Wednesday, December 7, 2016

Gaetano Guadagnini 1831 - A Restoration

I had no previous experience with guitars built by Gaetano Guadagnini II (1805 - 1852) so when Lucas Harris invited me to accompany him to look at an 1831 Guadagnini that was for sale I didn't know what to expect. This photo, taken that day, shows the guitar in nearly as built condition. Only the wooden tuning pegs had been replaced by rear mounted mechanical tuners.  I always thought ivory tuning pegs were archaic.  But here they are, a standard feature on a fairly plain guitar. There was much evidence the guitar had been played for years, but also signs that it had recently been neglected. A few cracks in the top had been repaired - there is a dated repair signature on the underside of the top - but, unfortunately, many more cracks had developed. The top appeared not to have been finished originally in any way and consequently there were several spots that were heavily discolored, I suppose, from contact with the player.  There was also a thick layer of grime.

String length: 640mm
Body length: 460mm
Upper bout: 275mm
Waist: 200mm
Lower bout: 335mm
Sound hole
diameter: 90mm

The maple back is in excellent condition and is finished in a vibrant deep brown-red that is most likely a shellac based spirit varnish. The finish has worn away from the edges of the contour and most notably from the bass side upper bout in a manner that suggests long use by a single individual.  The neck and peg head are ebonized  maple. Note the wear mark from the player's thumb.

The top and back are both strongly arched measuring 7 mm at the lower bout. This degree of arch is maintained at the waist and upper bout. The back also slopes from the tail to the neck joint making the tail block deeper than the front block. Strongly arched backs are not unusual but such an arch on the top is a unique feature. The construction of the arch is also unusual. The soundboard is not domed like Spanish guitars,  but bent side to side from the center line along the entire length of the top. The neck which sits flush with the fingerboard continues along the same plane.

The side ribs are contoured, top and bottom , to accommodate this arching.

The rib height including the top and back thickness :
Neck joint 65mm
Upper bout 63mm
Waist 72mm
Lower bout 69mm
Tail 85mm

Unfortunately these features do not show up readily in the photos.

Let's Get to Work!
Here is a close up of the top showing the major cracks. Others are too small to show up in this photo, but all needed attention.  This work was the focus of my restoration.

Several cracks ran under the bridge and you can see an example under the right arm of the bridge. I thought it was necessary to remove the bridge in order to do a proper repair.  There was already a slight lifting of the bridge at one point so it was easy to slip a knife in the opening. I dabbed alcohol on the knife blade and allowed it to wick into the opening. Alcohol crystallizes old hide glue, weakens the bond and allows the joint to be separated safely.

Older cracks had been filled with shims and several were secured with interior wood cleats. Since there were many cracks to repair it was necessary to remove the back of the guitar. I found an opening between the back / rib joint and wicked alcohol into the joint like I did for removing the bridge.

The back came off fairly easily although there were several difficult moments in the area around the bar ends where they let into the lining. Several bars became loose. Once I had successfully removed the back I finished by removing all three bars.

After one hundred and eighty-four years the back retained its arch free of its bars. I knew it would be some time before I re-assembled the guitar so I made a arched pine form to hold the back. You can see it in the background. The side ribs of any style of guitar have the tendency to change shape whenever the top or back is removed so I made a collar to hold the guitar while it was dis-assembled.

There were several areas where the lining was loose or broken and another location where a bar end had punctured the side rib.

The guitar does not have maple binding around the back. The finish has uniformly worn off the edge of the guitar  exposing the light colored maple.

 I took care of these problems immediately. A simple fix with a spring clamp and a piece of  cork was sufficient in some cases but a two part contoured caul and c-clamps were necessary for others.

It is interesting to note evidence that this guitar was built quickly: the notches in the linings for the harmonic bars ends are too large and the lining for the top is cracked at the waist but not broken through.

This photo shows the external collar that I made to help retain the contour of the guitar once its back was removed. The harmonic barring is a simple pattern as is the shape of the interior blocks. There is a repair inscription between wood cleats in the lower bout.

