The Georgian Bedstead, Part Four ~ Turning and Carving the Lower Bedpost Sections.

          You can’t afford to take any chances with expensive wood, especially if the supply is limited.  I took a short piece of pine, made it up in the same way as the posts just described and put it on the lathe to turn a trial of the urn that adorns the upper end of the leg section.  This way you see whether the theoretical design does in fact please the eye.  If so, then from it you take exact dimensions in order to copy the form four times.   If you are lucky enough to own a duplicating lathe, then the trial piece may be mounted as template which will give you four exact copies mechanically.

Next job is to make a careful drawing for the carving on the urn and transfer this design to the workpiece. I intended to carve four acanthus leaves around the urn.  The transference can be done with carbon paper, but I prefer to make small cardboard templates and trace around them.  The first templates I made were too skinny – I hadn’t allowed enough for the varying circumference of the urn. Before starting to carve I made a measured drawing of the trial urn, which serve as the standard for four posts. Some carving on the trial piece is a good idea to test the concept, but there’s no need to carve the entire urn.   I went about halfway around. This is helpful too as you learn and note which gouges will do best at the different parts of the carving. After carving the same motif for a while you get in the swing of it and will automatically reach for the right tool, but at first this is difficult to remember, so when chopping in the trial carving I labelled a copy of my acanthus drawing at every different curve with the identity of each tool used.

A final note. I am taking a large number of reference photos during this ongoing project.  I am posting up the most relevant with this blog, and a very few of the most interesting ones on my Flickr site.  For those interested in yet deeper detail I have all my pictures in a folder 'on the cloud' at my "Dropbox" site.  Upon request I will be happy to send you a link to my Dropbox folder, and here below is a link to the 'Georgian Bed' set on Flickr:  http://www.flickr.com/photos/a_henwood/sets/72157635480669119/ 

The Georgian Bedstead, Part Three ~ Making up the Lower Bedposts.

                         

 A need for large washers was mentioned in the last posting. I wasn’t satisfied with the washers available to me at the hardware store, so chose to make them myself.  I had a pair of ancient gate hinges made of strap mild steel of the right width. I was planning to use 5/16” threaded rod for the connections, and so I drilled  spaced holes 3/8” diameter along the centre of the hinge strap. I countersunk these holes to facilitate easier entry of the connecting rods. I then hacksawed the strap into square washers. I believe the square shape provides better thrust distribution for this application. Furthermore, it was a simple matter with a straight bit to rout out recesses in the post halves to house this shape.

   

    The only way I know of to make precisely true central cores down long sections is to make the post in two halves, routing or dadoing a channel down the middle of the inside surface of each half.  I did this, making a half inch diameter central channel.  I modified this channel by hand carving to a hexagonal shape at the point where the collar nut would lie, so that it could never turn in the wood. Behind (downstream) of the nut I laid in a short section of dowel to stabilize the collar against longitudinal movement.  The photo shows all this assembled, ready to be glued up. When gluing up I left a piece of threaded rod inserted into the collar nut so as to keep it perfectly aligned during the process. I oiled the rod in case of contact with the glue, and held the other end perfectly centred in the channel with a scrap of vinyl tubing.
     

 For specialised work of this kind I use a powdered resin glue, which is mixed with cold water into a thin syrup.  This glue is extremely strong and waterproof: ‘marine grade’.  It sands well and cleanly, so the line will be invisible at finish. Far better in my view than the poly-vinyl type of glue.  As you look at the photos, you may be wondering (worrying?) about the pattern match as I lay up these pieces.  It is obvious that there is a poor match in the side visible above.  Wherever possible I ‘book-matched’ two pieces to form one of the legs.  My original stock being 2x6 inches or 2x7, I was able to do this in most cases, which meant that I ended up with three good sides to the leg.  Only two sides of the lower leg section are visible; the poorest side ends up on the inside. After planing, jointing, ripping the plank in two and final dressing, the resulting glued up posts ended up approximately 3¼” square. I took that down afterwards to a smidge over three inches square.

Concerning Measurement.

      You may have noticed that I sometimes give dimensions in the old ‘Imperial’ units of inches, and sometimes in Metric.  In England, my generation grew up with the Imperial system, not only just yards, feet and inches, but also a rich trove of other units passed down from time immemorial: in school we gained familiarity with rods, chains and furlongs, and learned how these related to areas such as the acre, and longer measurements like the mile.

             I was already adult before the Metric System was introduced nationwide. The changeover in the United Kingdom was never entirely achieved.  The same attempt was made with more, but still limited, success, in Canada. This transition was made more difficult  by the presence of our neighbour to the south. The U.S.A. does not yet embrace the metric system, so proximity obliges us to be able to function in both systems.

