Well I started drafting this post in late Spring. It is mid-summer now. So much for spontaneity in posting! Still the cleanup was actually done in spring.
I need to revive my lathe. After an initial honeymoon with my wood lathe some 20 years ago it has sat neglected in the corner of my workshop. As I mostly make contemporary furniture, the lathe is only useful for creating cylindrical elements and for years I have been making these by hand-planing and sanding. Now I need some shaped cylindrical elements so the lathe must arise.
For the last 18 months the lathe has been used to support a stack of blackwood and figured eucalypt. In fact that whole corner of the workshop had become a bit of a dumping ground. Fortunately I didn’t photograph it before the cleanup!
The spring clean turned up a few interesting items.
Safety guard for Wadkin dimension saw
This is the safety guard for my Wadkin saw. I don’t use it because I find it safer to have a clear view of the blade
. What’s impressive to me is the size and solidity of the components of these old machines. This guard weighs almost 5kg even with some aluminium parts! I should have put a scale in the photo but the guard is about 150mm (6″) high. The arm that supports it is 1.5m long and so has to be pretty hefty itself - it’s a steel rod about 40mm in diameter and damned heavy. I don’t know if the guard has its the original paint job but anyone who loves old Wadkin machines will find the logo as ’stimulating’ as I do. The guard is sitting on the sliding table of the saw which itself weighs over 1000kg. You can see the dark but shiny patina that has developed on the milled steel surface over its 50+ years.
Rip fence for Wadkin dimension saw
As an aside - this is the rip fence for the saw. As you can see, the fence slides longitudinally on a dovetail and tilts to 45°. The position mechanism is a delight to use. You slide the fence to roughly the right location and lock it using the cam lever at the back. Micro-adjustments can then be made by turning the knob at the extreme right. The final location is then locked with the cam lever at the front. The inset scale is highly accurate. You can confidently rely on it for at least .25mm accuracy - probably better.
My first dovetails
This box has been sitting on a shelf behind the lathe for some time. It held a few parts for my trimmer router and some other tools. The box deserves, and is now getting, a little more respect. It represents my first serious attempt at dovetails.
Back in 1990 I was considering getting some formal training in woodwork. I was looking around the world for somewhere to study (including Krenov’s College of the Redwoods) and found that one of the best teachers in the world - George Ingham - was teaching right here in Canberra. I dipped my toe in the water by enrolling for a week-long summer school to make a dovetailed box. The course was run by George and his partner Pru Shaw. This is the box that I produced.
It’s quite a simple, even austere, box. The wood is plantation mahogany from Fiji. The top and bottom are MDF. Many of the other students were perturbed that we were using a man-made board for the main surface of a box that took so much effort and care to make. But the choice made perfect sense to George. Much of new furniture design especially in the 20thC has been driven by the arrival of new materials or technologies - think about steam-bending technology, plywood, lamination, plastics. In his own work George strove to get people to consider MDF for its own merits rather than just as a stable substrate for flashy veneers. Nevertheless some students insisted on a traditional look and veneered the lid.
My eyes were opened to a whole new level of excellence during this workshop. In addition to acquiring some skills I got a glimpse of what one could strive for and an insight into the importance of systematic approaches to the processes of woodwork.
I now have this box close to my handwork bench and use it to store my more delicate components such as fine brass hardware and supplies of mother-of-pearl. It holds a place in my workshop more in keeping with its role in my development as a woodworker.
The cleaned and re-organised corner
And now the result. Shelves were erected to store some of my more valuable sticks of wood. The lathe is usable again and I have some table space for saw accessories.
January 3rd, 2009
An exhibition of the work of George Ingham is being planned for October 2009 - A Singular Talent: a retrospective exhibition of the work and teaching of George Ingham (1940 - 2003). In conjunction with this, another exhibition - Chairs of the Alumni: the ongoing narrative - is being organised. Former students of George (that includes me) are invited to exhibit a recently made chair.
To quote from the invitation: “The exhibition Chairs of the Alumni will draw on the intakes of the years 1983 to 2003 inclusive to serve as a vehicle through which George’s teaching might be celebrated. It is hoped that the contemporary pieces will underscore individual maturation, and in the exhibition they will be counterpointed by George’s own teaching pieces and drawings.”
