1 Detailed Dovetailing
1 Detailed Dovetailing
1 Detailed Dovetailing
Dovetailing
A practice joint, a Moorish dovetail and other info.
Table of Contents
Detailed Dovetailing ............................................................... 1
Some Surprising Information about Dovetails........................ 3
Tails or Pins First?................................................................... 8
A Set Of Practice Dovetails ................................................... 12
Dovetailing Detailed - A Practice Joint ................................. 14
Sawing Dovetails .................................................................. 16
Dovetail Tips in Pics - Sockets ............................................... 22
Dovetail Tips in Pics - Pins .................................................... 27
Where you start can matter ................................................. 32
Chopping up to a line ........................................................... 39
Scribing and Sawing Dovetails ............................................ 44
A Secret Mitre Dovetail Joint................................................ 48
A Secret Lap Dovetail Joint ................................................... 50
Dovetails For a Hinged Box .................................................. 52
Moorish (Bermudan) Dovetails ............................................ 53
Groovy Dovetails .................................................................. 58
Decorative Dovetails ............................................................ 59
Some Surprising Information about Dovetails
Notes from The Forest Products Research Laboratory tests (by KS
Walker) reported in Woodworker January 1958
Caveat
Note that this means that the bottoms of the lapped dovetail sockets
would be rounded, and the hidden parts of the tails likewise. Hence the
tests are not strictly comparable to hand-made joints (or router jig
imitations), but it seems (to me) reasonable to assume that the principal
conclusions will apply.
All dimensions other than the included angle were kept constant.
The width of the dovetails were constant at half-height. Board width =
6", thickness of tail bearer = 1/2", pin bearer = 3/4", pin height = 1/2",
pitch = 1", included angles were 15deg, 21deg, 27deg, 35.5deg. He
pointed out that the British Standard 1186, quality of workmanship gave
14.5 deg as the minimum included angle. It does (did?) not quote a
maximum angle.
Species
The applied load was between 1000 and 1700 lbs on a 6" wide joint
in mahogany. This was regarded as the point at which the joint would
have failed in service. By continuing to pull to complete destruction, it
was found that joints could bear between half and one ton.
At first slip (ie the above point) all failures were the same, ie
compression of the tails fanwise permitting the slip. At ultimate
destruction, there were several modes of failure.
Glued joints
Mahogany only used. Tested four days after gluing with an urea-
formaldehyde glue and separate-application hardener. All joints
regardless of angle failed by shear along the short grain, leaving wedge-
shaped sections behind in the sockets.
The surprise
An unexpected result was that the ultimate load for the larger angles
should be less when the joints were glued than when dry. They seem to
put this down to the wedging action of the dry joints putting the tails into
compression across the grain, thus increasing the shear strength across
the grain.
Tails First
Marking out
However, the marking tool will be working onto end grain, possibly
with a reduced chance of the tool following the grain.
It is again possible to scribe the lines for the pin sides on the outside
face.
To overcome this some people offer the tail bearer to the pin bearer
before they open the sockets. They insert their dovetail saw into the kerf
and with a backwards pull, use the leading teeth of the saw to scribe the
end of the board
Tickling up
Prior to the scribing stage it is prudent to check for any small
inaccuracies. This also means working inside a cramped space. If you
make lapped dovetails (half-blind dovetails to the Murricans) it is easier
to make the tails first.
Pins First
It is easier to make a serious mistake. You have to take care that the
broadest part of the pin is on the inside of the job, ie the face-side
Scribing the tails from the pins means holding the pin bearer upright
against the face of the tail bearer. This can be difficult if the job is a big
one, eg a bookcase on which through dovetails are visible from the
bookcase sides.
Since in this case you are scribing onto face grain, the marking tool
can be prone to following the grain.
There is usually plenty of room between the pins for the pencil or
other marking tool to to reach. The results of the scribing are easily
verifiable.
However, the scribed lines have to be made on the inside face of the
job - see opposite, above. Working from the inside face can mean that
you can be not quite sure about the quality of the kerf on the other side.
Tickling up
It is easier to see and rectify any small inaccuracies.
