Reproduction Walnut Hall Tree Part 3: Assembling The Hall Tree

This is the third post concerning the reproduction of a 19th century hall tree. The previous two post have dealt with designing and making individual components. This third post will follow along with more of the same as well as show the assembling of all of the components to make the hall tree. One thing that always surprises me when constructing a piece of furniture is the interrelation of all of the individual components. If a change is made or an error occurs, the result affects the entire piece. Sometimes this is a change for the better or an error that can be corrected, and other times it can have deeper consequences, like having to make a new part from scratch. I liken the process to walking on a tight rope. It requires concentration and a steady, focused hand to persevere. That being said, mistakes inevitably happen and it is good to know how to deal with them when they occur. Now enough with the philosophy, and on to the project!

There were several components that I decided not to draw in SketchUp (see first post) but rather to copy directly from the piece. These were appliques that were superficial to the construction of the piece, but enhanced the design of the hall tree. The first of these was a piece of molding that had an edge consisting of half circles cut into a cove. The Molding is pictured below:
The first step was to make the portions of the molding made by the router. This consisted of the decorative lines running from side to side and the cove itself. This was a very small piece of molding in terms of width, so I left a lot of extra waste material so I could have better control of the piece as I worked on it and also I could keep my hands away from the machinery. Once the routed parts were completed, I flipped the board over and drew the half circles on the back side of the cove using a compass. This next photo shows the half circle layout on the back of the board.
I then took the piece to the band saw and cut the circles out. The photo below shows what that looks like. After that the last step was to remove the extra waste material by ripping it on the table saw.
On either side of this molding is an applique that works as a transition piece from the molding to the frame of the hall tree. It's purpose is solely visual. Below is a photo of one of these pieces followed by another photo showing the pieces made for the reproduction. The profile of these pieces were cut out on the band saw.

Next to those pieces, and applied to the sides of the hall tree, were some decorative pieces with carved hints of a volute. All of the carving on this piece was simple and were typical of factory carving of the time. The intention was to add decoration and to hint at the idea of a highly carved piece. Most of the carving was shallow and was similar to drawing. The original applique is pictured below followed by the ones I made for the reproduction.

These last applied components are found in the middle of the hall tree and flank the central raised panel on either side. There were cut out on the band saw and then little turned domes were applied to the faces of the components. here is a photo of one of the originals.

Here are the two pieces for the reproduction after being cut on the band saw.
This photo shows the applied domes being glued in place.
This photo shows the finished product.
Now that all of the applied pieces were made it was time to assemble the hall tree. The first step in this was to take all of the rails and stiles and lay them out on top of each other with the tenons overlapping the piece it would mortise into. This allowed me to accurately layout where the mortises were going to be. Below is a detail photo showing this. Rather than being flat, the pieces were stacked so I could trace around the tenons on to the face of the piece below it. The photo that follows shows the entire back assembled in this fashion.

After the mortises were all cut, I dry-assembled the back to make sure everything fit together nicely. The next photo shows this. It is followed by a photo of the rails and stiles for the bench assembled. This was an exciting moment because for the first time I was looking at an assembled item versus a pile of parts!

After all of the mortice and tenon joints were test fitted and fine tuned, I fit the raised panels in place and fine tuned their joinery as well. This photo shows the frame dry-assembled with the panels in place.
Before gluing the back together, I needed to apply some wood to the long back stiles that would stiffen these stiles up and help keep the back straight. I wanted these piece to be straight and in a search for a flat surface to glue them on, I remembered a lathe in the shop with a seven foot bed. The bed provided the flat surface I needed so I glued these strips together clamped to the bed. Below are two photos of this process.

After that was completed, I removed the clamps and glued the hall tree together. There were a few appliques as well as the seat I left out, but most of the gluing happened at this time. Below is a photo of the hall tree clamped while the glue was setting followed by a photo of the hall tree standing on it's own four feet!

