Sunday 19 January 2020

Building a 13 Course Lute, Start to Finish, 13: Gluing in the Belly (After a Dozen Little Jobs get Done)

Hello friends, and welcome back.

In today's post I have a number of small and relatively unspectacular jobs to describe, all leading up to one big and quite spectacular job--gluing in the belly of this 13 course lute.

In our last episode, I glued and shaped the sitka spruce fingerboard spacer, and then made final action adjustments by slightly lowering the edge rib. Along the way, I checked the action by stretching out lengths of nylon fishing line from the bridge to the nut, using graduated wooden spacer blocks to measure string heights. Everything looked in good shape for gluing in the belly.

As I finished up the last post, I said I would have about a dozen little jobs to do before I could tackle the big one. That was just an estimate, not a precise count. So here's what I did--let's see how many there actually are.

1. Make a belly drawing.

This is a full-scale drawing of the belly done on a sheet of drafting paper. It's not a theoretical document, but a practical record of the belly as I've made it before gluing into the body. I first trace around the belly, and project the locations of the bars using a straightedge. Then I measure and record belly thicknesses and bar dimensions. I also include information such as the amount of curvature I have put into certain bars during gluing, the weight of the belly, and so on. On the right hand side, I've left space to write comments on the various thicknesses and characteristics of the materials I've used on the whole lute, such as the material and thickness of the ribs, the materials used for the bridge (and the weight of the bridge), and impressions of the soundboard gained through tapping and flexing throughout the joining and thicknessing processes. I also write down specific musical tones and overtones that I hear when tapping both the barred belly and the body, and any impressions I have about (as the case may be) the freedom, restriction, richness, complexity, loudness, balance, etc., of the sound of the belly as I tap it in various places, both on and off the bridge. This drawing is for my files, and I have one of every lute I've made.

Projecting the location of the bars.

Measuring and recording belly thicknesses.
2. Make and glue a label.

I have three lutes on the go, so I will need three labels. I use some medium-thickness watercolour paper, and a jar of black ink into which I dip a calligraphy pen, and just sit down and write labels for about 5 or 10 minutes. I am by no means a skilled calligrapher, so I just make a bunch and choose the ones that look best, that are clear and have a nice balance and energy. Nobody's going to spend too much time looking at my label, but I want it at least to be legible through the screen of the rose.

I make sure to stick it down at exactly the position of the rose.

3. Chamfer the front edge of the top block/fingerboard spacer.

This is a standard procedure I do for all lutes, and it is meant to assist a future repair person who might, at some point (far off in the lute's future, I hope) want to remove the belly. The chamfer allows a palette knife to slip more easily between the belly and the block/ spacer.

 Note that I don't extend the chamfer all the way over to the edge rib, but instead finish it a few millimetres away in a 'lark's tongue.'

4. Drill a pilot hole for a strap button.

I do this after the final action adjustment (i.e., after trimming the edge rib) so I can be sure to locate it a specific distance from the edge of the bowl. I make the hole big enough that I can fit a small reamer into it. (I'll fit the strap button after varnishing.)

5. Glue a paper patch over the top block/fingerboard spacer.

This is a piece of handmade paper that covers the top block/fingerboard spacer and extends a little way up the neck. The 'tongue' of the belly will eventually be glued down over it. During the life of the lute the belly will be securely glued, but should a repairer eventually need to remove the belly entirely, the layer of paper will make removal in this area much easier. I have had occasion to remove a belly that was not glued with this paper layer--that is, glued wood-to-wood--and it took a long time, was destructive to both the belly and the top block, and necessitated some careful rebuilding of the belly before re-gluing. It's much better to think ahead, and be kind to a future generation. (A good motto in all we do, in my opinion!)

I saturate the paper with glue and rub it into place--much like I did when I glued down the rose pattern, in in episode 6.

When the glue's dry, I feather out the edges of the paper with files and a small sanding block.

6. Make a caul to fit the top block/belly tongue.

With the paper layer in place and the edges feathered, the top block area is in its final shape. Now I can create a caul that will fit that area perfectly, and hold down the belly tongue securely when I glue it.

This is my top-block gluing rig, in two pieces. The top-block caul itself is on the right. It's a piece of jelutong that I've cut out to fit the area exactly, then carved to fit the countour of the fingerboard spacer/top block. I've covered it with a layer of cork, then added a strip of masking tape along the edges to make sure that it will press down the edges of the belly when I glue it in.

