bartomos,
First, I hope you're not taking any of this personally. I'm trying to add to the knowledge base and having a technical debate will bring out many thoughts, facts and ideas that no single person can contribute alone.
You misunderstood my comment about sophistication and seem to think I'm trying to elevate my own intellect. I'm not. To say that chassis design was not very sophisticated at that time does not mean, nor do I think, that the engineers of the day were not as bright or capable as today's (or myself). It's a matter of technology available at the time. You mention bridges so l'll take that as an example. Way back when (late 1800's -early 1900's) it was difficult to determine bending loads in indeterminate structures, so they designed the joints of members as simply supported ones (i.e. pin connections that could rotate) making the calculations possible. The result was a structure far heavier, costlier and less efficient than one possible with today's analytical tools, almost all of which are the result of computing power. They even used to have grease fittings at the joints because they were afraid of what would happen if the joint was subjected to a bending moment.
Chassis design is no different. You are correct, an engineer from one industry can transfer to another, the science and mechanics are the same. But it's still far better technology available today than back then. 60 years from now they will no doubt look back at today's analytical methods and tools and chuckle.
Yes, a bolted joint can be made to work, and you point out some definite advantages. The OP said he could weld (or had access to qualified welders and NDT methods to verify the quality), and welding will be more efficient (and faster) than bolting, hence my suggestion.
It is almost impossible (I don't know of one, but I'll leave the possibility open) to strengthen something by adding to it without simultaneously stiffening it. I should have been more specific and instead of saying "flex" said "twist". My mistake. You pointed out correctly that his planned engine may well weigh less than the original. So there is no need to strengthen or stiffen the frame to handle the weight. If there were, you are correct, the L, a C or even an I would do the job just fine, correctly placed. However, that engine can produce significantly more torque, tending to twist the frame along its longitudinal axis.
The L and C shapes you suggest adding all in fact do add to the strength and torsional rigidity, and your 5 yrs of success certainly prove you added enough to achieve your goal. They just take more material and weight to accomplish the same result as a shape that closes the C of the frame into some closed shape like a box or tube. For resisting twist, a circular tube is the most efficient shape. The use of a C shape for the frame is economics - it's easy to make by stamping it out of heavy gauge sheet, takes no welding, the competition was doing the same and drivers didn't know any better.
The concept of stiffening a chassis to improve ride quality is fairly straightforward: keep the frame and body very stiff and let the suspension do the work of creating a smooth, controlled ride. Using the frame to act as a flexing member is less than ideal as it is an undamped member, continuing to vibrate after a deflection (pothole) - not good. That's why God invented shock absorbers.
Today's automotive chassis engineers know full well how to design even stiffer chassis than they currently produce (think Formula 1), but the economics of costs vs. consumers' demands for ride quality vs. competitive offerings dictate the end result.
I'm not ready to eat that crow just yet, but it never hurts to fire up the grill.