Right: Whether you make tails first, or the pins first, you will scribe
the marks from the surface that is not visible on the finished job. In the
'tails first' case, this will be the surface indicated on the diagram by the
downwards-pointing face-edge mark. In other words, the pins will be
marked onto the shape of the underside of the tail bearer, ie the part you
do not see when the job is glued-up.
This implies that the mating part must have sockets with flanks that
are at right-angles to the face side (but see my tweaking tip).
If you have seen the text in A Bookcase in Oak you will have noticed
that pins and tails can be made so that,until the surface of the job is
planed during 'cleaning up', they lie recessed slightly beneath the
surface, as also illustrated below.
It is usual for the pin spacing to be made closer towards the sides of a
carcase. This is a precaution against the panels warping over a period of
time.
I've shown fairly clumsy pins. For fine work they can be made
somewhat narrower.
Don't make them too narrow or you will find that starting the saw will
be very tricky. Also, if they are made too flimsy and the job turns out too
tight, they can be ripped from their sockets when the job is knocked
apart.
Dovetailing Detailed - A Practice Joint
It is best to use a fairly resistant wood. The fibres of many pines will
crush too easily when you try to pare the pin and socket floors.
American Whitewood (Tulip tree wood) or a mild oak or mahogany
would be suitable.
4. Use the dovetail templet to mark out the pin sockets (or tails if you
prefer).
Sawing Dovetails
Use a dovetail saw that looks like a tenon saw but has a thinner plate
with fine teeth. Use one with a closed handle. Complete beginners might
find a slightly blunt saw easier to use. (Less likelihood of jamming).
Some years ago I somehow got into the habit of holding a dovetail
saw this way. This brings the action is more directly in line with the
teeth, yet the vertical component of the cut is more controllable.
Make sure that the wood is as low as possible in the vice or it will
vibrate. If sawing tail sockets, you could try tilting the wood in the vice
so that the line in question is vertical, though some people say this
makes it even harder.
Check the lighting. The line should be between the light and the saw.
Grip with the thumb against one side of the of handle and the index
finger on the other.
Check that the tip of the saw, your wrist and shoulder joint lie in
the same plane
Grasp the top of the workpiece and rest the plate of the saw against
the tip of the nail of your extended index finger of your non-sawing (N-
S) hand.
Balance the sideways pressures between the saw and finger so that
the teeth will be in the waste but just grazing the side of the line.
(Whatever tool was used to create the line, you will have to saw to the
side of the line. 'Splitting the line' would be a serious error).
Lift the saw against its own weight and apply a featherlight touch,
otherwise the sharp teeth will dig in. If this happens you will then press
like fury and things will go wrong from there onwards. A forced saw will
probably follow the grain.
The conventional grip
Left: Make sure that the wood is as low as possible in the vice or it
will vibrate. If sawing tail sockets, you could try tilting the wood in the
vice so that the line in question is vertical, though some people say this
makes it even harder.
Check the lighting. The line should be between the light and the saw.
Grip with the thumb against one side of the of handle and the index
finger on the other.
Check that the tip of the saw, your wrist and shoulder joint lie in
the same plane
Grasp the top of the workpiece and rest the plate of the saw against
the tip of the nail of your extended index finger of your non-sawing (N-
S) hand.
Balance the sideways pressures between the saw and finger so that
the teeth will be in the waste but just grazing the side of the line.
(Whatever tool was used to create the line, you will have to saw to the
side of the line. 'Splitting the line' would be a serious error).
Lift the saw against its own weight and apply a featherlight touch,
otherwise the sharp teeth will dig in. If this happens you will then press
like fury and things will go wrong from there onwards. A forced saw
will probably follow the grain.
Start sawing at the back edge with the saw making only a very small
angle with the wood surface. Otherwise, if starting on the front edge, the
the gullets of the teeth tend to lock into the front corner.
Take your time. If you make progress slowly, the saw will not go
too deeply with each cut and the natural feedback process will enable
you to correct the saw orientation.
Once started, saw with the teeth inclined somewhat upwards. The
saw will work more easily because you are sawing more with the grain
than across it. Continue until you get to the shoulder line.
Right: Check the socket floors. Ensure that they are not hump-backed
and are square to the face side.
