In Part 1 of his article (see Issue 35) Bob Lyell started on the front quarter panels to (as he put it) “gain experience and confidence”. Part 2 (see Issue 36) saw him moving to the rear quarter panels. In Part 3 (see Issue 37) Bob made a start on the doors and in particular the door edge flanges. Part 4 deals with the door skins and fitting them.
Over to Bob……………
Although the entire outer surface of the tub is formed of compound curves, making its shape so attractive, each curve is only slight and each panel so narrow that no special action is required. Clamping, forming and nailing the edges being sufficient to persuade the Aluminium to follow the gentle curves of the Ash in two directions.
However, the much larger and relatively unsupported door skins do require special attention, I cut mine some 3 inches’ oversize and took them together with the Ash frames to a craftsman who curved them to fit in both planes by using an English wheel.
I added 3 pieces of wood to the frame, a strip of plywood so it could be held firmly in a vice and 2 blocks for panel locating screws in what would be surplus material, vital because once work has started the Aluminium skin panel must not be allowed to move by a fraction.
Rolling the top edge over was the most difficult job yet. I studied the shape of the top rail because whilst it sounds obvious this is what I needed to persuade the now valuable Aluminium skin to tightly hug as I worked it around the edge. I blended the various curves into each other by gentle sanding until I was both satisfied with the shape and that both doors were the same.
Clamping strip set just away from the edge and 2 additional flush fitting blocks for panel locating screws.
Clamped tight and located with 2 additional screws, work started in the centre of the rail with the Aluminium being worked out to each end. On the rear half, like the rear quarter panel, the metal wanted to form a straight tubular roll but as the outer face of the door curves from front to back, the skin had to be encouraged to ripple and shrink as it went over.
Starting to roll the top edge over.
The very attractive and complex shape of the front half needed the most effort because the metal wanted to bridge the gap and once I started I was no longer able to see the wood surface that I was trying to follow. I thought I had the shape twice but each time found I could work the metal down further before I heard the reassuring sound that I had reached the surface of the wood. Having the other bare door as a reference was a big help, as was keeping the metal soft by frequent annealing.
Eventually rolled over, the next operation was to shape a hardwood former to clamp the skin tight against the top rail so the return edge could be cut to a substantial 15mm (for rigidity) and folded over ready for nailing.
Before final assembly it is important to appreciate that clinching the skin around the flanged edge serves two purposes, to secure the skin tight against the outer face of the ash and to secure the flanged edge tight against the aperture face of the Ash. Both are vital for a good door fit.
So the challenge is how to form the skin into a 90 degree return before folding over and clinching when the aluminium flange is not strong enough to hammer against. The skin must not be allowed to crease against the edge of the Ash as that error would be impossible to recover from.
I decided the answer was to make yet another tool to add to my collection, an Aluminium block which would clamp the flange against the door frame whilst providing a robust face to fold the skin against. The following illustrations show the process in six steps.
2. Clamped to establish the cut line
3. Aluminium block added
4. Folded 90 degrees
5. Folded beyond 90 degrees
To fit the door flange without damage, the Aluminium block required a filed radius to clear the inner corner (I used a thin packer instead) and a curved edge to match the curve of the door, its thickness was the same as the width of the flange. Later I would also step one end and rotate it 90 degrees to accommodate the reduced width needed in the tight radius corners.
First I snapped the hinge reinforcement panel into place and secured it with countersunk wood screws. Then I fitted the 2 flange edges, clamped them tight in position and nailed them into place
through holes drilled in their inner faces. The top of the front flange was trimmed level with the wood to allow the skin to overlap and at the same point the protruding flange edge was tapered into a tight radius.
Then I fitted the skin still oversize, clamped it down tight along its top edge and nailed it into place. With the door sitting skin down on a cushion the edge was marked and cut to size as illustrated in stage 2.
I annealed just the edge of the door skin, applying the flame to the outer face so as not to singe the wood. The skin distorted slightly but fortunately returned to its original shape as it cooled.
I cut out both door hinge slots, back to the rebate in the wood using a junior hacksaw and file, which provided the easiest starting point for folding the skin over, the straight steel section between the hinges. The 90 degree fold was achieved in several stages but without further annealing as illustrated in stages 3 & 4.
Next I tackled the other relatively straight sections which again were folded to 90 degrees, the skin remained in contact with the Ash and started to become much more rigid.
Working a straight section to 90 degrees.
Finally, I turned my attention to the sharp corners, carefully applying the same principles in small steps.
Stepped end used to fold the tight corners (one clamp removed for clarity).
Effectively the whole process was repeated but using a thin strip of steel to hold the flange tight against the aperture. By carefully directed hammer blows the edge could be encouraged to fold over much further without distorting the skin. This is illustrated in stage 5.
Folding the skin beyond 90 degrees, the flange edge proved strong enough.
To finish I had considered buying clinching pliers to complete the job but having tried hammering a small area against a hardwood block I decided to continue that way all around the door. The tight corners needed annealing again to help them shrink.
That left two final jobs, first the top front joint where the edge meets the skin, achieved by cutting the skin to a tight joint and TIG welding which did not appear to burn the wood. Second, the short section above the top hinge where the already tightly curved skin has to be turned over 180 degrees. By this stage I had invested so much time and effort that I had to find a safe way to make this final fold so I decided to make a cut through the curve, fold the flat portion through 180 degrees and TIG weld the cut edges back together. The weld filed down smooth.
180 degree return at top rear by cutting and TIG welding (cheating).
My thoughts upon completion are that I am glad that I did it, learned a lot about Aluminium, have a lot of respect for the craftsmen who produced the original bodies at a rate of 1 per hour (I wonder how many manhours per body?) and finally for anyone thinking of doing the same I hope that I have at least revealed what is involved.
Above and below: two ‘shots’ of one of the finished doors.
The finished skinned tub with the nearside door fitted.
Ed’s note: Bob is hoping to have the car finished by next Spring. Judging by the progress he has made so far, this seems to me to be a realistic timescale.
I’m sure that Bob won’t mind me saying that in skinning the doors he was able to refer to an original one from TC0750.