Sunday, November 4, 2012

Typical Marquetry Restoration

I really enjoy working on period marquetry surfaces, in particular those made with sawn veneers before 1800.  The quality of the materials, range of exotic materials used and, most importantly, the thickness of those materials make conservation and restoration of those surfaces more successful, when you know what you are doing.

Last week I started the restoration of a marquetry panel on the base of an English tall case clock, made around 1700.  This clock, like most of those clocks from that period, had lost its feet along with the lower 6" of the carcase, including the side panels.  When it was last restored, in the 1920's, in London, the worker removed the surviving marquetry from the lower part of the panel and replaced the pine board underneath.  Then he replaced the marquetry surface as best he could, saving most of the elements.  He also replaced both side panels of the base, and added a double molded base where the round feet were.

Inside the case, I found a lot of powder post beetle damage, so I had the clock fumigated by a professional in a chamber for 48 hours with methyl bromide gas.  Methyl bromide gas is the best way to guarantee full kill of the eggs, as most other fumigants kill only the adult and larvae, leaving the eggs to start the deadly process over once they hatch.

Here is the clock, with the sides removed.
Here is the face of the panel, showing the problems caused by substrate shrinkage.  The lower piece which was added in the 1920's has pulled away from the panel, and there is a lot of wood loss due to the bug damage.  Much of the marquetry surface is loose, as the glue has dried up, and the only thing holding these pieces together is the mastic and surface dirt and finish.
First I carefully removed the panel from the clock, so I could work on it.  Now you can see the condition of the marquetry and finish.
On the back you can see the loss of wood, as well as the newer lower board.  There were also two small pieces of wood applied by the repair man to old the cracks together.  They didn't work and were easily removed.
A close look at the restoration effort by the London worker in 1920 shows his technique for saving the marquetry surface.  He soaked off the elements with heat and moisture and saved them.  However, the background ebony was damaged and lost.  He replaced the background with a similar dark wood but was unable to properly cut in the cavities for the original inlay.  He used various chisels to carve out larger cavities for each of the elements and then filled in the surround with a dark mastic, which has now cracked.

I first worked on gluing the substrate boards back together.  This required cleaning old glue and dirt from the joints, using a toothing plane to prepare the edge, and gluing the three boards back together.  This also required a slight adjustment of some of the surface elements, as the substrate was no longer the original size, but had shrunk across the grain, as is normal.  Once the substrate was stable I turned it over and started working on the marquetry itself.

I was fortunate to be a participant in an international group of marquetry restorers in Paris during the early 1990's.  We called our group ADEN, which was a combination of the Musee des Arts Decoratifs de Paris and the Ecole Nationale Superieure Des Technologies et Industries du Bois, located in Nancy, France.  We did a lot of research on modifying protein glues, age testing, rehydrating old glues, and other specific problems related to conserving old marquetry.

One of the papers published in March 1997 specifically discussed this problem of the clock:  Amelioration des operations de recollage des placages et de redressement des panneaux-supports lors de la restoration de panneaux decoratifs.  Short translation: how to restore old marquetry panels.

Out of this research came the idea to make and modify protein glues, which I used to develop Old Brown Glue.  Although OBG is now sold nationally and used for a wide range of woodworking projects, I developed it originally to restore marquetry, which requires a long open time and deep penetration properties, as well as the ability to rehydrate and bond to old protein glue itself.

The first step is to place paper towels on the surface and add clean distilled water.  Watch and wait a few minutes or longer for the water to soak into the mastic and under the loose elements.  When you can see that the mastic has expanded and is raising up from the surface, you stop and remove the paper towels and surface moisture.  Then you flood the surface with Old Brown Glue and rub it into the marquetry.  Be careful to hold down all the loose elements while at the same time pushing the glue around, allowing it to suck under all the loose edges.

This method is usually done with the finish intact.  It is not necessary or desirable to remove the surface shellac or finish for this to work.  Note that the glue will penetrate only where the finish is damaged or lost, and only where the marquetry elements are loose or missing.  That is one of the neat features of this process.  You can conserve the original finish, if it exists and still reglue the marquetry in place.

Note I am holding down the floating loose elements with my left hand fingers while I am rubbing the glue over the surface with the right hand.

At the same time, I have heated an aluminum plate on top of a kerosene heater.  I place two layers of clean newspaper on the plate and the lay the marquetry panel glue side (face side) down on the paper.
The heated plate cannot be too hot.  If it burns your skin it will burn the glue.  150 degrees maximum.

Place the heated plate with the panel into the press as soon as possible.  I use gloves so I can hold the plate.

Place some boards on top and press.  Leave overnight.  Clean off the paper with cold water.