The next step was to make the lid for the box at the bottom of the case which would also act as the seat. This lid comprised of a frame that went around the back and sides of the lid and the lid itself, which had breadboard ends to strengthen the lid. Below is a photo of the breadboard ends being glued to the lid.
Once the glue dried on the lid the clamps were removed and the ends of the breadboard ends trimmed. The next step was to attach the hinges to the lid and the board that went across the back. After the hinges were in place, the boards which frame the lid on the sides were glued on to the back. I glued them using the hall tree so that they would match the rest of the piece as well as possible.
Now that the entire seat was assembled, I needed to produce the rounded molding along the edge of the seat. The next two photos show the seat set up for this procedure. I clamped boards on either end of the seat to run into with the router. This prevented grain tear out and the boards were discarded after they had served there purpose.The last photo in the set shows the seat with the molded edge.



The last step in the assembling process was to attach the seat and arms to the hall tree. The arm posts had turned tenons which joined them to the seat and the arms. The seat and the arms were attached to the back with screws and I used dowels to attach the seat to the seat frame below. The photo below shows the seat and arms being glued in place. It is followed by two photos of the fully assembled hall tree. The next and final post will deal with the application of the finish and show the final product!

One note: I have left the lion carvings that are featured in the first post off of the hall tree in these photos because with the lions in place, the Hall tree is too tall to stand in my shop!

Mid-Atlantic Ladderback Arm Chair (ca.1780) Part 1

 Recently I have started to repair and restore a Ladderback or Post and Rung Arm chair which I have dated to being manufactured around 1780.In fact , this chair could have been made anywhere between 1750 and 1800. Dating a chair correctly is tricky business because this style of chair was made from the early 18th century through the beginning of the 20th century and continues to be made today. At first glance I had a feeling that it was probably made in the 18th century, mainly by looking at the finial design, but it is extremely difficult to be sure of such a thing. My suspicions were latter strengthened through close examination of the joinery, which I will discuss later. The really amazing thing for me about the restoration of this chair is how much I was able to discover about it with close examination.

A similar chair appears in the book Southern Furniture 1680-1830:The Colonial Williamsburg Collection  by Ronald L. Hurst and Jonathan Prown. There are several stylistic differences between the two chairs, but over all the construction methods used are very similar. One striking similarity is how low the stretchers are on the bottom of the chair. This was a feature I immediately saw in the chair I am working on. two notable differences are the arm design and also the fact that the Williamsburg chair has a square seat while the chair I am working on has a trapezoidal seat. These, I believe are all stylistic differences, perhaps particular to a region and or a particular maker. below is a like to this book:
















The chair is made up of two woods, Maple and Hickory. Maple was used for the posts of the chair, while every other component was made from Hickory. The chair came with a worn rush seat which was not original. Here is a photo of the chair as it came to me.

 

Above is a photo of the side profile of the chair. If you notice, the chair back is set at a slightly skewed angle from the rest of the chair. This allowed the back to be a bit reclined for more comfortable seating. Other ways of achieving this were to steam bend the rear posts so that they were bent a little in the middle or to use a method of multi-axis turning to achieve a bent rear post. Early American chairmakers seemed to prefer a straight leg set at an angle (which involves some interesting joinery angles) to the other two methods. In the 19th century, chairmakers commonly used the steam bending technique to achieve the raked back. This alone is not a good way to date a chair, because there examples of all of these methods throughout the 18th and 19th century.

The photo above is taken from the top of one of the rear posts. It is hard to see, but the leg is slightly oval in shape. Originally the posts would have been turned round on a lathe. as with all woods, the posts have shrunk along the grain so that the circle has turned into an oval. As I will later show, this natural function of the wood is used to the chairmaker's advantage.

