Roger wrote:
The message
from The Natural Philosopher contains these words:
Both ridge and purlins only serve to stop the rafters moving sideways.
(Euler buckling etc,). They have very little structural effect beyond
that. The rafters transfer all the loads to a downwards and outwards
force at the eaves level: That's where you need serous material.
Sorry TNP but I don't buy that argument. Purlins are often very
substantial pieces of timber and with a heavy roof they need to be. Very
few roofs have a pitch greater than 45 degrees so if the load is
resolved into rightangle components only half or less is being
transmitted directly down the rafter.
So where else does it go?
Unless your purlins are actually PROPPED all the weight of the roof is
transmitted via the rafter ends. There *is* no other point of support.
Purlins are simply there to either stabilise the rafters, or if above
them, to transmit te roof weight TO the rafters.
If all that was needed to be done was to stabilise the rafters a dinky
bit of 4" x 2" (likely the same section as the rafters) nailed flat
against the rafters would do.
Thats geberally waht is used IME.
I am quite sure that if I nailed a few bits of 4 x 2 across my rafters
and then cut out the purlins most of the 20 or 30 tons of stone on my 30
degree roof would promptly join me inside the house. Rafters are only 4"
x 3" (or 4 x 2 where they have been replaced) but the clear span is
never more than about 4 feet (which is perhaps half what you might
expect for a lightweight slate roof).
Well I guess wasted a lot of time doing an engineering degree then.
Look. purlins are only there in mast cases to spread the load from the
roof weight TO the rafters
http://en.wikipedia.org/wiki/Purlins
Definitions of Purlins on the Web:
* A horizontal structural member spanning between beams or trusses
to *support a roof deck*. In slope glazing, purlins are the horizontal
framing members.
http://www.inspectorwebhosting.com/c...erms-m-q-2.htm
* The horizontal members of the roof structure used to *support and
attach the roof panels.*
www.ameren.com/Environment/adc_ev_glossary.asp
* Timber members spanning over trusses to *support cladding or
between trusses to support loose timbers.*
www.wolf-system.co.uk/glossary.asp
* are the horizontal roofing members, which connect bents.
www.trilliumdell.com/vocabulary/
* Horizontal beam along the length of a roof, *resting on
principals and supporting the common rafters or boards.(
http://www.bbc.co.uk/homes/property/...glossary.shtml
* In architecture or structural engineering, a purlin (or purline)
is a longitudinal structural member in a roof. Purlins support the loads
from the roof deck or sheathing and are supported by the principal
rafters and/or the building walls. ...
en.wikipedia.org/wiki/Purlins
In every case it is clearly specified that these rely ON the principal
rafters and the joists (binders) to support the weight. They act merely
to transmit the weight *to* them.
If your rafters are VERY widely spaced, they need to be substantial.
But apart from a 'starved horse' looking roof,removing them wont cause
the roof to fall down generally.
I repeat, in a conventional trussed roof, any longitudinal members are
there to stabilise the rafters in the correct (laterla) position, and
act as local beams to transmit localised stresses to them, in the case
of the ridge and exterior elements - like the tile battens.
All the weight of the roof transfers to compression in the rafters and
tension ion the binders, the triangulation so defined forming a net
downward load on the binder ends at the eaves.
There is some net bending force on the rafters as well, but a long
purlin does very little to affect this unless its supported at the
ends,. If less than substantial rafters are used, or widely spaced one,
the better approach is W braves from the off binder centers to midway up
the rafters, or vertical posts between the rafter centers and the binders.
Or more relevantly when using the roof space as a room, you can add
another set of binders at 'loft ceiling' level bolted to the rafters.