that's it for this time out. I must confess that I debated for a long time about whether I should share this little insight I had into how to join this extension. One part of me felt like it was a trade secret, and if I gave it away freely, I might be doing myself or my livelihood some harm. However, I've come to the conclusion that it's best to share these insights when one can. After all, everything I know about lute making came from many people who have shared, and to this day still share, their knowledge generously with me.

And what's the worst that could happen? That there's a sudden rash of 12 course lutes being made with secure extensions? That a lute maker somewhere sees my solution to this problem and uses the concept in some new way to tackle a whole other problem? Or that I see some other maker's solution to a different problem, and apply it in a new way in my work? I'd be pleased with any of those outcomes. Happy lute making--and happy sharing, everyone.

New Work, and a Bit Unusual, Too

I see that it's been a very long while since I posted anything here about my new work.

I can't write about everything that's gone on in the shop over the past year, but I want to give some highlights that will show the kind of work I've been up to and the kind of challenges my clients have set for me. I want to talk specifically about two people, both of them excellent musicians who have very clear ideas about the kind of music they want to play, the kind of performing circumstances they'll encounter, and the kind of lute they need to fit the bill.

First up is Evan Plommer, of Sarnia, Ontario. This is the second lute that I've made for him, and just like the first, with this one he set me an interesting task. Again, he was looking for a 12 course lute to explore the world of accords nouveaux or 'transitional' tunings; this time out, he wanted a larger-bodied lute with a longer string length and lower pitch level to explore some deeper sonorities. (The lute has a string length of 79cm.)

Mug shots!

 

I think the body itself may have been the germ of this idea for Evan. It is that of a bass lute originally designed by Ray Nurse--he built an 8 course version of it a number of years ago, which Evan admired very much, so he asked me to design a 12 course lute around it. (The design is from a lute body dated 1589 by Magno Tieffenbrucker, in the Boston Museum of Fine Arts.) Ray kindly lent me the mold, and I completed the lute in the late autumn of 2015.

As you can see, the body has a quite flattened profile--Ray's concept in designing the bass lute with this shape was to enhance its projection. I think it works very well--the combination of string length, body shape, and the material of the back (11 ribs of honduras rosewood) give the lute a mellow yet robust tone.

One of the challenges of building a lute like this is the stepped extension. I've made a few of these lutes now, and I think I've found quite a secure way of attaching it--secure in the sense that it's not likely to separate from the neck unless there is some kind of impact (and let's just utter a small prayer right now that such an event will never come to pass), and secure in the sense that it doesn't bow an inordinate or unpredictable amount due to the tension of the strings. (I will share my method of attaching the extension in a future blog post, so stay tuned.)

I've found that this deflection due to string tension is more of a concern the bigger the lute gets. The string tension itself doesn't change much whether the lute is big or small. Instead, the problem seems to be that the longer the bass strings, the more room one must allow between the strings at the extension end of the lute so that they don't clash against each other when the basses are played.

The amount extra that one must allow between the courses is minute--less than 1mm extra for each course--but when it's added together over the four courses of the extension, it becomes a relatively substantial amount.  The main problem is that the more space there needs to be between courses, the more the extension must skew outward, that is, must point toward the bass side. The strings pull to the side, as well as forward, meaning there could be problems if the extension isn't built robustly and attached securely.

But everything seems fine so far, and the lute's been under tension for over a year. I hope to see it again sometime soon--I'd love to take some measurements and see how much the extension has deflected during that time.

On to the second player, and the second lute. The player is Ronn McFarlane, one of the best-known and best-loved lute players on the scene. I've been listening to Ronn for years, and enjoying his concerts and the classes on lute technique that he's taught at the Lute Society of America Summer Festivals. So it goes without saying that I was thrilled when he contacted me to see if I'd be interested in making him a new lute.

Now, I knew that Ronn had been playing one main lute for a lot of years--a 10 course that Ray Nurse built him I think in the early 2000s--and that it had been a long time since he'd commissioned a new lute. And indeed, once we started discussing the design features that Ronn wanted in this new lute, it became clear to me that he had done a lot of thinking about the kind of instrument that he would need in his career from this point forward. The result for me were a lot of small but crucial challenges in the design and the building of the lute.