Here's what it looks like in a dry run: two wooden cam clamps supply ample pressure.

And a side view. 

I mentioned in the last episode that one of the things I needed to watch out for when shaping the top block and fingerboard spacer was whether the belly, with its arched upper braces, would fit closely, especially at the outer edges. Now I have my answer: the belly fits well.
Okay, I think I'm just about ready to go with gluing in the belly. Just a few more things to do before I start--less jobs to do than just a checklist of things that need to be in order.

7. Dehumidify the workshop.

This is part of my routine in the workshop everyday. Weather conditions are such that I need only keep my small space heater running on low to maintain a humidity in the low-40% range. Today it is at exactly 42%, which is perfect for the job I'm about to do.

8. Prepare the glue.

The night before gluing the belly, I soaked two normal batches of hide glue. They will both be ready for heating and use in the morning.

9. Give the bowl and belly a final brushing-out.

Just to ensure that no little bits of dust or wood shavings will be rattling around once the belly's glued in.

10. Tape the belly in place with masking tape, spaced approximately 2" apart.

The belly should fit exactly. I take one last look all the way around to make sure that the belly will fit down tight onto the edge rib. I also make sure (one last time) that the outline looks good, and that no ill-fitted bar ends are going to elbow out and make a bump in my beautiful shape.

11. Assemble all tools and materials on the work bench.

I will need: a soft pad (a folded camp blanket covered by a sheet of soft cloth) along the edge of the bench; the top block gluing caul assembly, including two wooden cam clamps; 1" masking tape dispenser, clamped to the bench edge; an electric iron, on a medium-high setting, along with a cotton cloth folded in four; hide glue warming in my glue pot; my glue brush and palette knife. I think I'm just about ready.

My glue, glue brush and palette knife. Note that I've cut off most of the knife handle, so that it's easier to use (and less likely to tip over my glue jar.) The glue, by the way, is very dilute: I've probably added twice as much water to it as I would for a normal batch of hide glue.  
And there is one more job to do before I can glue, bringing the number to a magic dozen:

12. Chose music to listen to.

This operation will take about a half an hour, so I need to choose the right tunes. Here's what seemed right today.

And so we begin. Most of the photos you will see below are from a staged series of dry-runs. Once I get started, it's nearly impossible for me to stop work to organize a timer shot, so these photos are the next-best thing.

Here's the view as I begin. I hold the lute in my lap, against the padded bench edge. The glue is close by, and just beside it, my tape dispenser is clamped.
Starting on the bass side, I locate the first bar above the rose, and from that point down toward the bottom of the belly, I lift the tapes. 

With my left thumb I lift the edge of the belly slightly, then dip the palette knife in the glue pot and get some glue into the seam. I work in about 4-to 6-inch segments; I don't want to slop a lot of glue in, only make sure that there is some on both sides of the joint, the belly and the rib edge. 
Here's another view of my dry-run demonstration. Once I have sufficient glue in the seam, I will pull down the tapes and reattach them, then add more pieces from the dispenser. 

As I work my way around the belly, I turn the lute in my lap, and hold it securely against the bench.
Here's a shot from the actual gluing process. I secure the glued area with a complete coverage of masking tape, but I leave the leading edge of the glued joint untaped so I can work a little glue back into it before moving forward.

Eventually I get all the way around the belly to the treble side, to the first bar above the rose (where I started gluing on the bass side.) Now I can lift up the belly tongue, and use the palette knife to spread glue under it and down each side to the point where I've already glued.
With the whole perimeter securely taped, I can now get my top-block caul in place and tighten up the clamps. I check for good fit and squeeze out around the top block area.
Finally, I run over the whole perimeter with the medium-hot iron on a folded cotton cloth. I don't want to press down with any force; I only want to apply a little extra heat to the glued joint.
And that's about it. It's pretty simple once all the prep work is done, and you get down to doing it. The belly is taped in place before the gluing begins, and you only work in small segments at a time, so there is no question of misalignment or movement.

Although it's a simple and secure process, there's a certain finality that I feel at this stage. It feels a bit like a capsule has been sealed, with all one's experience, knowledge and hopes for the future inside. So much preparation has been built-in to this point; all the work I will do from now on will be to bring it to fruition.

I'll see you again soon!

Thursday 2 January 2020

Building a 13 Course Lute, Start to Finish, 12: The Fingerboard Spacer; and, Preparing to Glue in the Belly

Hello, friends--I hope you're enjoying a fine holiday. I'm in the midst of travels these days, but I thought I'd put together a little treat to celebrate the new year.