The next step is to create all the missing elements of the marquetry.  For this, it helps to have a French chevalet to accurately cut the elements, using sawn veneer that matches the original species.  There is a neat trick I use which works perfectly to capture the shape of a missing inlay piece.  I use thermal FAX paper, which is nearly extinct, since all modern fax machines are paperless.  I have a large supply of thermal fax paper on hand.  Place the fax paper over the missing area and burnish it with a hard smooth tool.  With experimentation you will find just the right tool.  Be sure to tape the fax paper tightly in place so it doesn't move.  After you capture the outline of the element, you use a light table to transfer that shape to a piece of tracing paper.  You cannot use the fax paper to cut the piece with the chevalet, as the heat of the blade will create a large dark shadow in the thermal paper.  Glue the tracing paper onto the proper veneer, which is held in a packet with a supporting board and some grease paper.

Place the elements in a tray, using the original fax paper as a guide so that you know where they go.

Place the new elements into the marquetry picture using Old Brown Glue to hold them.  Put again in the press with two layers of news paper and a heated plate.  Clean up the paper with cold water.

Note that there are still pieces of paper on the new elements.  This is also removed with cold water and gentle scraping.  Once the surface is clean you can begin the process of coloring the new elements to match the old.  Different chemicals, dyes and stains are normally used to balance the color and shellac is used to seal in the results.  Here is the panel during the coloring process.  It is not finished.  I will glue it onto the clock and finish all the other small repairs before I complete the polishing,  The original finish of the clock will be conserved with a French rubbing compound, called a "popote."  When the clock is done, next week, I will post a final picture.

I hope this illustrates clearly how you can approach such a project with a systematically proven process.  I am not the only person who uses this technique.  I can assure you that many workers in Europe, both in museums and private practice, are using these methods on a regular basis.  The goal is, of course, to conserve the great work of the masters for the future appreciation of collectors not yet born.

UPDATE:  Here is a photo of the clock put back together, cleaned and polished with a fresh coat of wax.


mokusakusensei--woods teacher said...

I am wondering how hot your glue is. Anything over 104 or so is pretty warm to the hand. I had this same question on the previous video where you spread the OBG over the surface. I realize, that the glue does not have to be as warm as hot hide glue, but it does have to be warm. Do you have an idea the temp? Thanks

Anonymous said...


Do antique restorers/repairers generally sign the work when they do a repair? How do you know that this clock had a repair in London in 1920?

W. Patrick Edwards said...

Old Brown Glue is best used at the same temperature as hot hide glue, 140 degrees. However, its viscosity and working parameters are a function of temperature, and the range of normal use is above 100 degrees and below 160 degrees. Higher temperatures give a longer open time and lower viscosity.

OBG can be heated up and cooled down as many cycles as you want without any loss of quality.

For this project the glue was heated to the temperature of the hot water from the sink tap, around 120 degrees.

Sometimes restorers do sign their work. More often clock repair men sign and date their work. In this case the repair was dated inside the case, and the clock was purchased at that time in London.

Chuck said...

Close inspection of the last two photographs lead me to ask some questions. There appears to be significant improvement to the background in the last photo. Did you replace any of the background veneer or just smooth and arrange the mastic? I will be very curious about your color palette and tools for applying it. Are you using water colors? transparent? Finally, a "popote" implies something of a secret recipe or mixture of not well known ingredients. Any clues?

A most interesting and worthy project. I am amazed at your ability to readjust the dimensional problems caused by aging and "weathering" of the piece.


W. Patrick Edwards said...

The last two photos show the panel before and after I removed the paper from the new elements. Some of the new elements are light (boxwood) and some are dark (ebony). That is why the background looks better in the second photo.

I use ground pigment powders which are carried in shellac. This allows me to color the new elements to match.

The popote is a rubbing compound. Each finisher makes his own, and there are several commercial products available. In America you can use some of the automobile rubbing compounds with similar results. Be sure to use the finest rubbing compound available.

JC said...

Seeing and reading about this type of restoration is amazing. I'm so happy to have found this blog. I'm curious to know if you will be re-repairing old veneer repairs such as the piece in the upper right corner, or if you will simply blend it in using additional dyes/paints to conceal it?

I can't wait to see the rest of the clock.

Will the base be restored (with bun feet) or left in close to its "as found" state?

Also, isn't "popote" kind of the same as "concoction"? Hahaha! It helps that I speak French.

W. Patrick Edwards said...

Restoration and conservation as a business is always a compromise. If it were in a museum, it would be pulled from the gallery and put in the lab where it would sit for a year or so while it was analyzed and conserved. At that point all decisions about prior restorations would be discussed and considered.

As a business, I need to consider time and cost much more than a museum. Also, this clock has been through some serious restoration some 90 years ago which significantly changed its condition. The primary wood on the case is walnut, but the restorer who repaired the base used another wood, completely replacing both sides of the base, as well as the crossbanding on the front of the base.

Also, the bun feet were lost and the new molding was added, which, for some reason, he chose to use walnut to make.

The client was concerned with the structural integrity of the base, as well as the loss of elements. My job was to repair those areas as well as clean the finish.

Yes, a popote is a concoction of stuff, made by each restorer to revive the polish. It serves to remove dirt and wax, as well as slightly abrade the surface of the existing polish to restore the shine as much as possible. We import and use a popote supplied by the firm HMB, located in the 12eme arrondissement.
Use a little popote and a lot of rubbing.