>>How much of that 34 years was spent doing, for example, moldless composite construction?
I built two moldless composite structure wings for a successfull Top Alcohol dragster in the 80's. I scaled Rutans Long-easy canard design down using a different airfoil. The first one failed at the attachment points because tire shake was much more violent than we assumed (no real data available). The second was the one we used to win the TAD class at the NHRA US Nationals in 1988.
>>You can build a lot of composite structure for what it costs to buy a good TIG or heliarc welder.
True but you can also build a lot of steel structure for what it costs for weighing scales, fabric shears, brushes, squeegees, rollers, mixing cups, rubber gloves, peel ply, vacuum pumps, mold release, bagging materials, wire foam cutter, rheostat for said cutter and all the other sundry tools needed for composite work. All process require tools, some composites tools may be cheap but there are a lot of them and a lot of them are consumeables. And by the way TIG and Heliarc are the same process.
>>None of the steel tubing spaceframes in SCTA are certified based on any sort of engineering analysis. They're all one >>off designs where the only requirement is they use tubing of the specified diameter and wall thickness with the >>requisite number and placement of hoops in the rollcage.
Technically true, but the generalized tubing diameter, wall thickness, tube placement and reinforcement requirements can give the sanctioning body confidence that the structure will protect the driver in a crash. Such generalized rules don't exist for composite structures.