             Anyway, when forty years ago I began to make furniture for a hobby, I began by using Imperial units.   I quickly found though, that when crafting small items, working in millimeters was far simpler, more precise and less likely to result in error. On the other hand, when measuring large components I found feet and inches easier to visualize.  So I ended up using both systems, often in the same piece of work. So, as an example, I may refer to a floorboard being four feet long, nine inches wide and 22 millimeters thick.  Happily, I am supported in this choice of method by the availability of measuring tapes and rulers which have Metric down one side and Imperial markings on the other. Obviously, other people have arrived at the same solution. I trust the reader can make the jump.

             The reason I mention this now is because you will see in the next part of 'The Georgian Bed' that I habitually use whole millimetres when I am measuring the diameters and spacings of the various parts of a turning.  Easier to read when laying calipers on the rule, and no chance of mistaking 11/16ths of an inch for 13/16s!

The Georgian Bedstead ~ Part Two, Drawings and Design Factors

    The bed being fabricated in Canada, but destined for England, I had to plan carefully so that the various components would be small enough to be taken over as checked baggage on the airplane.  This is a tough constraint, but not insurmountable.  The bedposts would end up with a final height of about 86 inches (see drawings).  They would be made in sections which would be screwed together, and could be taken apart in future when required for moving.  In order to get the bed into use as soon as possible, I planned to make the lower post sections first,  take them over and source material for the bed rails and structural head and foot boards in the UK, since they would be too big to take on the flight. 
           Here are the first drawings. At this stage I  planned simple head and footboards, the same width and thickness as the side rails, so as to be able to bring the bed into use quickly. The design for a panelled decorative headboard is presently in early stages. We may leave the footboard as shown, the original having just such a simple arrangement.
          
The first challenge was to design a very strong, simple, foolproof connection between the bedpost sections.  I have built poster beds in the past, connecting the sections variously with hanger bolts or sections of threaded wood. Neither of these methods proved to be a perfect solution.  From a lifetime of experience Jim gave me his advice, which was to connect them with a long length of threaded steel rod, through a continuous hollow central core running the length of the post.  Right away, I had to modify this plan, because the posts would be connected with the rails and boards of the frame by means of bedbolts passing crosswise through the centre of the lower post sections. A longitudinal rod would block that.  So the rod fastening the lower post section could not be allowed to run down to the floor.  I solved that difficulty with a plan to embed collar nuts in the heart of the legs into which the rods would be screwed. The thrust would be borne by large washers abutting the collar nuts; (see drawing).

The Georgian Bedstead ~ Part One, Finding the Material

The original bed was made from mahogany and we wanted to use the same material, so the next thing I needed to do was find a supply of this. I am most fortunate to have a friend here in Niagara-on-the-Lake who is a cabinetmaker, historian and restoration specialist; a lover of architecture and old buildings. He has a large workshop on the edge of town and a great stock of all the materials relating to these interests. He has old Victorian window glass, antique hardware and fittings, you name it. Most importantly for me, as it turned out, he has a wide range of timber, accumulated over many years; some of it has been reclaimed from old structures and is decades old. When I explained my project he told me that he had some mahogany. I could take a look and see if it would be suitable. I went around and found the wood to be perfect for my purpose, and there was enough of it. Good and generous man that he is, he allowed me to buy it. Mostly, the wood was in planks two inches thick, by six or seven wide and eight or nine feet long. Jim has heavier, more powerful machines than I, so he helped me with the first cutting, in preparation for making the bedposts, which was to be the first task.

Building a Reproduction Georgian Bedstead ~ Introduction

In April of this year I decided, at my daughter’s request,to build for her a reproduction four-poster bed of Georgian design. We had seen the bed (photo above) at a country house in England: Hinton Ampner, owned by the National Trust and open to the public. I wish we had made a close study of this beautiful piece, and taken measurements too, but the decision to copy the bed was not made on the spot. However, we took a number of photographs in and around the house that day, and fortunately several of the bed. From these photos I was able to deduce the size and proportions fairly closely, and when I came home to Canada I drew detailed drawings as a prelude to the work.

Building a Shutter

The last time our old house underwent a serious renovation was in 1971. That was when it got a nice new set of exterior window shutters. Forty years is about as long as these items will stand up in our climate, and all of ours have been showing their age for some time. From time to time I have tried to repair and stabilize them, but this spring it got to the point where one shutter was so rotten it was beginning to fall apart. Something had to be done.

Not too many people make wooden shutters anymore. I called around and finally found a company (some distance away) that still builds them. I phoned them up. 'Give me a ballpark price' , I asked. 'Send us your specifications and we'll quote', they said. Well, I thought, by the time I write all that out I could be well on the way to building it myself. So I thought I'd give it a try.