So - there is my new project in a nutshell! It’s quite a while since I designed a new chair. In fact the last was these conversation chairs exhibited in 1997. Damn that’s a long time ago. So this challenge is overdue.
Conversation Chairs by Mark Woolston
I hope to take my readers along on this journey over the next 12 months - showing the design ideas, the ergonomic tests, the design resolution, the making, the catalogue photography - right through to the exhibition.
Hope you come along for the ride.
September 29th, 2008
Ming chair by George Ingham 1997
This wonderful piece of design and superb execution is a chair by the late
George Ingham. George was the foundation head of the wood workshop at the Canberra School of Art - and, I’m proud to say, was my teacher and mentor.
This chair featured in an exhibition, The Chinese Chair, curated by Rodney Hayward and Nigel Lendon at Craft ACT that “sought to sketch by means of the lineage of the Chinese chair, an understanding of design as being a trajectory of information through both time and space“. It also has a special place in the hearts of those of us who knew George and mourned his passing.
The chair is obviously making strong reference to the Chinese tradition of chair making but melds this with distinctly western techniques and some of the very recognisable vernacular of George Ingham.
There is considerable acknowledgement of the importance of ergonomics in recent western tradition. The profile of the back splat is improved over the typical Chinese chair so that the main support is provided to the lumbar region of the sitter. The angle of the seat and the gentle curve at the front make this a very comfortable chair. The woven seat with its generous width and the carefully placed continuous arm also make the chair a pleasure to sit in.
The construction is a delight to anyone who has wrestled with the severe demands of joining seat to chair legs and back in a way that will withstand the weighty guest who loves to rock their chair back on its hind legs (no one who has ever made a chair would do this!) while also trying to make the piece look elegant and light. The seat corners are mitred with three substantial splines so that there is a large area of long-grain gluing - large enough that a hole can be drilled through the corner joint to accomodate the leg.
The joints are all based on the “spindle in a socket” technique that pretty much defines the traditional Windsor chair that George would have been so familiar with. The solution starts with the stretchers in the undercarriage. This style of square timber transitioning to a thinner round that fits into a drilled socket is very characteristic of George’s work. This joint is used in one of his designs for a student piece - a small side table that teaches edge-jointing, mortise and tenon and these spindle joints. The style is then continued into the very clever joints at the corners of the seat.
Most chairs rely on carefully designed, executed and glued mortice and tenon joints to provide the strength required for the joints between leg and seat and back. In this chair a hole is drilled through each strong splined corner of the seat and the round of the leg simply passes through each hole. Gravity and geometery lock the joint tight. This joint would survive any failure of the glue over time.
The round of the leg continues through the seat to become the support for the continuous arm. The jointing theme is continued here. The round is reduced to a smaller spindle which fits into a socket drilled in the underside of the arm.
The chair is completed by the shaped back support which is inset with three pieces of stone. The considered and beautifully executed transition between the top of the splat and the arm of the chair is typical of George’s thoughtful craftsmanship.
September 29th, 2008
A recent project (Sixty by six) used simple mitre joints on the four corners of the ‘cabinet’. I’ve used a technique for clamping this joint that I learnt from George Ingham.
Fun with clamping mitre joints
I’ve really got no idea whether this is a commonly used technique but I can’t recall ever seeing it documented and suspect it may not be widely used. I’m very interested to discover how other woodworkers deal with this problem.
Regardless of the type of mitre joint - whether it contains hidden dovetails, is the mitred corners of through dovetails or is a simple mitre relying solely on the glue - good clamping pressure is required for a pleasing visual result and a strong joint.
I often use simple butted mitres with no fancy jointing. Obviously this limits the strength of the joint as the gluing surfaces are half way between long grain (strong bond) and end grain (weak bond). In many situations this is not strong enough. But under the right conditions it is. A wood that has good general gluing properties (eg not oily) and has a reasonably open end-grain can provide a suitably strong joint at 45° to the grain. Clearly the stress that will be exerted on the joint is important which generally means this joint is suitable for smaller works. This joint relies entirely on the effectiveness of the glue so the gluing process is all important. The wood surfaces must be clean and perfectly co-planar. The glue must be fresh and strong. And the clamping pressure must be adequate - and this is the punchline to this post!