The harmonic bars are spruce with the grain standing up. They measure 9mm thick and 15mm high and are arched as I previously described.

 Is the date 1928? The cleats from the earlier repair were well done and secure so I left them in place even though I was going to replace all of the old shims. My policy on securing cracks is to use many small cleats with the spruce grain perpendicular to the crack rather than fewer larger ones. Had I used larger cleats this guitar would have had a lot of extra wood in its top which I think is detrimental to the tone. The V is a directional arrow and is an aid to me in placing the cleats.   I clamped each cleat individually using sound hole clamps and plexi-glass cauls. Using this procedure stabilizes the area of the soundboard and insures that the two sides of the crack remain level.  In this photo the cleats are in their rough shape. I left them that way while I fitted and glued the shims. Then I went back and feathered their edges into a smooth low contour.

This photo shows several types of cracks and my treatment for each. Below is a photo of this area of the guitar before I started work on it.

Rather than trying to make a tiny shim to fill a narrow crack I enlarge its width to about .75mm and make the edges parallel. Most cracks occur in the light colored soft wood of the annual rings and are usually fairly straight. The thin dark lines are denser and denote the late summer growth. These harder grain lines serve as a convenient boundary. Not only are they a visual aid but their hardness acts as a 'fence' to guide my knife. Usually I work free hand. There were several wider cracks or cracks that crossed an annual grain line. Because of their width I treat these differently. I use a metal ruler to scribe a pair of parallel lines that are wide enough to encompass the crack. I then gouge out a channel with a veining tool to a depth that is about 2/3 the thickness of the top. This method provides a 'floor' for the shim to rest on and a third gluing surface that provides extra strength for the joint.

I choose the wood for my shims from an old stock of very close grain soundboard scraps and try to match the color of the original soundboard. I can cut very accurate and clean edged shims with my miniature table saw. I bevel one edge of the shim slightly so that it slips more easily into its channel. It needs a good fit but not a tight one. The shim will swell when the glue is applied and it may not fit in the channel. A certain amount of fitting is usually required. I widen the channel a little or narrow the shim a little. When I'm satisfied with the fit I work quickly, brushing a thinned mixture of hot hide glue into the channel and pressing the shim home. I wipe away the excess glue with a hot damp rag, followed by a dry rag and then cover the repair with painter's masking tape to slow the glue's drying which allows the wood to swell a little resulting in a very snug fit.  Everything takes time.

There is still a lot to do, but that's enough for now.

All photos by the author

Tuesday, November 15, 2016

November Update - Guadagnini Stauffer Voboam

Lucas Harris with his 1831 Gaetano Guadignani

My shop was shut down for three months from July to early in October for renovations to our house. Once I got back to work I've been busy.

I finished the restoration work on an 1831 Gaetano Guadagnini that I started several years ago. I knew this was going to be a challenging, long term project. There were numerous  cracks in the soundboard - some that had been shimmed and more recent ones.  All needed attention.

The edge of one side rib had received a nasty bump, cracking the rib and fracturing the interior lining. One back bar was split and another was loose. Mechanical tuners had replaced the original wooden pegs damaging the peg head and naturally there were many unexpected surprises.
I'll report on my restoration in detail in the next week or two.

I have added a search function to this blog. The search box is located in the sidebar between the Blog Archive and About Me. Now you can quickly find previous posts. Type in an instrument name or an historical maker's name. All of my posts associated with your entry will display in the order of your choosing.

Screen Grab from the catalogue of the Gardiner-Houlgate auction house.

Carlos Agonzalez, referring to my posts on building a Martin Kaiser theorbo (late 2011), sent me this link  click here to the auction house that was selling a theorbo similar to the Kaiser theorbo  E.24 in Musée de la musique, Paris. Without making an attribution to Martin Kaiser Gardiner-Houlgate points out the similarities. Both lutes have the same build date of 1609. The veneered design on the back of both necks is an arrangement of ivory and ebony chevrons.  Furthermore, the string lengths of both lutes are similar, especially the diapasons. The catalogue photos are not precise enough to see the subtle construction features of the theorbo extension but there seems to me to be a general agreement in design. Also, although the bowl of the Gardiner-Houlgate lute does not appear to be flattened like the Paris instrument, the way the ribs come together over the rear of the bowl seem to have been given the same treatment.
Photos of E.24, the Paris Kaiser, are available here . Type Martin Kaiser in the search box.