Let's have a look first--then I'll tell you about some of those challenges.
 
This is a 10 course lute as well--I expect that this number of courses gives Ronn the maximum flexibility he needs to use the lute in all the different playing situations he encounters as a professional lutenist.

One of the main stipulations he had when we were initially talking about the design of the lute was the string length--59cm. As you may know, that is a fairly short string length for a 10 course lute. (My usual model for a 10 course lute tuned to g', a shrunken Tieffenbrucher C45, comes in at around 63-64cm.) Now, it's easy to find a model of renaissance lute with a string length of around 59cm--the 1592 Venere and the Hieber lute both come to mind--but a 10 course lute would not work on these models, for a couple of reasons. First, those bodies are a little too compact to handle 10 courses--they don't really have the resonant capacity to deal with all that sound. Second, their necks have room for only 8 tied frets--and Ronn was adamant that this 10 course lute have room for 9 tied frets (and he wanted the ninth to "tie easily," he said.)

So, that sent me back to the drawing board. The solution I came up with for a lute with a relatively large body, relatively long neck, and relatively short string length, was a Sellas archlute--the same small liuto attiorbato that I had used for the first 12 course lute that I made for Evan Plommer, in 2012.

As you can see in the pics above, the body of this lute is quite broad and short--almost as if the folks in the Sellas workshop in the early 17th century had taken the outline of a small-bodied 7 or 8 course lute, and just inserted a spacer or wedge in the middle to broaden it (Robert Lundberg talks about this  design concept in his book Historical Lute Construction.) At the same time, as you can see in this side view, the back of the body is quite flattened--which enhances sound projection, but also, fortunately, makes the lute more comfortable to hold and play.

Even with this rather squat body, however, getting the ninth fret to tie easily was a bit of a trick--I had to raise the position of the bridge by a few millimeters, and change slightly the profile of the lute's back, right where the middle ribs meet the back of the neck, making them gather in a slightly steeper curve. I got it to work, though--as you see here, a most relaxed ninth tied fret.

One other important design consideration was the peg box. Ronn is very much a travelling musician, and often that means he needs to be able to take his lute into the cabin of an airplane as hand luggage--and the case needs to fit into the overhead bin. A 10 course lute of the usual design has a long peg box, and a deep case to fit it--too deep for Ronn's needs. The solution: make a shorter peg box by having a chanterelle tuner, a main peg box that carries 7 double courses, and a bass overrider that carries courses 9 and 10.

 

As you may have noticed, the overrider is not based on any particular historical design. The idea of basing an overrider on some ancient 13 course lute where you'd normally find such a thing (an Edlinger, say, or JC Hoffmann) seemed quite anachronistic to me, so I followed Ray Nurse's lead in the lute that he build for Ronn in 2000: I designed a functional (and hopefully attractive and natural-looking) piece of equipment that will get the job done with a minimum of fuss.

Two final features to mention: I installed a set of Pegheds, the geared, mechanical tuning pegs; and I installed a K&K Pure Classic pickup inside the lute. Both were Ronn's special requests, and I consented to them without question.  The Pegheds make a lot of sense for Ronn's situation: I cannot imagine anyone playing a lute more, and needing to tune a lute more (and more quickly and accurately) than he does. As for the pickup, Ronn plays many gigs in many different situations--solo, with the Baltimore Consort, or with Ayreheart, electric or acoustic as the case may be. The pickup allows him maximum flexibility to use this new lute in any situation.

So that's it for me for this week. I 've suddenly realized that throughout this blog post I've been typing the words "Ray Nurse" quite a lot. Who is this mystery man, you may be asking? Well, I'm sure many of you have met him, or perhaps seen or heard his work--he's been an amazing lute maker, musician and scholar for many years. This is what he looks like, standing in the doorway of my little shop. Ladies and Gentlemen: Ray Nurse!

New Lutes, February 2015

I have two new lutes to tell you about today, and honestly, I couldn't be more proud of them.  Both in their ways are new models for me, which is always a challenge, and to have them turn out as well as they have done is simply thrilling.