The vector of clamping force should be at right angles to the surface of the joint
It is very difficult to get a good clamping result unless you are able to exert enough force at right-angles to the surfaces being glued. Picture frame clamps in their various forms are very poor at this. Crossed sash cramps are very hard to set up effectively and still don’t provide the pressure in the correct vector. Angled clamping blocks that rely on friction (such as sandpaper-covered blocks) may work but are fraught with potential glue-up disasters and ultimately compromise the amount of pressure you can apply.
My method is to glue clamping blocks firmly to the outside faces of the elements being joined allowing clamping directly across the perpendicular of the joint.
First the choice of blocks. I use MDF because it is softer than the wood I’m gluing so unlikely to compress it. It also breaks up in layers which makes removal at the end very easy. The shape is a triangular prism, that is a right triangle in cross-section extruded to the exact thickness of the elements being joined. You must make the blocks large enough to accomodate the pads of the clamp when they are positioned so that the pads are centred on the desired lines of force.
Make the blocks large enough to accommodate the clamps
I use my normal yellow glue (Titebond II at the moment) to attach the blocks. Some makers use only a dab of glue and little pressure. I prefer a stronger bond and cover most of both surfaces and just rely on the suction pressure created by rubbing the surfaces over each other until they grab.
Make sure you leave the blocks long enough for a good bond before proceeding to the next step. I try to organise my work so that the blocks can dry overnight.
As always, do a complete dry run of the clamping setup before applying any glue to the work. Then glue up as you normally do - and use the mitre blocks to achieve solid clamping pressure across the mitres. You should be able to get strong joints and a minimal glue-line.
This glue-up required sash cramps for the tenons along the side as well as the clamps on the mitred corners
Cleaning up the glued mitre is quick and easy. I do it in three stages. First I remove the bulk of the clamping block. I use a bandsaw for this.
Remove the bulk of the sacrifical block on the bandsaw
I then remove most of the remaining material with a wide, thin chisel. This usually leaves a small amount of block material glued to the work.
Remove the rest with a chisel
Some material is left.
The remaing MDF and glue is simply planed off leaving a clean joint.
Plane away the remaining material and glue
The cleaned joint
clamping a narrow mitre joint
Gluing narrow mitres was illustrated in an earlier post.
I’m very keen to hear how other woodworkers solve this problem and hope that this technique might be of some help.
August 16th, 2008
The dovetails
This is part 3 of a series of articles based on the construction of The Wedding Present. If you’re only interested in my dovetail methods, just read this. If you are interested in the context of the dovetails, start here.
As it’s a while since I hand-cut dovetails I did a little practice work. (I’m a traditionalist and cut the pins first and mark the tails from the pins.)
 |
 |
| Pins: Pins cut in test piece |
Tails: Cut from the pins |
Tools
This lead to a re-examination of my tools. My dovetail marking guide is homemade from a piece of 1½” brass angle. While it has served me well for many years it has some drawbacks including not being as symmetical as I’d like. While searching for a replacement I came across the tools of Chris Vesper.
 |
 |
| Old: Old dovetail guide made from brass angle. |
New: Chris Vesper’s dovetail guide. |
I bought his dovetail guide and found it handles very well. The extension of the reference bar on each side of the guide makes the marking of pins near to the edge of the piece much easier than with my angle guide. (The red ‘X’ on the 1:6 guide is just to remind me that I should be using 1:8 in case I accidentally turn the guide around.)
Another acquistion: Chris Vesper’s ‘double square’
While I was at Chris’ site I picked up his marking knife (I’d also been just managing with a sharpened length of 1″ hacksaw blade for 15 years - still useful for some applications) and his ‘double square’. The latter is a small square that has a movable blade like a traditional large combination square. As well as being a beautifully finished tool that is a pleasure to use, the adjustable blade has proved its worth many times in the short while I’ve used it. Woodworkers will know the annoyance of putting a narrow board into their vise for edge-planing and then reaching for their 2″ square only to find that the edge of the board is just less that 2″ above the jaws. Rather than reposition the board in the vise so that the square fits, you can just shorten the blade of Chris’ double square to check your edge. The narrow blade shown in the photo interchanges with the larger blade. It is extremely useful for checking squareness in tight confines. This is exactly what you need when checking the squareness of your tails if you cut very narrow pins as I do.
