Friday, August 4, 2017

Schelle theorbo - Conclusion

My last post on the Schelle theorbo focused on the historical instrument and there are still a few points I want to make about it. But this time I want to show you photos of my replica.

The bowl before the finish was applied

I have always kept a few boards of bird's-eye maple on hand - it is plentiful in Canada. But a few years ago I picked up a special slab measuring 6cm x 27cm x 94cm. All of it loaded with bird's eyes.

   From this I was able to cut consecutive matching ribs.

The neck on the original is  a conifer wood veneered with ebony. At the request of my client I used English maple painted black. In this photo the bowl has been French polished.

My extension is made with one piece of quarter sawn poplar. The back of the extension is arched side to side for its entire length so the center is the thickest part. The thickness preserves the strength of the piece but tapering reduces its weight and makes it less clunky looking.


I copied the dimensions of the  string trough on the extension but omitted the treble string bracket.

The design of the theorbo head is unique and unlike the 17th century Venetian examples.

It still curls back on itself but, as the above photo shows, but the string trough is narrow and cramped like the arrangement on swan neck baroque lutes.

It is well documented that many 16th and early 17th century lutes were rebuilt as baroque lutes or that baroque lute makers favored the shape of the earlier models for their new lutes. Sebastian Schelle's baroque lutes are built in this style.  The same characteristics can be seen in the design of the bowl of his theorbo.

Sebastian Schelle 1728
The Schelle theorbo bowl seen in profile is strikingly similar to those made by Laux Maler in the second half of the sixteenth century.

Laux Maler ca.1550

 This is the Laux Maler, E. 2005.3.1 in Musée de la musique, Paris.

The Maler bowl sits lower because the two edge ribs have been cut down making them less wide than the adjacent ribs. Also the lute's top was removed at this time. Note how, on both lutes, the back profile sweeps toward the neck in a gentle curve without an abrupt bend at the front block. The rear of the bowl bows out slightly rather than ending in a perpendicular to the lute's face. The contour of the rib joints are striking similar to one another.

Compare two other Maler lutes by clicking here and draw your own conclusions about the similarities. Both are in the Germanisches National Museum Nüremberg as companions with the Schelle theorbo. Bye the way,  one of them MI 619, was rebuilt as a baroque lute in the Schelle/Widhalm shop, ca. 1740.

A video that includes Daniel Swenberg playing my Schelle theorbo in the Bach at One Series can be seen at  Trinity Church Wall Street


All photos by the author unless otherwise noted.




Thursday, June 1, 2017

Sebastian Schelle Theorbo

    At the beginning of April I finished a model of the large bodied late baroque theorbo by Sebastian Schelle, 1728. Back in 2001 I built my first Schelle theorbo. That instrument was quickly followed by orders for a half dozen others. At the time I found this surprising because the Schelle is so different from other theorbo models. It isn't a flat back multi-rib triple rose striped yew job like the Italian instruments built a hundred years earlier. It is huge. The bowl is 11 wide ribs of bird's eye maple. The belly of the original measures 64.7 centimeters long x 40.2 wide while the bowl is 19.4 deep. The string disposition is 8x1, 7x2. The string lengths are 160 cm and 88 with10 tied frets on the neck. My model's string disposition is 8x1, 7x2 at 140 cm and a 9 fret neck with a fretted length of 86cm. The bridge holes can also accommodate 7 fretted singles, but more on that later.
A technical drawing of the instrument MI 574 is available from  German National Museum Nüremburg .

German National Museum Restoration Photo

After I had built the first two or three models I had the opportunity to visit Nüremburg and examine the lute.
Even though the technical drawing and descriptive notes are clear and thorough there were many features I wanted to view first hand.

The technical drawing and my photo show that the bowl is lop-sided. The second rib on the photo's left (lute's treble side) leans in while the corresponding rib on the opposite side has a straighter contour. The effect is particularly noticeable by the third rib on each side of the bowl. The measurable difference, comparing the two arcs, is 5-7mm.