Thank you Carlos for bringing this item to our attention. This is an interesting lute that might otherwise receive little notice.

I've made progress on the Stauffer model that I described in my last post. It will be finished soon and I will describe its construction soon.

And there is a new Voboam underway.

Here are some of the positive benefits from the renovations to my shop space. The bandsaw sits in a formerly unusable area that was a hallway. A wall was removed and I was able to spread out the power equipment in the formerly cramped machine room.

I re-organized much of my shop including cleaning out a neglected area under a stairwell, installing panelling and mounting shelves. Now all of the incidentals that cluttered my benches have a home.

All photos by the author except as noted.

Thursday, July 21, 2016

July Update - a new Stauffer 7

This photo was taken by long time friend and aspiring lute builder, Bill Good. I was exhibiting at Boston Guitar Fest in the middle of June. On the table is my 7 string Lacôte. It was the subject of my last post. And getting a trial, my newest guitar, a 7 string Stauffer.

Many of my lute and baroque guitar clients have 6 string romantic guitars so my plan was to built two 7 string guitars; a Lacôte and a Stauffer and to exhibit them, side by side, at various festivals and private gatherings in the coming months. The Lacôte was ready in April in time for the 60th anniversary celebration of the Toronto Guitar Society. In early June I finished a model of a 7 string Stauffer (photo at left).

I try to build instruments that are faithful to surviving historical  models and often I can use examples that I have personally examined. However, 7string Stauffer guitars are rare and I know of none in public collections. Late last year I bought a copy of Stauffer & Co. by Erik Pierre Hofmann, Pascal Mougin and Stefan Hackl, Les Éditions des Robins. The book (shown below) is an impressive survey of the work of Johann Georg Stauffer, his son Johann Anton and other builders who worked in the  Viennese style from the early 19th century through to Ludwig Reisinger in the 1930s. Importantly, the cultural and musical milieu of Vienna of the era is discussed at length.

Sixty guitars are presented side by side in three photos; face, back and side. Each guitar is documented and its significance is discussed in English, French and German. The book is large, measuring 38cm x 31cm. The text and photos are conveniently placed on facing pages. Unlike many sources that list only the height and width of their subject, these authors include external measurements, 20 in all, for the 60 guitars on a two page appendix that immediately follows the last guitar entry .

I was delighted to find among the entries a 7 string guitar built by Johann Anton in 1827 with a string length of 64cm.  The guitar is a beautiful example of Stauffer's, design, choice of materials and craftmanship. This information provided not only a solid factual foundation but also the inspiration for my replica. I filled in details using my research on Stauffer and other Viennese guitars that I have studied in public and private collections.

After the Boston Festival I visited clients in New York City area for a guitar tasting. Daniel Swenberg watches as David Thompson tries a Panormo replica.

I was re-united with some of my earlier guitars. From the left; the new 7 string Lacôte 2016,  a 61 cm  Johann Anton Stauffer from 2004, an original La Joue 1820, a 1787 model Benedid that I built as a 6 course in 2001 and recently converted to a single 6 and a Panormo  from 2014.

I didn't take enough photos of the construction of my 7 string Stauffer  to dedicate a post to it but I have started a 6 string version of the same guitar. In the future I will publish a post or two about it including the details that are relevant to the construction of a 7 string.

Last week we started renovations on our house that affected my shop. My work has come to a stand still but the renos are going well and I'm hopeful, if impatient.



Wednesday, May 18, 2016

Building a Lacôte 7 String

I finished my model of a 7 string Lacôte. Its construction follows the same principles as the six string Lacôte I discussed in four previous posts, the last on March 10, 2013. Navigate through "Older Posts" at the bottom of this page to view these.

