*
Saturday afternoon. I re-fixed all the flashing using Torx wood screws to lock them safely into
place. The screws went through the thin aluminium as if it wasn't
there. I finished off with the tin snips to tidy the overlaps. Now I
have to see if it will support the heavy rubber. I shall drill though
the nose section of the flashing and hold the rubber sandwich with small bolts /screws
and Eternit "washers." 
These are aluminium domed washers
with a roofing felt backing. They are used for fixing fiber reinforced
cement roofing panels to their supporting battens. I just happened to have some left over after a
roofing project. They will be ideal thanks to the large area of pressure
being located right on the outside of the circle formed by the washer.
The felt inner washer will stop the sharp aluminium edge from
penetrating the rubber like a circular cookie cutter. It all sounds so
easy!Only practice will tell if it works. Right now I have no idea how I will hold up the long strip of heavy rubber while I drill and bolts it to the flashing. I may end up having to use short, separate lengths of about 62cm or 2' for each of the 16 sides. Though I'd rather not. An overlap between short sections might help to keep the rain and draughts out.
I can always remove the flashing sections to attach the rubber then replace each one in turn. Or, I could choose a manageable length of several/multiple sides. Which can be temporarily supported with timber props off the veranda floor while I bolt or screw it together.
I have made no attempt to fix the flashing sections together. [Yet!] Hoping to rely on the rubber to tie lengths to each other. It is pretty resistant to stretch over a 10" edge length. I have plenty of waste aluminium if it proves to need backing plates between flashing joints but hoped to avoid too many visible screw heads. Pop rivets are neater and do not project much on the outside but do on the inside. Which would deform the rubber. I can't use normal pop rivets for the roofing washers because the overall sandwich thickness is far too great.
The geometry of multi-sided shapes have the advantage that they resist deformation. However, the flashing sections are only supporting their neighbours if tied firmly together. Without any physical connection they will tend to sag downwards under a load. Resisted only by the bend in the thin, aluminium profile at the dome edge.
When the drip noses of the sections butt against each other the resistance to downward deformation increases dramatically. The same would occur with riveted links between flashing sections. Though in this case the geometric reinforcement is all in tension. Suggesting I really should add connecting plates if the rubber is unable to provide a link.I should flatten the nose piece sections to avoid cutting the rubber through with the sharp edge over time. Wind rock or mechanical movement might easily cause the sharp edge to wear through the rubber.
I have cut off most of the points of the octagon top ring to provide skirt clearance. I shall probably smooth the straight saw cuts off to reduce friction with the rubber skirt. The ladder and its supports are currently preventing me from reaching the front [southerly] points.With the timing of the sealing so weather critical I can't risk needing to rebuild the ladder just to regain access to the outside of the dome. Wednesday onwards should reach 65F with sunshine. That might be enough to dry out the plywood covering to allow me to go ahead with the sealing.
The trapdoor now has two sliding door bolts fitted. These ensure the heavy hatch cannot drop either way. I had been tying a loop of string cord around the hatch and ladder handrails. Which was a time consuming nuisance. The bolts stop firmly against the big stepladder's side rails. Preventing any risk of my being "brained" or crushed by the considerable weight of the larch hatch. Fitting two bolts, instead of one, is simply extra security.
Click on any image for an enlargement.
*
No comments:
Post a Comment