Occasionally, I will incorporate features like this in my model since they obviously occur as part of the historical building process. But the difference was too large and I decided to make my mold symmetrical by using the fuller bass side contour (photo's right side).

The technical drawing provides six cross sections of the bowl. Having chosen to replicate the bass side I laid out the half sections along a central plywood spine.

Here I am using both clear plastic tape and duct tape to hold the rib joints together. I like the advantages of using clear plastic packing tape: I have a clear view of the joint, the tape is ultra thin so, with an open mold, I can feel the joint between my thumb and fore-finger and correct a mis-alignment if necessary.  Plastic tape however is not strong and in this situation I want the extra strength that duct tape offers.

Here's why. I want the ribs to develop a pronounced scallop effect across their width. This occurs naturally when the joints are papered over on the inside of the bowl. But it is possible to enhance this feature. I achieve this by chamfering the bottom edge of each ribs so that when they are assembled the joint is like an inverted 'V'. Closed and tight fitting at the top (the surface of the bowl) but open at the bottom or inside of the bowl. When the joints inside the bowl are papered over the drying glue and shrinking paper pulls the bottom edges together and forces the two sides of the inverted V upward, bowing the surrounding wood  and creating a more pronounced scallop.

The reflection of light on the rib joints accentuates this effect on the original lute.

The rose is set into the soundboard and is made of a wood other than spruce, perhaps basswood. This is noted on the technical drawing.

Apparently Schelle followed the traditional method in preparing to cut the rose. Once the blank disc of basswood (?) was inserted into the soundboard the rose pattern was glued on the inside covering the joint The pattern was then cut from the inside out. This is visible on a restoration photo of the inside of the soundboard showing that the paper pattern covered the joint of the soundboard and inserted rose.

Schelle's rose is not precisely cut nor is the drawing precisely rendered. The design is somewhat fanciful. I photocopied the rose pattern  from the technical drawing and using a light screen made a full size tracing, altering details as I went. I did not follow Sebastian's example and I cut my rose directly from the table but I did not carve a border.

The bridge is located close to the rear of the soundboard but its original position can be seen just forward of that. Apparently the bridge was relocated rather than re-glued in the original spot because the wood was too badly damaged. I do not know if this was done during restoration or at an earlier date.

There is little doubt that the Schelle bridge is original as the shape of the bridge tip matches the footprint of the original position on the right side.

   My client wanted a slightly wider spacing between courses and extra string holes so the lute could be single or double strung.  This increased the width of the bridge by more than a centimeter so I diminished the size of the bridge tips to help compensate.

The pegbox has 14 pegs for 7 courses although in this photo of the original lute the nut is grooved for a single first and the first bass peg does not carry a string.

The treble string is mounted in a bracket attached to the side of the pegbox. This style of side bracket is often a feature of swan neck baroque lutes as it allows the thinnest string to run straight over the nut rather than angling around the pegbox cheek and on to its peg.  It does not however alleviate he problem of  the build-up of string windings that jam into the narrow opening. Here a frustrated lutenists has chopped away the inner part of the pegbox.

For reference, here's a closeup of a similar bracket on the triple nut extension of a Jauch that I built several years ago. This bracket was constructed as part of the pegbox cheek and a rectangular slot was cut through for the string. You can see that it allows little room for string windings.

The Schelle theorbo is probably best known  for this feature - the neck  can be folded back on itself to allow for easy transport. The iron hinge is the center of a simple mechanism. The theorbo extension is in two parts connected by a simple half lap joint.

Note that this series of photos show the theorbo extension upside down.

Here is a closeup and you can see the half lap. The hinge pin sits toward the bowl.  Each half of the two part extension is notched so that one over laps the other. The hinge swings back on itself from right to left.

I photographed this while the lute was upside down on the table

The theorbo head is nearly 18 cm long so there is plenty of room for the eight diapasons.
The extension is constructed with two types of wood and although both are painted black the quality of finish is different.

For example, the peg box has a smooth satin finish and in areas where it is worn a nice patina has developed. In contrast, the finish on the theorbo head and its half of the extension is dull and rough. There are tool marks on the sides of the head.

