A Structural Engineer’s look at Timber Framing

By John Davis

Richard Harris in his excellent book ‘Discovering Timber Framed Buildings’ suggests that “it is generally misleading to analyze traditional carpentry using modern structural analysis”, James Bentwod in his SPAB pamphlet ‘Repairing Timber Framed Buildings’ only mentions engineering calculations in cases of eroded timbers and suggests a repairer with experience can “rely on an intuitive feeling for the structural stability of the frame he is working on”.

Do Structural Engineers have a contribution to make to preserving and even illuminating the function of timber framed buildings or should everybody (as one Conservation Officer told me he did), try to keep Engineers well away from old buildings?

I am probably too closely involved to give an unbiased answer to these questions but I can say a few things about timber framed buildings that have struck me when working with them.

It is said that the essential character of nations remains the same and Tacitus (the Roman historian writing of the German tribes who later as the Angles and Saxons came to England) said of their houses ‘They do not even make use of stones or wall-tiles; for all purposes they employ rough hewn timbers, ugly and unattractive looking’!

The English liking for timber frame has a long pedigree!

When a Structural Engineer who has worked on modern designs first encounters traditional timber framing the first reaction can easily be one of bafflement.

Why are the floor joists and rafters wider than they are deep rather than the other way round as in modern construction? (see Figures 1 & 2) – the traditional way would seem to be just be a real waste of timber.

Fig 1
Modern style timber frame construction (Early 20th Century Felixstowe).
Note sawn regular sized softwood timbers (generally depth much greater than width of joists and studs), and simple nailed joints. In this case studs are tenoned into plates, but later butt jointing with skew nails is generally used.


Fig 1
Traditional timber frame construction
(House at Kelsale) joists of width greater than depth housed into main beam with mortice & tenon joints.

Why are the wall plates and the rest of the wall framing timber so large?
Why are all the roof timbers so small – and usually – so distorted?

A little more investigation will reveal the intricacy of many of the joints – some have a complexity more reminiscent of those puzzles where you are given half a dozen elaborately shaped pieces of wood and are asked to make a cube, than the product of a workaday carpenter.

(see Fig 3 & Fig 4)

Fig 3
Traditional Timber Frame Construction Main Beam /Post joint, formerly jettied (post now removed)
House at Kelsale


Fig 4
Diagram of traditional wall plate scarf joint.
House at Fressingfield.

As well as the complexity of the joints, they appear far stiffer than needed to resist the likely loads on them. Richard Harris said that a giant could lift up a timber framed building and turn it upside down without breaking it up. The obvious question is why on earth would they be constructed like this? Even in the remoter parts of Suffolk giants have rarely been known to invert buildings!

The English, and dare I say especially their businessmen and farmers, are not known for spending more money than is strictly necessary on their buildings, except when using the building as a symbol of status, or in present day parlance giving it ‘pavement appeal’. It should be remembered that when built, most buildings will be erected to fulfill a function at a price, and nowadays a great deal of ingenuity and hard work is put into finding easier and cheaper ways of making houses. A good example is trussed rafters – a brilliantly cheap method of putting pitched roofs on houses, (but a method unlikely to last for a thousand years).

So were master carpenters simply ignorant in the middle ages constructing heavy and elaborately jointed timber buildings, or were they so bound by tradition they could see no other way of constructing a house?

I am sure the answer must be no. At any point in history, the most economic solution to a problem will depend on the availability of materials and the technology available to process them. In the time of the dominance of timber framing up to the 18th Century, oak (which was almost invariably used for timber framing) was plentiful and other timbers were unlikely to last as long in the conditions of use, (no reliable ways of keeping moisture away from the outside of timber frames were available).

However the means of converting the oak into houses were hand tools, saws, axes, and adzes; and oak could only be easily worked with these tools when green (i.e. before drying out from its natural moisture content to the lower moisture content likely within a house).

The reason for the elaborate jointing I suspect is hinted at in the Spring 2001 issue of the “Eavesdropper” in an article by Alan Bayford reporting Andrew Moore’s talk on the Management and Conversion of Timber for Buildings. He describes how timber twists as the material dries and this locks mortice and tenon joints together. This is correct as far as it goes, but I believe the real reason for the elaborate joints is to prevent the timber framing from twisting and turning out of line while this shrinkage takes place; the locking together of the joints is a way of achieving this. Some of the timber members then end up seemingly oversized simply because they are sized to allow the construction of the complex torsion resisting joints rather than simply to bear the loads applied to them.

(see Fig 5 showing a traditional torsion resisting rafter / plate joint and Fig 6 showing the equivalent modern detail).


Fig 5
Diagram of Traditional rafter/ wall plate joint.
The dovetail on the underside of the rafter allows compressive force in rafter to be transferred into the wall plate and provides torsional restraint to the rafter.
House at Fressingfield.


Fig 6
Diagram Photo of typical modern cut roof rafter / plate joint – rafters are notched over the top of plate with no torsional restraint.
(early 20th Century Felixstowe)

Even today Structural Engineers know that when designing for instance timber roof trusses, the critical thing is to get the joints right, and often if there is enough timber to adequately form the joints the remainder of the structure will be satisfactory.

I was recently asked to look at a failed jettied gable built onto a house around the 1950s. Here a large oak beam spanned between two concrete brackets and a blockwork wall of about 5 metres height was built off it.

The oak beam had twisted by at least 10 degrees over its length and the twist was sufficient to lift the beam at one end, so it was bearing on one corner on the supporting brackets rather than bearing on edge, in spite of the weight of blockwork on top of it.

This gives some idea of the forces within the timber twisting it, which have to be resisted by the framing joints.

Certainly when you look at the traditional joints – mortice and tenon, lap dovetail, the splice joints used in wall plates, they are all torsionally stiff i.e. they are able to resist this twisting effect.

It would be interesting in due course to examine the function of each joint in detail and hopefully a later article will attempt this.

Of course resisting shrinkage and twist is not the only function of the joints in traditional timber framing, they also transfer loading through the frame and help tie it together against loads including wind loading and outward thrust from roofs.

What are the lessons to be learnt from this? Firstly, as well as the beauty and historic interest to timber framing there is also a structurally

The elegant and appropriate use of the materials available, and the craft tradition refined over many hundreds of years gave elegance and intelligence to the way timber was turned into buildings. What at first appears a heavy handed way of using oak turns out to be far from this.

Secondly, when repairing timber framing, the repairs (especially if in green oak) need to respect the way the building works as a structural frame, and to be jointed in such a way as to lock together with the remainder of the framing and not allow it to subsequently distort, or twist off line as the green oak dries out. This is where a Structural Engineer’s knowledge of the way timber behaves when loaded and how frames work can be put to good use in helping to produce appropriate repairs to existing timber framed buildings – or, for that matter, help in building new ones in the traditional manner (fig 7).