First up is a 12 course lute, for Lucas Harris, of Toronto.  In addition to being a conductor and very busy accompanist and teacher, Lucas is the plucked-string player for Tafelmusik, the well-known Toronto baroque ensemble.  Lucas got in touch after seeing the 12 course lute I made for Evan Plommer in 2012 (you can look at pics and a description of that lute here).  He was looking for a new lute that he could use in the many different musical situations he works in: accompanying singers, in ensemble, and as a soloist.  A real workhorse, as he said, a lute with a beautiful and powerful voice.

Well, I hope I've been able to make that for him.  I completed the lute three weeks ago and shipped it to him in Toronto; a week later, he took it with him on tour with Tafelmusik to Australia and New Zealand.  I haven't heard from him since he left (and here's hoping that no news is good news), but I'll expect a full report when he returns.

Built to Lucas's specifications, this lute has a string length of 65cm, and tunes at A=415.  There are eight tied frets, with the ninth a wooden fret glued just on the neck.  We debated a bit over which historical model would make the best body for this lute, and after considering things like overall size and shape, number of frets on the neck, and string length, we found that the C45 Tieffenbrucher archlute body fit the bill best.  Now, I've made a number of 10 course and 8 course lutes on a smaller C45 shape, but this body would have to be somewhat larger, and to make it, I needed to build a new mold.  I based it upon Grant Tomlinson's design for a slightly enlarged version of the C45 that he has used in his shop for many years.  Like the original C45, Grant's lute is made with 31 yew ribs; I designed mine to have 21 ribs, in homage to a couple of ancient lutes: the Hans Burkholzer (SAM 44 in the Vienna KHM) and the 13 course lute by Thomas Edlinger in the National Music Museum in Vermillion, SD (NMM 10214).

Here's how it looks:

   

The 21 ribs are, in this case, of Honduras rosewood, and the spacers are of holly.

I don't know about you, but I get a real kick out of the 12c extension.  It seems like such a slight little thing, and with four bass courses strung to it, there is a fair amount of tension on it.  All of this, and it's attached only to the tiniest corner of the bass side of the peg box.  This is the second of these extensions that I've done, and I've become quite confident of my design and construction of the thing. It's well-built and solidly attached, and even under tension, it hardly bows or moves at all.  Have a look:

Okay, a couple more photos of the whole lute--just a couple.

The second lute I recently completed is a 7 course for Doug Asherman of Oakland, CA.  Based on the smaller C45 mold that I've used for numerous 10 and 8 course lutes in recent years (you'll see many examples if you leaf through some of the older posts on this blog), this lute is a new model to me because I've never done a 7 course version of it before.   I was pretty sure it would turn out well, but I wasn't sure how well it would turn out... and I must say, I like it very well indeed.

The back is of yew wood, with sycamore spacers; the neck and peg box are veneered in ebony.
The string length is 62.5cm, and there are 9 tied frets on the neck.  So far, so much the same as I've done before; but this lute's a little different.  Besides being only 7 courses, this lute is the first I've built in a very long time to be completely strung in gut, and I think the results are absolutely spectacular. 

The trebles and octaves are gut from Boston Catlines; the basses are, on the 5th course, a copper-gimped, and on the 6th and 7th courses, silver-gimped, all from Gamut Strings.

I don't know if it's the gut, or the lute--it's a bit of both, I'm sure--but this lute has a warmth and clarity that I don't think I've quite heard before in one of my instruments.   The yew back adds a shimmer and quickness to the sound that just isn't there with other woods.  All of that, and at the same time the lute has a very big voice, very well balanced and free.  I think that sometimes yew, as a wood for the back of a lute like this, is the victim of a certain kind of prejudice: players tend to think of it as having a nice, rich sound, but one that's perhaps not quite as strong as a harder wood like rosewood or maple.  On the basis of this lute, and some others I've built, I think that prejudice is wholly undeserved.   Yew wood can have quite as strong a voice as either of those, and at the same time a richer and more complex one too.

For anyone who would like a look at some more pics of these instruments--and all the others I've built over the last few years--I would direct you to my flickr feed, here.  For now, I'll just add a couple more detail shots of this lute's back.  The yew wood that I used for this lute was just spectacular.  I'm constantly awed by its beauty.

What's up next?  More fun, and more adventures: a 10 course lute on this smaller C45 mold, and a 12 course bass lute.  I can't wait to get back to the shop.

Take care, and I'll talk to you again soon.