Obviously I am fascinated with the historical theorbo and it has been my pleasure to share these photos from my time in the museum. I had intended to show more photos of my Schelle and I will.
There is more to say about this instrument too so I'll put that all together in my next post.
Also, the new Voboam and Stauffer are about finished so there will be stories about those.

All photos by the author except as noted.


Sunday, April 23, 2017

April Update

The last several months has been a busy time: After I finished a model of the 1690 Voboam guitar that my client picked up at the end of January I started another one but with changes in the aesthetic and internal construction. I am nearly finished with a new Stauffer based on an early (1822) "Legnani" model. And I finished a model of the large, powerful German theorbo by Sebastian Schelle 1728.

Long Island Guitar Festival photo
Early in April I attended the Long Island Guitar Festival at LIU Post. Harris Becker, the festival director and owner of my 7-string Stauffer, invited me to show my guitars. Raphaella Smits was one evening's featured performer in a program of Ponce, Mangoré and Mompou brilliantly played on a 1980 John Gilbert. Raphaella performs regularly on 19th century guitars so this was an opportunity to ask her to critique my work. We met after her master class with Harris and guitarist Huy Thanh Nguyen. Raphaella was generous with her time, thoughtful and direct in her comments. There were many take aways; nut width, string tension and string choice among the practical.

The inspirational was the gift of her CD of Antonio Jimenez Manjon played on a 1899 Vicente Arias. The music is at times "dreamy" but also "rough and virile" to quote adjectives Raphaella uses to describe his music. The guitar is very lightly constructed with a long string length of 72 centimeters, Its tonal characteristics, to my ear, embody the period's spirit: vitality and sentimentality. The guitar's uniqueness beckons adventurous luthiers.

My 1822 Stauffer (the neck is in the background) has progressed since this photo was taken. The purfling is finished and the body is just about ready for French polishing.

I had built several examples of this large theorbo, using the museum plans, before I examined the original in Nüremburg in 2005.  Earlier I thought there were several odd features in the instrument's construction. These were clarified when I studied the original. My post will include material from that visit. The string lengths of my model is 86 and 140 cm while the bowl is 41 by 65 cm.

The baroque guitar  that I finished in January was my usual 1690 Voboam model . But I made several changes in the aesthetics and construction features . My newest Voboam will have different features too. I'll compare my work on those two guitars  and explains my thinking about them in a future post.

All photos by the author except as noted.


Thursday, March 2, 2017

A New J. A. Stauffer

Early last summer I finished a Johann Anton Stauffer 7 string with an adjustable neck and a René Lacôte 7 in time for several instrument exhibitions. The story about this is  July Update - a new Stauffer 7.  At the time I was working flat out and didn't take enough photos to fill a post about the Stauffer. Now I have finished a six string version of the same guitar with plenty of photos so I can show you the similarities and differences in the construction of the two versions. Also, I described building a Stauffer terz guitar in three posts in November and December 2015. Since the three guitars share the same major construction features I omitted many steps in this post. Consult the Terz posts to see these steps.

Building a Stauffer Terz Guitar

Part 2

a Photo album

 I have been surprized to find very few structural differences between  6 and 7 string guitars in my investigations. I'll describe those shortly. The obvious feature is the suspended 7th string. So let's start with that.

The 7th string is mounted on a simple appendage that mirrors the figure 8 design of the peg head.  The peghead of the original guitar was clumsy because the neck was made unusually wide probably to accommodate the wishes of a client.  I don't think this was done to strengthen the neck because Stauffer eight string guitars built around the same time had standard necks. Since I intended to use a standard nut spacing decided to design a new peghead. I wanted to position the nut on the new peghead so I could use the same diameter of string as the 6th  with both at the same tension. Remember, a string will remain at the same tension if the pitch and length are in the correct ratio.


Harris Becker photo

The usual tuning for the 7th string is D, a full step below the 6th string. I did the math and found the string length that was necessary to make the guitar's regular nut  (640mm) equivalent to the second fret of the 7th string. The answer  is 718mm or 78mm from the regular nut. This placement allows the 7th string to be the same diameter and tension as the 6th.