When a chairmaker was plotting the layout for where the stretchers would meet the posts, he needed to have a way of marking where to drill the mortises for the tenons on the stretchers. Pencils were not as readily available as they are today, so alternative  methods were used. Traditionally, when a post was on the lathe, the final step in the turning process would be to incise lines at the points where the post and stretcher would meet. this would either be done with the pointed end of a skew chisel on end, or by use of a stick with several sharpened metal points fixed to it being placed next to the spinning blank. the points would incise the lines exactly where the chairmaker wanted them. The benefit of this later technique was that the measurements were predetermined on the stick with the points and the scribe lines could be repeated exactly from post to post.  Neither of these methods were used by this chairmaker. The scribe marks on this chair were made on the lathe while  the blank was spinning using wire to burn the wood by friction. This permanently darkens the wood and serves the exact same function as the incised lines. I have heard of this technique in turning before in turning, but this is the first time I have ever come across it being used for this particular application. Above is a close up shot of two of the scribe marks on the chair. They mark the top and bottom points of the mortise for one of the slats in the back of the chair.

This photo is a closeup of the finial. With more research, it might be possible to find out the region and possibly the maker of this chair by close examination of the finial. The lines on the finial were once again made by use of the wire burning technique.

As stated above the rush seat on the chair was not original and was in need of replacement. When I started to remove the old seat I noticed that old newspaper had been used for stuffing between the layers of rush. The paper was balled up and very brittle from age, but I was able to open enough to find out that the seat had been put on in 1934 in Xenia, Ohio, a town near Dayton. The paper was various copies of the Xenia gazette. One copy had a date of Friday October 20th, 1934. above are two photos of the paper I found.  

This photo shows the chair after the seat was removed.

What followed next was very interesting and told me great deal about the manufacture of this chair and the amount of attention to detail that the maker put into his work. The techniques used below also confimed in my mind that this chair was made during the 18th century, since that is the time period in which they were most used.

The chair was loose in spots and so I set about knocking the chair apart so I could clean the joinery and re-glue the chair. When I started to dismantle the chair, I found that the tenons for the chair had a specialized design. after a little research this is what I have learned.

Chair makers of the 18th century used two techniques to make a locking tenon that would keep the chair tight, even when the glue dried. One utilized a split tenon that was wedged from the interior. The other process is as follows. The chair maker would make the tenon of the stretcher slightly bulbous, sometimes carving a shoulder onto the inside of the bulb. The stretchers were made of Hickory in this case. The sides of the tenon would sometimes be carved down flat so that the tenon was slightly wider in one direction that the other. The tenons were then heated by putting them next to a wood stove on just outside of a fire in an effort to remove the moisture from the tenon causing it to shrink. Meanwhile, the posts for the chair would be made from green wood and the mortice for the stretcher would be made using a Spoon bit. Below is a photo of a spoon bit:

Spoon Bit

The spoon bit used with a brace would drill out a hole for the mortise. With a little movement, the chairmaker would cause the bit to cut a hole slightly wider in the center. After the mortise was drilled and the tenon was dried, the tenon would be hammered into the mortice with a little hot hide glue. The moisture from the hide glue would cause the tenon to expand into the mortise causing it to lock in place. As the posts cured, the wood would contract around the tenon making the fit tighter. The flattened sides of the tenon prevented the post from cracking at it shrank do to pressure. The end result was a tight fitting joint for the ages that relied largely on the forces of mother nature.

Any person who has ever glued a chair will tell you that the sections that loosens up the most are the stretchers that connect the front of the chair to the back. This is due to the motion of people sitting in the chairs  and leaning back, etc. To accommodate for this, the chairmaker oriented the posts so that the maximum amount of shrinkage would happen around these stretchers, minimizing the amount that the chair loosened. The result was that although the stretchers from side to side were loose, the stretchers from front to back were very tight and needed no re-gluing, just a little wedging in spots.

It is very impressive that this amount of consideration went into the manufacture of this chair. It clearly shows that it was the intent of the chairmaker of old to build a chair that would last for generations to come.

Below are some photos of the tenons.

This photo shows the tenon with its carved shoulder. The end of the tenon was rounded like the end of the spoon bit.

The flat side of the tenon can be seen in this photo.