Because You Asked: Making a Lute Mold (6)

Hi again.  Today's the last installment in the series of how to make a solid lute mold.  We've built a big block of wood, carved it down into a shape, drawn rib lines and carved rib facets--all we have to do now is refine these lines and surfaces until we're satisfied. 

Simple enough to say it, but in practice I've found the refining to be a very long process.  It demands a lot of concentration, decision making, and intentional looking--from different perspectives, in different kinds of light.   It also requires you to mentally step away from the fact that you're sculpting a shape, and think, instead, of how bent wood ribs will actually fit together on the mold.  The fact is, you can stop work on the mold anytime you like, and make your mold any shape you want, but it doesn't necessarily follow that your mold will be easy, or even possible, to build on.  With that in mind, what I want to do today is give some techniques and hints to help you get a good-looking and practical shape for your mold.

The first thing to remember throughout this process is not to try to do everything at once.  Instead, think of working through drafts, or versions, of the mold.  What we have right now is a first draft (and a rather lumpy one at that.)  We'll do some shaping, remove a modest amount of material, then have a careful look at the whole.  We'll make some decisions, then go on to a second draft, where we'll do the same thing.  All the time, we're thinking of moving toward a shape--not just landing there, but moving toward it in careful steps.

In general, it's best to be conservative at this stage.   It's much easier to remove material from the mold a little at a time than it is to add material if you've gone too far.

Refining the Rib Lines
You'll remember that before we put in the rib lines, while we were still working on the overall shape, I suggested that it was natural and desirable to leave the surface of the mold slightly 'humped up' between cross sections.  I also said that eventually we would be bringing down these areas in a very controlled way--and that's what we'll do now.

The best way to see these humps between the cross sections is by looking at each rib line separately, not from the front or the side, but directly across the rib line--from above, with a strong light (your bench lamp) above you, and all other lights shut off.  You can also lay a dark cloth on your bench below the mold, to provide a contrast against which the rib line will stand out.

Sighting across the rib line, you'll see a series of humps between the cross sections.  This is good.  The idea now is to lower those humps gradually and create a smooth, continuous line.  To do this, use a sanding block (with, say, 180 grit paper) to gently take them down.  Be careful not to sand down your cross sections--these should be preserved.  Sand off only material between the cross section lines.

It's best to have a plan of attack when you're doing this work, rather than just randomly sanding away at various rib lines.  Here's how I proceed:  I mentally divide the mold into quadrants--treble and bass sides in front of and behind the largest cross section--and I begin by working with one front quadrant, then the other, and then one back quadrant, then the other.  I might work from the outside rib in toward the middle, then from the middle toward the outside rib.  However, even though I'm working on one rib line at a time, and one quadrant at a time, I'm always keeping an eye on the whole, trying to see how the rib lines are shaping up all over.  

Here's a look at the mold after we've had a go at all the rib lines.

You can see that where the rib lines have been sanded and flattened, the pencil line has been obliterated.  However, the tattooed pin holes are still clearly visible.  With your flexible curve, re-establish the pencil rib line; then as before, use your finger plane and curved scraper to remove material between the rib lines, to establish a good flat surface between them.

That's round one.   How's it looking?  A bit less bumpy, a bit more streamlined?  Good--we're making progress.

Now take some time to really look at how the mold is shaping up.  Look not only at the rib lines you've been working with, but also the facets you've carved between them.  Look at each one separately: does the shape curve gracefully, curling and flattening in an organic way?   Sight across each facet, from one rib line to another: are there bumps on one side that don't match the other?  In other words, will you be able to bend a strip of thin wood to a shape that will fit that facet closely?

Turn on the overhead light, then turn it off again.  Use your bench light to highlight certain spots; then back it off to get a good sense of the whole.  Walk to the far side of the room, then come close again.  Turn the mold around, look at it from above, behind, in front, below.  Take the mold over to the window, and look at it in natural light, both direct and indirect.  Think.  Brood.  Sigh.  Furrow your brow.  Scratch your chin.  With a pencil, and mark some spots where you might want to flatten the rib lines further.  Then take a deep breath, take up your sanding block, and go on to draft number two.

A Couple of Tools
A couple of tools that will make your life easier at this stage are a contour gauge and a sliding bevel.