As the title of this week's post indicates, I'll be dealing with two themes today--gluing and shaping the fingerboard spacer, and preparing to glue in the belly. They are very closely related, and really need to be done in tandem (and I think they're best described in tandem), since the shape of one essentially dictates the shape of the other. 

At the outset of this series, a reader posed a fundamental question that I've been coming back to ever since: how different is building the baroque lute from building the renaissance? My answer: very different, and though the differences are many they all find their source in the shape of the fingerboard. If you've been following the series, you'll know that at practically every step of the way I've been making allowances for and anticipating its arrival--I've taken account of it in checking the neck angle and developing action; I've arched the belly in subtle ways by gluing curved braces across it; and I've curved the bridge slightly, during both the making and gluing of the bridge, to match that belly arching. Now it's time to see how all that preparatory work pays off.

For those who may not know, one of the distinguishing characteristics of a baroque lute is its sometimes very highly curved fingerboard. The amount can vary, but I've designed this lute to have a fingerboard arch of around 4.5mm at the nut end, and around 2.5mm at the body-neck joint. Compare that to a renaissance lute, which, depending on how many courses it has and consequently how wide the neck is, might have an arch of around 1mm or less at the nut end, and about 0.5mm or less at the body-neck joint. On a renaissance lute, this curvature is achieved by gluing down a thick piece of ebony for a fingerboard--between 3 and 4mm thick--and planing the curve directly into it. However, this option is not available on a baroque lute, since for the curves I'm looking at, the piece of ebony would need to be somewhere around 7mm thick. The result would be a) a very heavy neck; b) a ton of work to shape the piece; and c) a terrible waste of a precious material.

The solution to the problem is to glue and shape a piece of strong, stable, light material on the top of the neck--a fingerboard spacer--and then, ultimately, cover that with relatively thin piece of ebony (as we'll see in a later episode, around 2mm thick--barely more than a rather thick veneer.) The material I use for this fb spacer is sitka spruce--which is strong, stable, light, readily available locally, and easy to work. Here are some process photos to show how I glue it on.

On the left is a caul that matches the twist I've planed into the neck's top surface (see post #8 in this series for an explanation of that.) The fingerboard spacer is on the right, and will be glued onto both the neck and the top block. It's a bit oversize in length as well as width, so I've masked off the upper portion of the body to keep glue from dripping down inside.

After lots of dry runs (where I figure out how many clamps to use, what sequence to use them, where to set them, etc.), I'll warm the parts up with a hair dryer, paint glue on both surfaces, and then put them together and clamp. With such a big gluing surface things will tend to swim around a bit during clamping, so I need to check frequently as I tighten to make sure all the pieces end up in the correct position.

I paint extra glue on the outside of the joints--as I've mentioned before, the glue shrinks as it dries, and will help to pull together the edges of the two pieces.

Careful planning--and lots of dry-run rehearsals--help me to achieve a nice orderly result.

Next day, I clear away excess glue on all edges, and then trim the piece. I begin by cutting off excess length at the top block and the nut.

Then I trim the sides--first with a knife...

...and then with my trusty low-angle block plane (and a lot of wax on the sole.)

I leave a small bit of excess along the sides, but trim the ends very closely.

Now with templates I can lay out the fingerboard profiles at the body end... 

...and at the nut end.

And then I can get to work putting a good shape on the thing.

It can be handy, when doing so, to chamfer down to the line with a sharp chisel.

As I near the shape, I start working with rasps, then files, then finally a sanding block.

Although the surface is curved in cross-section, it must be kept very close to flat in long section. As I'm working, I draw (and re-draw) lines that follow the layout of the courses in order to keep track of the flatness. 

The straightedge held on each line, with strong light behind, shows the developing profile clearly.
Here's what the edge of the shaped fingerboard spacer looks like (I put a piece of paper behind to show it more clearly.) The spruce is feathered out pretty precisely to meet the edge of the neck veneer.

There's the nut end profile (sorry for the lack of focus!)
And here is the profile at the edge of the top block. You will notice that at each side of the fingerboard spacer, the ends of the curved profile meet up, rather bluntly, with the flat top block surface. I'll need to ease the corners of the top block a little, to continue and blend in the curve gracefully.

In practice, that means I also need to ease (slightly) the edge of the outside rib. I use a sanding stick for this little job--just a strip of particle board with a piece of sandpaper stuck to the end.