It is worth mentioning that while I had the chair apart, I noticed that the tenons for the slats had a barb on the top edge of the tenon, similar to the tenons on the stretchers. These serve no function that I can see, because they are visible when the chair is together and do not lock the slats in place. One other thing I discovered while the seat was off and the chair disassembled was that there appeared to be remnants of blue paint. Many of these chairs were originally painted and refinished some time later on. It is my opinion that the chair was not painted originally due to the fact that the finial has a decorative scribe mark around its diameter. I do think that it was later painted and then striped and refinished at some point. It is possible that the paint is original and that would certainly be in keeping with the trends of the time.

The tenon of a slat with its barbed side and remnants of the blue paint. This photo shows the condition of the existing finish in detail. The crazing in the finish is largely due to use of linseed oil on the finish.

these last two photos show the chair disassembled and also being re-glued. The next post will focus on the restoration of  the finish and  the application of a new seat.

Walnut Eastlake Cylinder Desk (ca. 1890) Part 1

One of the pieces I have been working on recently is a cylinder roll top desk built in the Eastlake style around 1890. The piece is American, and uses Walnut lumber and veneers throughout the case. the secondary woods used are Cherry in the cubby holes on the interior of the desk, and a mix of Poplar and American Chestnut throughout the case. This piece is very similar in style to a desk that I worked on about a year ago. Her is the link to that piece:

http://johnmarkpower.blogspot.com/2010/05/walnut-eastlake-cylinder-secretary-desk.html

The first and most noticeable difference is that the desk I worked on previously has a glass front cabinet above the cylinder, where the current desk has a shelf. One of several other differences is that the current disk is smaller in proportion. If you notice, I dated the current desk slightly later than the first desk. The reason for this is that the desk I am now working on employs the use of machine made dovetails on the drawer as well as a joint called a sliding keyhole joint or a "french" dovetail. Both of these joints made there debut on the furniture market around 1890. Immediately Prior to this most pieces used what is called a Knapp joint. A history of the Knapp joint can be found in the following link:

http://johnmarkpower.blogspot.com/2010/05/knapp-joint.html

The desk I worked on previously had this type of joinery, which was used as early as 1870. With this information, I was able to identify this current desk as being made around 1890 because of its use of machine made dovetails. In fact, this was one of the very first machined dovetails made. What is impressive to me is how well these dovetails look and also how strong the joinery is.

One other note, if you are interested in looking at other examples of Eastlake style and a little history on the Eastlake Movement in Furniture, feel free to click on the Eastlake label on the right hand side of the screen.

Overall this desk was in decent shape, but it had several repairs that were necessary and also a very dirty finish. The finish beneath the dirt was in pretty good shape so I decided to restore the finish instead of removing it. Below is a photo of the desk as it came to me. The door on the right hand side is off of it's hinges.
The next two photos show the joinery used in the two drawers. The large drawer in the bottom section of the case uses the machine made dovetail and can be seen below.
The smaller drawer in the cubby section of the desk uses the sliding keyhole or French dovetail. This is a simple joint made using a table saw and is a good one for a small drawer such as this.
The case was loose in several places. The next two photos below show the top desk section being glued. This piece sits on top of the lower section and is held in place by dowels.

On the gallery section of the desk section, the board that creates the back section had cracked open. the next photo shows the crack being glued up.
One of the doors had a crack in one of it's stiles which was loose.The next two photos show the door being glued.

The cubby section of the desk was also loose. This photo shows it being glued.
The writing surface for the desk pulls out along a rabbited track. the rabbited groove in the lid had broken on one side sometime ago. This prevented the writing surface from pulling in and out easily. The photo below shows the broken portion on the underside of the surface being repaired with new wood.
This next photo shows the lower case being glued.
The desk surface was missing some appliques o the top of the desk, so I needed to turn two new ones. Luckily, I had several that were intact to go by. Below are two photos showing this process on the lathe and the finished product.
Likewise, one of the doors was missing it's escutcheon. I used the existing escutcheon on the desk lid as a template and made a replica. Below is a photo of the new escutcheon being turned, and then the finished escutcheon after the key hole was filed. The last two photos show the new escutcheon in place on the door. My next post will show the finished piece.

The turned escutcheon.
The escutcheon after the key hole was filed.