The contour gauge allows you to take an accurate impression of a curved shape.  It's useful in two ways: first, it allows you to compare contours from one side of the mold to the other; second, it can sometimes show a hump or a bump more clearly than just looking at the rib line itself.  This is especially true if you're trying to get a smooth transition over one of your cross section points, where the rib line tends to hump up on each side.

(By the way, comparing rib lines from one side of the mold to the other likely won't give you an exact match.  That's to be expected, and it's not a problem, as long as the shapes are comparable, and good on each side.)

The second handy tool is the sliding bevel.  This tool allows you to look at the angle formed by the outer rib and the bottom of the mold.  That angle will obviously change as you go from the widest point forward, or back; the sliding bevel will help you judge whether the rib angle is changing gradually and gracefully.  It will also allow you to compare the rib angle from side to side on the mold.

The sliding bevel is perhaps most useful around the bottom of the mold, as you develop the shape of the capping strip.  Again, the angle changes as you move around the mold, but the angles should be similar from one side of the mold to the other.  

The Bottom of the Mold
The area back of the widest section is the most difficult part of the mold to shape.  All you have to guide you is the longitudinal section; every thing on either side of that is up for you to judge.  It helps if you have a good set of photographs of the original lute body you're working from, showing the bottom section from various angles, but you might have only a photo of the side view, and another of the whole back.  Or, you may not have any photos at all.  In any case, while it's true of the mold as a whole, it's especially true of this part of the mold: you are a sculptor, and you have to imagine this shape into being.

I can't say that much to guide you here, except to re-emphasize the principles I've given already--remove a little wood at a time, try to see the back area not in isolation but as part of the whole, and at the same time pay attention to individual facets, to judge whether you will actually be able to bend a rib into the shape you've made.  One other general principle is that of all the ribs, the center rib will likely be the one to have the sharpest bend at the bottom end.  All the other ribs will have a somewhat shallower bend, and they will likely become a little shallower the further they are away from the center.

The Capping Strip
Maybe the best way to help find a shape for the back end of the mold is by setting up the capping strip.  We've already put in a pin line to show its location (remember, we did this when we first set up the rib lines)--now, let's draw it in fully, and do some work to integrate it into the mold.

Once I've drawn in the location of the capping strip, I flatten the whole area, from the top of the capping strip to the very bottom edge of the mold.  Why do that?  Won't that spoil the nice rounded shape at the bottom of the mold?  Well, it will, but the fact is that once I actually get around to putting a back together on this mold, I'll need to flatten the ribs in that area anyway to create a gluing surface for the capping strip.  If I flatten the area on the mold first, then when I put the back together I can bend the tips of the ribs to follow that shape.  I'll be giving myself a head start in my flattening work, and I'll also save having to remove a lot of rib material to do it (which might otherwise thin out and possibly weaken the ribs in that spot.)

The more you refine the capping strip area, and the rib lines and facets leading up to it, the better idea you'll get of the shape of the whole bottom end.

Things are starting to look better here.  I can't remember how many drafts I went through on this particular mold to get it to this shape, but it was a few.  Each time I touched up the rib lines, and each time I flattened the facets between them, my work got more and more precise, and the mold got more and more refined.

Late-Stage Chores
As the overall shape evolves, and comes close to what you might hope is a finished shape, there are a few late items to pay attention to.

Keep refining the top block, but when the shape gets very close to finished, switch to working with a file (a #0 Grobet file works well), and file from the mold onto the top block.  Remove the block to have a look at how close you're getting to the marked cross section here--you should be very close indeed.

Another late-stage task is to scrape the facet of the center rib.  You'll remember I advised to leave that one alone for as long as possible, to preserve the longitudinal section you built into the mold.  At a certain point, though, hopefully when the mold is very close to its final shape, that longitudinal section needs to be scraped away, as the center rib takes its place as just another of the ribs on this lute mold.

While we're on the subject of scraping facets, another important late chore is to hollow-scrape the facets between all the rib lines.   Up until now we've been content to scrape them reasonably flat, for the purposes of carving a good overall shape.  However, we now have to take this scraping a step further, in anticipation of actually fitting a bent rib against the mold.