I have measurements and photos of several Lacôte 7s and I examined another in detail several years ago. But I wanted to build a guitar that would appeal to guitarists outside the circle of period instrument enthusiasts so I decided to combine features from several guitars. I chose the body of one of Lacôte's décacorde guitars that is in Musée de la musique Paris, E.986.5.1.
E.986.5.1 Musée de la musique

The body is 443mm long, 248mm wide at the upper bout, 180mm at the waist and 310mm at the lower bout. The side rib depth measures 76.5mm at the front and 81mm at the rear. The front of the bridge is 126.5mm from the rear of the body. The rose diameter is 73.5mm and its lower edge is 335mm from the bottom edge of the body.
These were the measurements I made on my visit in 2012 and the ones I used building my model.

 Grant MacNeil, owner of The Twelfth Fret guitar store here in Toronto, has a passion  for romantic era guitars and there are always several in his store. I was fortunate that he offered me the opportunity to examine this rare 7 string string Lacôte before it was sold. The model has a short string length (580mm) and is now in a private collection so I didn't use it for my model. But the peghead is typical of Lacôte's work so it served as a fine example for my replica.

This closeup shows the similarity to Lacôte's  standard 6 string peghead design. He simply expanded it to include a side bracket to house the 7th string.

The original Lacôte tuners are mounted on plated metal strips that have corroded over time. Grant showed me photos of the detached peghead taken during the shop's restoration. The neck stock to peghead joint is a sort of concealed tongue and mortise. This appears only as an horizontal line across the narrowest part of the neck.

Rob Rodgers makes an elegant set of single Lacôte style tuners that were a must for this guitar.

The gears sit in recesses cut into the back of the head stock. The string rollers span the string slots and rest in blind recesses in the center spine. Installing each tuner takes time and accuracy. If I ran a production shop I would make templates and machine mill them. But I make one instrument at a time so I do it all by hand. I carefully laid out the design and used a scroll saw (Excalibur 16) to rough cut the string slots and recesses for each gear.

The sawed slots were clean but I erred on the side of caution and made them all a little too small. They needed to be enlarged and squared. I don't have a bench mounted vise so for jobs like this I improvise. The peghead is clamped to a squared block that is in turn clamped to the table. A second matching block is clamped some distance away. A long sanding stick with sandpaper attached on one face and both edges spans the gap. As long as I move the sanding stick perpendicular to the face of the peghead the finished work will be clean, accurate and square.
The string roller passages are blind, they do not go through the thickness of the peghead like the gear recesses and must be cut with a chisel. They require a lot of attention. The rollers on standard tuners fit snuggly in carefully drilled holes and yet smooth operation can be elusive. Tuner manufacturers even offer rollers with bearings mounted on the ends that fit tightly in the peghead holes but allow the rollers to rotate freely.
Since string tension pulls the rollers up toward the face of the peghead and back towards the nut the rollers have to rest precisely against the top and rear surfaces of their respective passages. Achieving this took some fiddling.  Initially, the passages for the rollers on each side of the string slots had to be cut undersize, slowly enlarged, testing the fit until they all worked smoothly.

Finally, I glued an ebony plate of the face and then opened up the string slots and cut the string ramps.

A matching maple plate conceals the roller ends in the center spine.

The two parts of the V-joint were  cut with the Excalibur scroll saw. Then I spent hours perfecting the fit.

I used Lacôte's iconic multi-piece design for both the rosette and edge purfling. But I added a later era style fingerboard and brought it to the edge of the soundhole increasing the number of frets to 19. Note in the previous photos of the historical guitars that both have short fingerboards with additional frets inlaid in the soundboard.

Twelfth Fret Guitar Shop

 Lacôte's choice of bridge design for his multi-string guitars is interesting. The bridge of the guitar from the Twelfth Fret is truncated. It is a simple rectangular block but otherwise contoured with a raised saddle area with the rear sloping to a thin edge like other Lacôte bridges.