New saw: Adria’s dovetail saw
I’ve never been quite happy with my Pax dovetail saw - mostly from its lack of balance in the hand. I was happy to discover on Chris Vesper’s site a maker of dovetail saws in Canada - Adria Saws.
Having received all my new tools I could no longer put off cutting into these beautiful pieces of wood.
Over or under?
The first decision in cutting dovetails is whether you mark up so that the pins and tails protrude slightly beyond the outside face of the joining piece or stop slightly short of the face. By far the most common practice is to allow the pins/tails to protrude by 1/3 to 1/2 of a millimeter. This makes planing the outsides flush very easy because you only need to remove the excess from the pins/tails rather than planing down the entire side. However it does mean that cramp-up is more complex because you cannot just cramp against the face but have to make a cramping block that will allow the protruding pins/tails to . well .. protrude.
My teacher, George Ingham of the respect for tradition school, always insisted on leaving the pins/tails just short of the face. After all if you can’t competently plane the face of your wood, what are you doing cutting dovetails! I’ve always been somewhat agnostic about this and tend to use the method that suits the project. I have also had a bet each way and let the pins protrude (because the cramping block for this side has plenty of leeway and lots of area to push against, and cut the tails just shy of the face because when you use narrow pins there is very little surface to cramp against. I also make the difference very small (like about 1/6mm) so that it is easy to clean up either way. The difficult grain in the pieces I was using meant that I didn’t really relish planing too much off the faces.
In this case I have chosen to cut the pins on the shorter side pieces so that the front elevation of the box shows only the narrow ends of the pins. The blockier tails show on the sides. When you use dovetail joints on drawers, the orientation is all important and the tails must go on the sides so that the inherent strength of the joint is used to resist the force of the drawer being pulled from the front. As there is no directional bias of forces on a box the choice can be based on asthetics alone. In this case I wanted the front to be uncluttered.
Cutting the pins
I mark the width of the joining pieces (+ or - 1/6mm) on the inside face of the pin pieces with a knife. The mark is transferred to the outside across the bottom of the piece (the small knife mark will be planed out when the base of the box is flattened. I use a fine pencil line on the outside as a guide until the pins are marked.
I lay out the pins on the end of the pieces using a dovetail guide and then extend the marks square to the end until they meet the knife mark on the inside and the pencil line on the outside. Once the narrow edge of the pins are marked on the outside face, the pencil line is replaced by a knife mark in the waste areas only.
Dovetails: Pins are cut and pared
The faces of the pins are then sawn and the waste removed with a coping saw. The base of the waste area is pared flat and square to the faces. I don’t believe in undercutting these surfaces in order to obtain a better final appearance (at least not deliberately!). If I learned anything from George Ingham it was to respect the integrity of the joint. This relates not only to the structural strength of the joint but also to the very reason we choose to hand-cut dovetails in the first place.
Marking the tails
The traditional rationale for cutting the pins first and then marking the tails from the pins is that the tails are marked onto the face of the tail piece. If you do it the other way around you have to mark the pins onto the end-grain of the pin piece. End grain is generally less able to sustain a clear marking line than face grain. This argument applies more to the softer woods used in traditional European furniture than it does to the hard Australian woods which can usually carry a fine line in their end-grain. Nevertheless I have always had better results cutting the pins first so I tend to stick with this approach.
Dovetails: Setup for marking the tails from the pins. The tail piece is held steady by a bench clamp (out of shot). The block to the left aligns the reference edges of the two pieces.
When marking the tails from the pins, it is always worthwhile spending a little extra time creating a stable and well lit arrangement. Trying to steady two pieces of wood and mark at the same time when shadows are working against you is a recipe for error. My setup for this job is shown in the picture above. The block on the left-hand side aligns the two pieces on their reference edges. A waste piece (with a very square end) is positioned on top of the tail piece so that its end aligns with the knife mark on the inside. Both pieces are held in place with a bench cramp.
