Hi Guys,
As Simon Cowal would say - "I knew that this was a good question" and the follow on questions and comments are equally thought provoking.
Re the original blind pegging question :-
There are some places in timber frame construction where a concious effort was made not to show pegs coming through to the interior of the building. This still applies to day and typically might include where a common rafter is pegged to a clasped purlin and where floor joists are tennoned into a spine (summer) beam.
Re mortice and tenon joints not being being employed on scissor braced trusses :-
This is not quite true. The scissors can and are joined using M&T at each end to the rafters but the ovelap joints are generally made with unpegged lap joints.
See photo of Bishop's Camera roof at Farnham Castle [dd 1375, Moir]:-
Scissor braced trusses with collars It is of course possible to use lap joints at the scissor / rafter connection points but one has to consider the sequence in which the rafters, scissors and collars are to be laid up, cut and assembled and especially where slightly twisted timbers are being used. This can lead to less than tight lap joints. These are some of the most difficult timber configurations to frame. The roof illustrated was a "cost not an issue" construction and employed only the best timbers to form 39 sets of scissor braced trusses set on 16" centres. The rafters feet are not pegged and simply butt into a shallow locaton socket on the wall plate with no overhang beyond the plate. There are no purlins or windbraces i.e. a "sans purlin" roof.
Re Ommitting pegsIt's probably worth examining the interupted butt rafters employed in an early 17th century butt purlin roof in Farnham, Surrey.
See :-
Butt Purlin Roof & interrupted Rafters You will note that the lower sections of the common rafters have been deliberately omitted where a lean too dormer roof has been framed. The lower rafters were designed to be mortice and tennoned and pegged to the underside of the butt purlin but the upper section of the rafters only to be mortice and tennoned with no peg. There is no ridge or windbraces in this roof. I have seen this on other buildings in this town so this practice of completely omitting the upper peg is quite deliberate. Your thoughts on this practice would be much appreciated.
One must consider how a joint is expected to work. It is not just about tension or compression it is also about moment connection. A M&T joint functions well largely as a result of the shoulders being pulled up tight by the use of a tapered draw peg. It is the shoulders of the joint that are important as this is where load is transmitted so even if a member goes into tension providing there is sufficient preload applied by the peg then the joint can continue to act as a moment connection. A single three quarter inch dia oak peg can sustain a double vertical shear load of nearly 10 tons which means that a significant degree of preload can be applied. Its worth keeping in mind that the preferred mode of joint failure is in peg shear such that tenon relish shear and mortice cheek pull out are avoided. Replacing a peg is cheap and easy.
Re Lap Joint versus Mortice and TenonIt is highly likely that Tim's surmise about the evolution of timber joints probably progressed hand in hand with tool development. This is true today with certain joints now being made on CNC joinery machines being impractical to achieve by hand (housed taper curved dovetail floor joist to beam joints).
That said there are still places where a lap joint is preferable e.g. as in the case of a cruck mantle beam (low tie cross beam). These joints are generally skew pegged and though the carpenter might anticipate that the lap shoulders will do most of the work it is more likely that this is actually done by the pegs since cross grain shrinkage will gradually loosen shoulder contact.
There is also the well known lap dovetail found at the end of a tie bean where an English tying joint is employed. These unpegged lap joints are known to shrink back and frequently come apart over time requiring the addition of some metal strapping to arrest any further movement. This is classically illustrated at The Olde Bell public house in Henley-On-Thames, Oxfordshire [dd 1325, Miles].
See :-
English tying joint failure ConclusionIt appears that carpenters were and probably still are prepared to rigidly stick to tradition regardless of whether or not there is a well established history of failure in service of a particular type of joint.
What is the alternative ?
Regards
Ken Hume