I had the idea that it would be a complicated job, so I pulled apart the old shutter to see how it was put together. It proved quite simple actually. Through mortises and tenons hold the rails and stiles together - shouldered at top and bottom, full width in the middle.

Before putting the frame together you must cut diagonal slots in the inside faces of the stiles which will hold the louvres in place. I used a dado set to cut these slots a quarter of an inch wide, at 45º. Not having a mortising machine, I chopped the mortises by hand, which proved a tedious task, especially when dressing them to try to get the tenons to fit. Since the stiles are only 2¼ inches wide, next time I will set up a jig and rout them, going halfway in from each side. Now I glued up the frame, wedging the tenons for greater strength and integrity. I used a one-part, powdered resin marine grade glue, which is mixed with water, is very strong and becomes fully waterproof in 48 hours. When the glue had cured, I ran a rabbett down the inside of what was to be the central edge of the shutter, so as to provide a neat overlap with its opposite number when closed. On the outside of this same vertical edge I ran a simple bead as per the original, to dress it off. You can see the rabbett in the photo which shows the trial installation of louvres below.

Next I had to make some trim, which goes around the two apertures, front and back, mitred so as to line up exactly with the inside edges of each louvre aperture. I wasn't able to exactly match the original, so I made a ½" wide by ¼" high cloverleaf profile using my old Sears molding head on my equally ancient radial-arm saw.

At this stage the trim is applied to only one side of the shutter frame, using finish nails and the marine glue. Now the shutter is turned over so as to fit the louvres, which will be prevented from slipping right through the slots by the side pieces of trim, which retains them underneath.

Having first cut the slots, it was now easy to plane the stock for the louvres to the exact thickness for a perfect fit. Using an extra-thin ripping blade, I resawed one-inch pine stock (19mm thick) exactly in half. This gave me two thin planks about 8.5mm thick. My slots actually turned out at 6.7mm wide, so it was a quick and easy job to finish plane the louvre material to this thickness.

In the nature of things, there were subtle differences in the actual precise length of the 39 louvre slots, so I now cut the louvres to length and laid them up in a trial fit, numbering them lightly in pencil. In cases where a louvre was slightly too tight or too loose for the slot I had cut it for, I would find another slot where it would fit just perfectly!

Until this point I had left the louvres slightly oversize in width. They are designed to sit slightly proud of the frame on each side, just about to the level of the upper surface of the trim. In practice, the frame stock being 1¼ inches thick, this meant that the louvres were to be two and one eighth inches (54mm) wide. I quickly accomplished this by clamping the louvres together in batches of thirteen, and running them through my 13" Delta planer on their edges as a block.

Remember now, that the louvres are held in place by the trim, yet stand proud at their edges. So now I cut tiny 45º shoulders on both sides, both ends of all the louvres. You can see in the photo how I batched this process also by laying a 45º support block against the fence of my saw, then stacking a whole bunch of louvres against this support. To ensure the shoulders were exactly the required length despite the variation in louvre length, I also fenced out at 90º on the table on the side I was cutting. This is hard to see, but also illustrated in the photo.

Now that the louvres were precisely fitted and shouldered, it was time to round them over, which I did with my newly acquired Lee Valley one eighth inch radius rounding over plane. What a little gem of a tool that is! One pass down each long corner of the louvres and the job was done.

I glued in the louvres, not so much for strength but for protection. I had noticed that these thin pieces of wood had decayed more at their ends, where damp could penetrate more easily; so putting on a generous coat of waterproof resin glue should extend their life considerably.

Almost finished now: all that remained was to apply the trim on the second side so as to hide the shoulders of the louvres, and lastly, to pin all the joints with wooden dowel.

I was reluctant to use commercial dowel for this purpose, feeling that an unknown species of wood might not survive so well as if I used the same cedar out of which I had built the frames.

It shouldn't be that hard to make enough dowel to pin six joints; and it wasn't. I split a cedar offcut to first get a stick about half an inch square. Then I whittled it with my pocket knife until it was more or less round and roughly 12mm in diameter. I puzzled over how to achieve a precise three-eighth round dowel from this. There is a tool, but I don't have it. It is a metal plate with serrated tapered holes in it. You drive a stick such as I had made through the appropriately sized hole. I looked around the workshop, and my eye fell upon the old handsaw that was one of the first tools I acquired, back in the late sixties. These old handsaws have holes in the tip end of the blade, so that you can hang them up on a nail. Sure enough, here was a three-eighth hole in a piece of hardened steel! I laid the saw flat on my bench with the hole over a benchdog hole and drove my stick through it with a wooden mallet. A mite rough to be sure, since the hole hadn't been made for purpose and sharpened, but a true round dowel. The job was done.