The  7th string nut is a piece of ebony 4mm thick, 13mm wide and about 17mm high. It is embedded in the face of the peghead and stands upright.  The design requirements are fairly tight. The 7th string peg can't be too close to the nut or it will make tuning difficult and it must be clear of the other pegs so as not to interfere with them. Everything has to fit somewhat elegantly in a limited space.

This is the template I designed. It is laid out on a half inch grid. The seven peg holes and 7th string nut positions are noted.

Here is an the interior of the Stauffer 7 just before the back was glued on. I used a barring pattern with three bars beneath the rose. I don't know how the original guitar is barred but both Stauffer's eight and six string guitars use this pattern so it must be appropriate for a 7.
I did make several structural allowances for the additional string.  The front block is wider on the 7 to counteract the lateral pull of the off-set 7th string. It measures 90mm x 43mm as compared to 75mm x 43mm for my 6 string version. The bridge plate is extended to the bass side to support the wider bridge. The tail block for the 7 is a little wider to provide more support for the pull from the wider bridge. 

The bridge is made a little wider to accommodate the seventh string. The spacing between the 6th and 7th string is also a few millimeters wider than the spacing between teach of he other 6. Nineteenth century makers used a wider separation than this but discussions with players has led me to adopt this solution. 

Also I  made a few changes in how I construct certain parts of the guitar. Until this last guitar I cut the recess for the neck in the front block before I glued the block in place. I worried this would lead to a mistake in alignment that would cause a lot of trouble later on. I devised a complicated system to check the alignment of the parts before they were installed. This was silly and way too much work. 

Now I assemble the guitar body to the point where every part is done except for gluing on the back. This procedure allows me to lay out accurate cut lines for the neck recess on the soundboard and the front of the sides.

After making the preliminary cuts I clean the waste with a chisel and finish the sides of the recess with various files and sanding sticks. The back of the recess should be square with the top of the guitar and this is easy to check with a small try square. The work goes quickly and accurately.

At this time I also drill the holes for the neck adjustment mechanism. In this photo the large clamp is securing a wood block that helps to align the drill bit for drilling through the heel. When this is finished I can align the neck in the front block recess, hold the neck to the proper angle and drill on through into the block. The anchor nut for the screw is then secured in the block and finally the back can be glued on.

Sue found this old photo of me cutting a flitch of English sycamore into back and side rib lengths for 19th century guitars. I'm still using the same wood.

All photo by the author unless otherwise noted.


Wednesday, December 28, 2016

Gaetano Guadagnini Restoration - The Conclusion

Here is a photo of the Gaetano Guadagnini 1831 guitar after I finished its restoration. In my last post, after a brief introduction and a description of its condition, I used nearly the entire post to demonstrate and explain the internal blocking and shimming of the numerous cracks in the soundboard. Since there were so many cracks I couldn't help but wonder whether the guitar would regain anything of its former character. My goal was to return this interesting and important instrument by a member of the famous Guadagnini family to its rightful place on the concert stage in the hands of a fine musician.

The bridge is not only a prominent feature of the guitar it is also an integral part of the barring system of the soundboard.  The bottom of the bridge is contoured to fit the high arch of the soundboard and its unusual size (225mm x 14mm x 20mm) stabilizes the soundboard arch and moderates the soundboard vibrations in the same way that an internal bar does. I needed to take particular care in re-gluing it.

Clamping the bridge was going to be difficult because  pressure would be exerted unevenly on on the front half of the bridge leaving the important rear gluing surface unsecured. I devised a caul to solve this problem. Bridge clamps placed through the sound hole could now be centered on the caul. Others clamps were placed on the bridge wings. An arched  caul provided necessary support internally. Clamping pressure was exerted equally across the length of the bridge ensuring a good bond.

Now I could turn my attention to the back. I had taken the three bars off the back and those had to be carefully replaced in exactly their former location. I had previously marked their location so that went smoothly.