As the kerfs were cut, a small shake in the end grain caused this chip
to fall away. Since this was so tiny, it was glued with a water-soluble
adhesive to a slip of paper. Superglue was applied to the side of the kerf.
The paper was easily manipulated into place.
The paper was torn away and finally removed after the socket was
opened.
Note the correct grip on this jeweller's saw. It has been fitted with a
coarse fretsaw blade. A coping saw will serve almost as well, though the
set can be a little thick for some dovetail saw kerfs.
Some people prefer just to chop out the waste, though this can
sometime lead to voids in areas if the socket floors. Providing that these
are in the central area of the floors, it usually doesn't matter very much.
However, if the sides of the job have to be shaped extensively, it can
matter if your plane eventually reveals the voids.
Working from each side and starting just below the sawcut, gradually
work down to the knife/gauge line.
A thump from the butt of the hand is often sufficient, but a mallet
might sometimes be necessary.
With the chisel horizontal, and also working from each side, start the
chisel on the crown of the hump you have formed. gradually work down
to the knife line.
Ready for the final stroke or two, the chisel tip is just resting on the
witness markleft by the knife line.
Note: If the flanks of the pin sockets are not parallel and square to the
face side, an ill-fitting joint looms ahead. Whenever tweaking a cut, try
to tweak to a definite line. Just blindly fiddling and faffing will often
make things worse.
A sharp marking knife is being used to scribe a new line to pare to.
Some canny vertical paring will be needed. A good reason for not
having the pins too slender.
If they are too slender, you won't be able to get the chisel into the
space..
A proper paring chisel (thinner than the bevelled edge firmer chisel in
use) will be ideal for this job.
Dovetail Tips in Pics - Pins
Tails-firsters cut the tails, which form the sockets for the pins.
The pins are then marked directly by scribing inside the pin sockets.
Since so many things can go wrong when scribing the pins from the
tails, use a ruler to aid registration.
One trick to get a good fit is to put the ruler's edge a gnat's whisker's
distance below the line.
This practice joint is supported by the jack plane and held in place by
a weight. The rule gets the alignment right.
Here an awl is used for scribing. Some people make a slim knife
from an old hacksaw blade.
As with the pin sockets, start paring at the hump formed by the above
sloping cuts and level the tail socket floors.
If these arises are left square, as the sharp arises of the pins drive
into the socket, the top edges can crumble and be carried into the socket
with the pins.
Burnishing and compressing the pins is not really very good for
adhesion, so gingerly mate the two parts down about a third of the final
depth. Well, that's the theory!
If you leave waste on the ends of the boards, you’ll need to use a
notched batten to apply the closing force in the right places.
This is the stage at which very slender pins risk being uprooted.
Once the joints are fully home, remove the surplus glue. The water
will do wonders for the fit of a good joint, though much less if there are
considerable gaps.
For stripey mahogany like this you need a very fine shaving
aperture.
I worked ‘pins first’, following the usual routine of marking the pin
profiles and their flanks and then used a dovetail saw to cut down to the
tail socket base lines.
The force of the final blow was taken by the nib of wood you can see
just below the chisel.
As the cut progressed, the un-supported wood fibres bent and then
broke so that the fibres were torn out by their roots, thereby leaving a
void.
During the final cut, the fibres were therefore hanging over thin air, as
it were, so the force of the blow also made them break out. (From
experience I knew that the chisel would crash through, so I protected the
bench top with a chopping board).
The gap made by the fracture is very close to the outer face of the job.
Its border is formed by the knife line.
On full completion of this stage, all the socket floors (lands) will have
similar cavities. The fastidious worker might find them unsatisfactory.
Do They Really Matter? A City & Guilds or GCE (UK examinations)
examiner might, when looking at test pieces, be somewhat aghast at this
sight, but as far as the strength of a proper job is concerned, the answer
might be ’really not very much’. After all, we do know that end-grain
does not adequately bond, so the loss of gluing area need not be very
important.
What really can matter, is just where the cavities appear. You can see
that one side of this particular cavity is only a knife line’s depth from the
wood’s surface, ie the outer surface of a real workpiece. The outer
surface is, of course the surface that almost invariably has to be planed
after gluing-up.
The cavity is revealed as the plane cleans-up the outside of the joint.