The above is the bridge from Lacôte's décacorde conserved in St. Cecilia's Hall, University of Edinburgh. The tips are broken and the 'eyes' are missing but glue shadows pinpoint the location of the lost features.

I used the décacorde design and re-sized it for my 7 string bridge. Note that I angled the saddle to improve the intonation.

St. Cecilia's Hall University of Edinburgh

I have examined three of Lacôte's décacorde guitars and each time I have been impressed with the practicality and aesthetics of the contours of their necks. All were constructed in mahogany with a smooth transition onto the heel.

The advantage to this design is that the short smooth transition from the neck to heel allows the player easier access to the upper frets.

The entire guitar was finished with French polish, allowed to harden and rubbed out with various compounds.

All photos by the author.


Wednesday, April 6, 2016

Railich Theorbo Drawing

I examined the theorbo by Pietro Railich, No. 1569  Brussels MIM in 2007. Since then I have used it as a model for a small theorbo with a fretted string length of around 80 cm. In preparation for building a solid mould for these instruments I made a series of templates based on my measurements, tracings, photographs and notes. I recently turned this work into a simple line drawing of the theorbo's bowl. It is available  here as a full-size printable download. The file size is 67 KB and the printed drawing measures 43cm X 70.5cm.

Click here to download the Railich Theorbo Drawing

Detail photo of Gottlieb's drawing

A full size drawing of the soundboard by Stephen Gottlieb is available from The Online Shop (MIM). It shows many details; the bar positions and their sizes and shape, top thickness in many locations along with observation notes.

I had a copy of this drawing for many years and thought the instrument would make a good small theorbo. It also closely resembled the larger Kaiser theorbo that I had been building and that appealed to me. But I couldn't find other information or photos. All of this piqued my interest and on a trip to Europe I decided to include a visit to Brussels.

When I got there I found a theorbo in three pieces. The belly was separated from the bowl.

The other lute is a mandora by Johannes Jauck No. 0251

The front block, neck and extension assembly was detached from the bowl.

And the bowl was missing its front.


The condition of the theorbo affected the measurements that I could make while I was in the museum's laboratory and most recently the method I could use to make a drawing of the bowl. Obviously I couldn't precisely determine  the length of the bowl. I calculated this measurement by taking the length of the soundboard from its rear edge to the position of the neck joint which is clearly delineated by the fret board points that project onto the soundboard (see previous photo). The contour of the bowl at the front block also can be determined by referring to the soundboard profile. For the continuation of the contour of the central axis of the bowl beyond the fracture I created the line by considering the angle of the neck joint, the thickness of the neck at the joint and the position of the joint relative to the length of the soundboard. I had recorded these values during my examination. Fortunately, although the bowl was separated from the soundboard and detached front its front block, it had retained its original contour! When I placed the bowl on Gottlieb's drawing of the soundboard the two matched perfectly. This is a tribute to the skill of the builder and demonstrates the importance of accurately bending and assembling lute bowls. A further note - although the bowl matches the outline of the soundboard the two sides are not quite symmetrical. The treble side from the widest point of the bowl to the interior edge of the front block gradually widens by 2mm and then tapers back to symmetry with the base side as it reached the front block. My drawing uses the fuller treble side for both sides of the contour.

 Those of you who visited my now defunct website,
'' may remember the story on my 'projects' page about this theorbo. Several months ago one of you wrote for information about the instrument now that the page is no longer available. I decided to write this post as a result of that request.

While checking my notes and assembling photos from various files I found that '' is online (!), available in the online archive, Wayback Machine .


Open WAYBACK MACHINE and enter '' in the search window. A window like the one  shown below opens. I clicked on 2011 (that is the first year that the archive scanned my site after I posted the Railich project. January 29 was one of the dates . Clicking on that date opens the index page to my archived website. All of the buttons are live so you can navigate to 'Projects' or elsewhere. Unfortunately not all photos display.

All photos by the author unless otherwise noted