Even though I had retained the guitar in a collar (see previous post) in an attempt to keep the contour of the side ribs stationary they splayed out slightly. When I tried to align the back it no longer fit. I wasn't completely surprised. The bass side rib bowed in slightly at the waist but out at the upper bout - a perplexing and difficult situation. I wasn't convinced that normal clamping pressure would hold everything in the correct alignment so I made a special clamping jig that spanned the width of the upper bout.

On the treble side upper bout it fit flat against the side rib and was held in place with a vertical screw clamp.

The bass side of the clamp was made to exert pressure near the gluing joint. This arrangement pushed the side rib into alignment while the neighboring clamps held the joint together.

You'll notice that the wooden clamp feet point in different directions. I made the clamps this way on purpose. The feet fit more or less loosely on the screw rods allowing each foot to conform to the any arched surface. Turned sideways the foot is more rigid and less of its surface is in contact with the subject.

It was clear to me the gluing operation would take a lot of time. Too much really so I decided to do the work in stages. First, I clamped everything in position without glue. Then starting at the neck joint - this ensured that the proper neck angle would be preserved - I released a few clamps and inserted hot hide glue with a brush, knife or artist spatula.  Once the clamps were re-applied I moved along to the next area, one side then the other. I treated the upper bout, waist and lower bout as separate areas. The technique worked almost perfectly. There were several minor areas where the back overhung the ribs. Whether the back or sides changed shape or shifted or I erred is difficult to know - probably a bit of each.

Originally, the peghead was fitted with wood tuning pegs but these had been replaced at an earlier date with mechanical tuners as shown.
The peghead angle is 20°.
I removed the metal tuners and plugged their  holes as shown. The holes for the posts of the metal tuners were different than Guadagnini's wood pegs so there was some over-lapping of holes which I thought might cause problems. To avoid this I laid out a slightly different peg spacing in order to avoid drilling through an area that was already weakened by over-lapping plugs.  Also, the gears on the metal tuners were mounted with protruding lugs that left deep ugly holes in the peghead wood. I filled these with a mixture of fine ebony filings and cold fish glue.

I sanded the rear of the peghead smooth and refinished it with shellac dissolved alcohol blackened with aniline dye. The end grain of the wood plugs resisted the new finish and many applications of the shellac solution were required. I used a chisel tipped artist brush to build up enough thickness of polish that could be levelled. Then I switched to a regular French polishing technique to finish the job.

The photo is of the finished peghead. The tuners are a planetary type peg.

Although I had finished all of the major restoration work there were a few things to tidy up. Two of the inlaid ivory frets were broken with fragments missing (the 9th and 11th), so I decided to replace these with ones I made using bone saddle material. I also levelled the frets and re-crowned them using typical fretting tools.

I did not clean the top before I started working on the cracks because doing so would have forced grime into the open wood, staining the edges of the cracks and making the subsequent shims even more obvious. Cleaning the top was part of my plan to blend the lighter color of the shims with the original color of the top. I dampened a new shop rag with water as hot as my bare hands could tolerate, wrung it out quickly but throughly, and wiped it over the top. Then I burnished the top dry with a wool polishing cloth. Since the wood I used for the shims had darkened with age this final procedure produced a good looking aged patina.

Note* I would not have used such hot water had I not been sure that all parts of the top were sound and secure.

After I finished re-gluing the back a few areas overhung the side ribs as previously mentioned. Levelling these with the side ribs removed the original finish. In this photo the finish had worn off the bass bout through years of playing. Since that was part of the guitar's story I left that alone as well as other wear marks. Closer to the neck was an area that my work disturbed. I renewed the finish by making a thick glaze that could be brushed on the bare wood in one coat.

Glaze was applied to the edge of the back in this area too

I mixed about a 3mm length of burnt sienna oil pigment squeezed from the tube with an equal amount by volume of spar oil varnish and several drops of cobalt dryer. I brushed this on with as little overbrushing as possible and allowed it to dry for several days. Then I sanded out any heavy brush marks with 600 paper and applied a thin wash of a shellac solution.

That finishes my description of the restoration of this splendid guitar. Here are two final views.

All photos by the author.