... this?

No, because this happens:

What was needed was a proper square tube bender used in The Olden Days, and I knew a chap not far from me who had the very thing....

... and with which we were able to complete both radiuses of the frames in one continuous piece. I first tried making the bends with heat. It sort of worked but, inevitably, in attempting to correct or tweak part of the curve, a kink would be generated. I think this happened because if the tube was unevenly heated, some sections were more annealed than others. And then the plywood template caught fire. I called an attempt.

The Rover has arrived! A handsome beast and the Doctor's Coupé body is the only one in existence. The doors shut with the reassuring clunk of a railway carriage door of old - it's a quality car. The 16/50 is rare enough - there are reportedly only nine known to the world, and the one in the Coventry museum is a saloon. The reason for the type's rarity might be best explained by a look at the engine.

At first glance I thought it was a twin OHC, but no, it's a pushrod engine though it's layout is eccentric. The inlet valves have conventional rocker arms operated by the camshaft on the nearside of the engine. Horizontal pushrods then reach across the engine to operate the rocker arms of the exhaust valves.

The drawings above will help to illustrate the configuration, and also the somewhat unusual steering arrangement. At the end of the column, the steering box is part of the crankcase. Then, a shaft runs vertically down from the box to connect to the steering arm. I haven't examined it too closely but I suspect it might have, in effect, two steering boxes. It's an engine that will need to be treated gently as spare parts will almost certainly prove unobtainable. It's a bit wheezy with a Zenith carb replacing its originally fitted Stromberg. There was an option of an SU whose body was at some 60° to the float chamber - I've never seen one, or anything like it before.

During the last aluminium welding job that I did, I went through two WP26 TIG torches, so I've invested in a water-cooled torch that I hope will last a bit longer. Taking the plug apart to rewire it for my particular machine, I noticed that there was no heat shrink protection on the two wires that share poles next to each other. I've alerted the supplier and hope that they might correct this potential problem - one I've had before with a switch that wouldn't turn off.

Some more progress on the magneto magnetiser. I've been given a site transformer to up the power on the coils. That should produce some Uri Geller results.