Because the above cavity was so near to the surface, it only needed a
few strokes of the smoothing plane to reveal the kind of disaster you see
illustrated.
Had I started chopping from the outside face, the same thing could
have happened, yet the void would have been near the inside face of the
job.
This might take rather more time, yet you can still get some speed
with better quality surfaces and no cavities. Chop vertically, eventually
going up to half-way through from each side, but with the bevel facing
the base line.
Over the years I have seen various stratagems that, while still relying
on chopping strokes, painstakingly seek to overcome the cavity problem,
but I fear that I have not yet been able to make them work. Maybe I did
not try hard enough!
Chopping up to a line
I've been driving a 1/2in bevelled edge firmer chisel into a piece of an
Elm look-alike (Japanese Sen, Kalopanax.spp). The idea was to replicate
a basic woodworking task - chopping out waste.
But note that the face of the cut has still been driven slightly to the
right of the line. Perhaps the error does not look very great?
.
It shows the discrepancy that I think it would be glaringly evident if
this gap had appeared on a real job. This time, the error is about 0.3mm .
They show that the distances between the cuts were such that quite a
thick section of wood had to be sheared with each chop. This made each
cut had to work nearly as hard as cutting into virgin wood and with a
similar, though reduced, effect
Approaching the line in closer steps has enabled the chisel to finish
just where it was intended
An Inevitable Cavity. When you complete the operation by chopping
from the opposite face, the inevitable result is the sudden fracture of
the last few fibres leaving a cavity in the end grain of the job. End grain
does not take glue very well at all, so for some classes of work the loss
of gluing area does not matter
Sometimes one has to take care that a cavity is not revealed when
for example, producing a test piece for an examiner, or by operations
such as sawing a box apart to form a lid, or perhaps planing its sides for
some decorative purpose.
In soft or thin material you could actually get away with a minimum
of two cuts, but take heed of this pic, the outcome of a bit of simulated
carelessness. The first cut established the essential void, but the chisel
moved slightly downwards so the pressure of solid wood behind the
half of the bevel uncontrollably rotated the edge, so forcing a wonky
cut.
Scribing and Sawing Dovetails
One can come across unspecific arguments about ‘Sawing to remove
the line’ or ‘Sawing to split the line’ or ‘Sawing to the side of the line’.
What is best really depends on what tool made the line, and what the
line actually indicates.
There are, of course, several ways you can scribe dovetails. Here, I’m
using a pencil in the style where you make the pins first. (Using a pencil
is more difficult when you make the tails first).
I chose a slightly blunt pencil to show that the width of the pencil line
does not unduly matter, just as long as you remember that for a conical
marking tool, the important side of the line will lie slightly outside the
footprint of the end of the pin.
In this instance, I used the method that uses waste at the ends of the
workpieces.
Note: A snag with the pins-first style is that when you saw the
sockets, you must work from the inside face of the workpiece. It is really
desirable to work so that you are watching the outer face, the face that
the world will see. In tails-first mode, although this need not apply work
on the show face.
As the pencil finishes its stroke, it flicks away from its guide and
leaves a curved end to the mark
To be able to square the line across the end of the job, you need to
extend the mark to the very end of the job, ie gently over-mark it.
Strongly resist the temptation to make it bolder and therefore wider.
It is all too easy to strike a line and discover too late that it wasn’t
quite where it was intended it to be, so make a tick where you think the
line should start, move the square to the side and check whether you
have got it right.
Saw at a gnat’s whisker away from the inner edge of the line. The
right-hand kerf is just right, the other is a shade to far away from the
inside of the pencil line.
Actually, the joint fitted without any tweaking
There are advantages in starting a saw cut at the corner furthest away
from you. Some people start on the front corner, but this means that you
have to start with the saw exactly aligned so that it matches both the
horizontal line and the line on the vertical face. Unless you are in
practice it is just too easy to get one of them wrong!
When you work this way, you can get a good start with the kerf
firmly established in the waste. As you continue to saw and gradually
drop your arm, you first have to watch the horizontal and then the
vertical line - not both at the same time
A Secret Mitre Dovetail Joint
To visualise this joint think of it as a common dovetail joint overlaid
with a series of thin panels.
