Hi Mike,
You can be forgiven for thinking that there is a forest of wood in that roof but as I recall when I calculated the volume of timber there was only about 750 cu ft used to cut the whole roof which has an internal span of 20 feet, external span of 26 ft and length of about 56 feet. This roof has double wall plates with a corniced flying plate and ashlars. The timbers are so dense that it's almost impossible to see the rafter peaks.
The main reason for the movement in the roof is probably down to the insertion of a 3 flue chimney which cut through the wall plates on one side and hence also through some of the scissor couples. In due course the chimney was removed and the roof yielded. This roof is in a castle and it has a number of other roofs intersecting and leaning upon it which also add to the deformation and yet at the same time provide a little more longitudinal stability. The roof was originally plastered over the 7 internal cants and presumeably this must have started to fall off centuries ago due to the various deformations. The original barrel valuted ceiling was then torn down and a new suspended ceiling inserted at tie beam height. This removed more than half of the longitudinal stiffening for the roof since the external tile battens are set at 4" spacing whereas internal lath is probably set at about 2" spacing. The metal tie rods are interesting in that since each of the scissor couples are independant i.e. no purlins then the metal ties will tend to provide only localised point load type pulling on the rafters and then rely on the external tile lath to transfer this restraint to the rest of the roof.
Inspite of all this abuse the roof still survives as a very early example of the kind of roof under discussion.
Regards
Ken Hume
Last edited by Ken Hume; 09/11/09 07:42 AM.