Dovetails: The tails have been marked from the pins
Three possible marking tools are shown - a pencil, a scribe and a knife (No. 11 scalpel blade in this case). I favour (and used in this case) the scalpel blade as it can make very clean marks, has good access to the tight spaces and doesn’t restrict vision too much. As it is double-beveled you need to be careful to maintain a angle that makes sure the cutting edge is hard against the pin while still making an essentially vertical cut into the piece. A fine single-bevel, pointed marking knife would be ideal here but I’ve not yet found one. The resultant marks are shown in the photo.
Cutting the tails
Dovetails: The tails are cut from the marks on the inside face
The tail markings are squared across the end of the tail pieces. The marked lines are then sawn down to the width of the pin piece. It is inherent in the geometry of constructing a dovetail joint that the transfer of the shape of the pins is to the inside face of the tail piece. This means that the tails must be sawn absolutely square to the face because it is the fit of the pins to the outside of the tail piece that we actually see.
Paring the tails
Dovetails: Paring the tails
The waste between the tails is removed (in this case with a jeweller’s saw because of the small space). The base of waste area is then pared. Placing a light source directly behind the cutouts can help with paring a flat base to the narrow space. The following picture shows a pared shoulder. You can see the knife cuts around the edges extending about 1/3mm into the wood. These marks are used to pare the rest of the shoulder flat.
Dovetails: Paring the shoulders
Fitting
Each joint is then tested for fit and tuned if required. These are some shots of the cut joints assembled dry.
Dry assembly: The slight protrusion of pins and tails can be seen
Dry assembly: Special attention is paid to the fit of the top of the joint as this the most visible part of the joint in the finished work.
Dry assembly: A completed joint ready for glue-up
Pre Glue-up
There are a few things to be done before glueup - like fitting the bottom panel, the runners and the magic catch. These are described in the following posts. However there are also some tasks specific to the dovetails that I will describe here.
The inside surfaces of the four sides must be fully sanded as it is impossible to sand these surfaces well after the box is assembled.
And you need to choose a method for cleaning up the excess glue on the inside of the joint. There are several ways to do this including washing the excess glue away with water and a toothbrush after cramp-up but before the glue goes off. I usually have too much to do in the short time available during glueup so I prefer to use the following method.
Find a thin sticky tape that you are sure will peel off the wood without leaving any gum residue. It needs to be thin so you can get the edges right into the inside corners of the glueup. I used to use a brown packing tape that was ideal because it was very thin and very waterproof. The latter means that the excess glue will not even stick to the tape and is very easy to remove. Unfortunately all the packing tape I have bought in recent years leaves lots of residue when you peel it off. This is as hard to remove as the glue. I now use a blue masking tape made for painters. It is slightly thinner than ordinary masking tape and is easy to remove.
Pre-glueup: Inside of joint taped to handle excess glue
The plan is that you apply the tape to the inside faces (the ones that are hard to clean up after glueup), making sure you place the edge of the tape exactly on the intersection of the joint. The excess glue will flow out on top of the tape and harden there. After the glue is dry, you can just peel off the tape and remove most of the glue. Any remaining glue is usually a fine sliver which can be removed with a marking knife. The intention of all this is to prevent the excess glue soaking into the wood and marring the appearance of the piece after the finish is applied. (I don’t have an image of the box sides with tape applied but the accompanying image shows the same technique used on the box lid.)
Glue-up
I use a yellow PVA glue for glueups like this where high strength, longevity and minimal glueline are the main criteria. In this project I used Titebond’s PVA for the first time.
Glueup: The dry run for gluing the front-left corner
The first two joints are glued in separate sessions so that I have the maximum time to attend to each joint. (The time from starting to apply the glue to having the joint in the cramps under pressure should be less than two minutes - then you have a few minutes to check and adjust squareness.)
Glueup: The last two joints (the two at the front of the bench) are glued at the same time.
The remaining two joints have to be glued in the same session (dictated by the geometry of dovetails and most other joints).
It is essential to do a dry run for each glueup session to make sure everything fits and that all the cramps and cramping blocks are adjusted and ready to go.
Part 1 The present
Part 2 The wood
Part 4 The lid
July 13th, 2007
