The more I chipped away at the subframe, the more I had that sinking feeling. So, rather than have the front end disintegrate halfway to breakfast, I got out the guessing sticks and angle finders to measure it all up.

It took me a couple of days to complete as the lines on the ruler moved about somewhat and it became clear in the drawing stage that they were up to their old tricks again - parts that were meant to slip effortlessly into an adjacent station weren't playing ball. However, after some head scratching, I realised that the problems were not all my own. A complex fabrication like this must have been put together on a jig, but the subframe wasn't exactly a precision assembly. Some of the internal pieces - the captive nut plates for instance - would have been pre-attached, though distortion would have been minimal. I suppose that fifty years of use might well account for the inaccuracies.

The hole centres were the most important aspect to get right. I turned up various sizes of plug and lightly centre-drilled their faces to give me a half a chance of getting in the ballpark.

These worked well and for the main structure I attached a protractor to the corner of the right-angled ruler to act as a datum and measured off the centres on their respective radials.

I couldn't think of another way of doing it. At the drawing stage I had to use a combination of radials and squared points to fit them all on the page.

It was a good plan to check up on my mischievous measuring equipment by producing a paper template before pressing ahead with the final drawing .

One of the steering rack mounting bolt holes was slightly out of kilter, but otherwise things seemed to be going quite well. Because the subframe was so crudely welded together at birth, some of the edges of the structure could only be guessed at and, as every time I picked the frame up to move or turn it over, more bits - sometimes worryingly large chunks - fell out. Clearly, I had only so much time to complete the task!

In order to get some idea of how the whole thing fitted together and how best to work out the sequence of assembly, I did a quick sketch of all the parts, excluding the captive nut brackets which could be easily located by their respective drillings. This would be handy for a chum in Amsterdam that I'd asked to model the subframe in Fusion. The bushings for the various bolts are made up of tubes with reamed collars on each end. Some of these I can turn up myself, but the big stuff will have to go to Chumley - that'll be a lot of sausages. I'll make up a jig from the original subframe and fine tune my arc-welding skills before I tackle the job. As for the more delicate items - the captive nuts and the bushing collars - I can choose to TIG.