Notes: A magnifier can be very useful when marking out the mitres.
Use the end of a combination square to test both vertical faces and the
floors of the sockets
A Secret Lap Dovetail Joint
• Accurately plane all ends. For the time being, assume that the end
of the tail bearer will finish flush with the outer face of its mate.
• Mark the socket depths.
• Form the rebate at the end of the tail bearer. Use a shoulder plane.
• Cut the pins. Saw as far as you can at an angle to the face.
• Complete the tail sockets with a bevelled-edge chisel. Skew-ended
bevelled edge chisels are useful.
• Since end-grain adhesion is almost non-existent, you can undercut,
but don't overdo it.
• Use the end of a combination square to test both vertical faces and
the floors of the sockets.
• Scribe, saw and pare the pin sockets.
• Trim the end of the tail bearer to form the listel.
• Work the moulding.
• Dry assemble to test the fit. If the joint is too tight, plane a few
shavings from the inside face of the pin bearer. If too slack, start again!
Dovetails For a Hinged Box
When a lid is formed by first making a box and then sawing around
its perimeter, some of us like to show off by having a mitre in the
corners.
The edge mitres either return a groove or allow the upper edge to
be moulded.
Please note that the surfaces marked with an asterisk are at right-
angles to the faces of the sides.
Note the little quirk and bead formed on the bottom. This will
inevitably shrink over a period of time, so the quirk helps to improve the
appearance of the gap that will be formed. Since wood rarely shrinks in
length, you only need to put the quirk on the sides and not on the ends.
Cut the grooves first. You can then align the flanks of the half pins
with the grooves.
Moorish (Bermudan) Dovetails
I got the urge to try one after seeing a photo of a Spanish chest. I've
since seen a description of one in Fine Woodworking magazine (Issue
35, August 1982) where they were called (I think) Bermudan Dovetails.
The style is said to be of medieval Moorish origin.
Prior to shaping, the pin-bearing board looks like a common lap
dovetail. The adjacent face looks just like a set of through dovetails. Out
of habit, I made the trial joint without waste on the ends of the pins, but
with hindsight it would have been better to have had a sixteenth of an
inch or so of waste.
Stage 1
Scribe and saw the pins more or less in the usual way. Note that
some cuts should be shorter than others, ie they do not reach the
shoulder line of the tail socket floors.
Make the tail sockets just like an ordinary lapped (blind) dovetail.
Stage 2
Prepare a template for the curved shape. It must have a baseline
indicating the location of the tail socket floors. Precisely locate the
template and scribe the shaped lap from the template. Use a jeweller's
saw to rough out the curve. Since using a chisel would be rather tricky,
finish with fine files, taking care not to fray the edges. This profile must
exactly match the template. If you go a bit too far, alter the template to
match your curve.
Stage 3
Offer the pin bearer to the socket bearer in the usual way and cut the
sockets.
Note that at this stage the design allows the pins to just enter their
sockets. Assemble the joint as well as you presently can.
Using a magnifier, locate the template absolutely dead above the face
side of the tail bearer and with a sharp awl prick out the curve.
Stage 4
Start with upward sloping chisel cuts working down to the gauge line.
Then pare horizontally as near to the awl marks as you dare. Nerves of
steel are obligatory. Work to just leave the awl pricks showing
Alternately paring vertically and horizontally, fiddle the profile as
best you can with paring gouges and chisels. Check that the rebate floor
is at right-angles to the vertical face.
The awl marks (fortunately) swelled tight again after the glue
squeeze out was washed from the outside surfaces.
Though it might not look so, the end-grain is slightly lower than
the upper surface of the workpiece.
Groovy Dovetails
C' illustrates how the mitre 'returns' the rebate.
Take care not to make them too tight since I have known them to be
extracted by their roots when trial joints are pulled apart.
The upper joint has structural value since the extra grip at the sides
could help to restrain a board that might cup, though this might really be
a bit of an academic point.
Because the half-pins only receive pressure on one flank, they can
easily be bent sideways and possibly will split. Make them a bit more
robust than the intermediate pins.
Anyway, they look pleasing to some people and can add a distinctive
touch to your work - English Arts & Crafts designers used them.