Here's what it looks like without me in the way. It's long enough to use lengthwise as well as across the body, and it keeps the top surface of the edge rib pretty flat while I'm shaping the contour. 
This is the kind of shape I'm looking for--the curve of the fingerboard spacer extends to the outer edges of the top block, and that curve blends smoothly into the forward part of the edge rib.

It's time to see how the belly looks inside the body. You can see here how the arched braces I glued to the upper part of the belly  match the arching of the fingerboard spacer. So far, things look good.
Two other things I'll be paying close attention to are, first, how the belly fits into the curves that I've just shaped into the top block/ edge rib area, and second, what the action looks like. This is the very first time during construction that I will be able put on an actual string and stretch it from bridge to nut to see what the true string height above the fingerboard will be. It's a bit of an exciting moment!

Here's what it looks like. I've marked the string spacings at the nut, so I know exactly where to stretch my piece of nylon fishing line. I've also marked the positions of the first and eighth fret, which is where I want to put down wooden spacers to get a representation of the action of the lute as it is now (and how much I need to adjust it before I can glue in the belly.)

Another view. Notice two things in this photo: my collection of very accurately made wooden spacers (I've made them in sets of ten, graduated in tenths of a millimetre); and the leather collar on the tip of my index finger, around which I can wind and pull tight the nylon line without it cutting into my finger.
A quick word about the method used here to check the action. The spacer I use at the first fret represents the theoretical height of the string above the finished fingerboard, so when I stretch a nylon line between it and the bridge, I can slide a spacer under the string at the eighth fret to derive the exact string height. I measure the action this way for the first eight courses only. What I'm looking for--what I'm hoping for, actually, when I do the preliminary action check at this point--is an action that does two things: first, rises very gradually by a small amount from one course to the next; and, second, is overall very slightly higher than the action I am hoping for, before I glue in the belly.

My initial assessment of the action is that, as I hoped, overall the action is slightly too high. However, I've also noted that the action of the middle courses--roughly courses 2-7--is a little bit low, relative to the outer courses. The arch of the fingerboard spacer is a bit too high down at the lower end, so I'll need to remove the belly and do a bit of reshaping.

During the re-shaping process, it's handy to use a contour gauge to get a clear look at how sweet the fingerboard curve is. (I also use this gauge to 'take notes' on the developing arch, by tracing the shape into my notebook.)

I write notes and squiggles as I work, to indicate where I need to remove (or not remove) material.

Here's how the action ended up looking after I had finished up re-shaping the spacer. As you can see, the action rises gradually--by 0.1mm from one course to the next. This looks very good to me. 
The action across the string band looks good, but as I said above, it's all a little bit high--about 0.2mm. This is a good thing--I've been planning for and anticipating this all through the building process. It means that I'll be able to lower the action, in a very controlled way, to get it to the exact place I want before gluing in the soundboard.

In order to lower the action at the 8th fret by 0.2mm, I'll need to lower the edge rib by about 3 times that amount--somewhere around 0.5 or 0.6mm. It's hard to say exactly; all a person can do is start shaving it down, lowering it a little, then taping the belly back in, and checking the action.
To give a little visual guidance, I use a marking gauge with a sharpened pencil lead to mark a line about 1mm away from the edge, all the way around the bowl.

Then I get out my low-angle block plane and get to work. Most of the material I remove will come from around the location of the bridge--say, the bottom third of the body.

As I come up around the sides, I'm looking to blend in with the curve already planed in the edge rib (the 'belly scoop'--see episode 2), including the curves that I've just made on each side of the top block.

I use the sanding stick to blend in the work I've done with the block plane, and make nice, even transitions all the way around.

The only way to get a clear look at my progress is to tape the belly in place, and check the action. I will do this two or three times--removing some material from the edge rib, taping in the belly, and checking--before I get the action to the place I want. 

I want to make sure the action has come down equally on both sides, there is still a good belly scoop, and the curves and contours I've shaped into the edge rib are smoothly integrated. It's tough to show this in a photo; here's my best effort.

I think this is a good point to leave things for today. There are still about a dozen little jobs to do before I can glue in the belly; I'll show them all next time, and I'll show how I glue the belly in as well. This episode has not been the most spectacular or dramatic, perhaps, but I trust it's shown the kind of detail and concentration needed to bring together a lot of different strands of the building process. What I've shown in this post is essential work in creating a playable instrument.

Happy new year, everyone!