One characteristic of ribs bent with dry heat is that they tend to go somewhat concave across the width.  This makes for a really cool look on the finished instrument, but it can make for some difficulty in fitting and gluing ribs--unless you account for it on your mold.  That's what the hollow scraping is for.  The wider the rib you're working with, the more concave it's likely to go, but narrower rib facets--like the ones I'm working with here--need to be hollow scraped too.  Check with a straight edge all over the mold to make sure that they are.

Beyond the Theoretical
At this stage, your mold is very close to being finished--you've used the cross and long sections you built into the mold to carve yourself an elegant shape, with graceful, flowing lines.  But what if, after all this work, and all that careful attention you've paid, the lines aren't graceful, and the shape isn't elegant?  Well, the short answer is: you've got more work to do.

You may have gotten to a point where the practical shape of the thing you've made bumps up against the theoretical shape you're working with.  In other words, the cross sections you designed in two dimensions on your drawing just haven't translated well to three dimensions on the mold.  That's okay--it happens.  Usually, for me, the problem has been that I've ended up with a shape that's a bit too big and boxy on the front end of the mold (the area approximately behind the rose on the finished instrument.)  I'm not too upset if that happens, because it just means that I have a little more material to remove; it would be far worse if the area was designed too small, because once the material is gone, it's difficult to put back.

If this is the case for you, and you've got more material to remove, then just do it the way you've been doing it so far--a little at a time, working both sides of the mold evenly, keeping an eye on the flow of the ribs, as well as the overall shape.  You'll get there eventually.  I've found that when I'm working the area behind the rose, it's very easy to go too far, and to weaken it (I mean visually, not structurally), so do be careful.

The Finished Mold
The mold is finished when you declare it to be--when you're either satisfied with the shape, or you're so sick of working with it that you can't bear to look at it anymore.  (They might be the same thing.)  Here is the final shape of my mold.

I think this shape looks reasonable--reasonable enough, at least, for me to build a body on it, which is after all the only real way to tell if the mold is good or not.  If it builds a good-looking back, then I'll be satisfied; if it doesn't, then I'll work on it some more before building on it again.  Most of the molds in my shop have been works-in-progress; some have gone through numerous versions before reaching their final shape.  I'm looking for a shape that's practical, but also beautiful, and sometimes those qualities can be elusive.

You can see in the photos above that I've done a couple of things to finish up: I've inked the rib lines, and numbered the ribs.  You can number them consecutively if you want--I prefer to number the ribs separately on the treble and bass sides from the center rib out (i.e., 0 in the center, then T1, T2, T3, on the treble side, B1, B2, B3, on the bass, etc.)  I find this system helps me keep things straight as I'm working outward from the center rib when building the back.

Here's another hint--cut some strips of mylar, stick them down with double-sided tape, and trace the pattern of the rib.  A set of these rib patterns is very handy to have when you're laying out shapes on your rib material. 

One last thing to do is give the mold some kind of finish: two or three coats of a spray-on lacquer sanding sealer work well.  (Remember to remove the top block before you do it!)

You are now well and truly done your lute mold, and you have my congratulations in advance.  Take yourself out for a refreshing beverage of your choice--it's on me.  Don't overdo it though--you've got a lute to start building in the morning.

* * *

Post-Script 

And there you are, for those who asked (and those who didn't): this is how I make a solid wood lute mold.  I told you it was complicated!  Perhaps it's unnecessarily complicated; no doubt there are many ways you can find to do this more simply, or more efficiently, or more precisely.  I encourage you to experiment, and find the method that works best for you.  If you have the time, I'd like to hear from you--let me know what your experience is like using this method, or if you've come up with ways to improve the process (I'm a fan of those who betterize.)

I hope this is a help to lute makers, especially those who are starting out.  When I was building my first lutes, there wasn't much information around on how to make this kind of mold.  I don't think a solid wood mold is necessary for all kinds of lutes, but for some of the more complex shapes of certain lute bodies,  I think a solid wood mold is indispensable.

Still, while it's a very important part of the process, making a mold is really only a small part of the total work of making a lute--and after all, the whole lute is the true test.  The mold shown in the pictures I've used for this series is for a 12 course lute that I made in 2012.  How did it turn out?  You can judge for yourself--here's the lute's happy owner, Evan Plommer, favouring us with a tune, his own lovely setting of Walsingham.

Until we meet again...