For instance, do you think it's really RPM that's getting you, or the damage that naturally happens when you slam a soft boolit into fast twist rifling?
Goodsteel
To arrive at the correct answer you must separate the two distinct parts of ballistics which you and many others lump into one. There is "internal ballistics" which is what happens to the bullet in the barrel during ignition, acceleration and exit from the barrel. This is where you are mainly focused. As mentioned before; you can have a perfectly cast and loaded concentric and balanced bullet in a cartridge loaded into as close to perfect rifle as possible and yet it can be sufficiently damaged during the internal ballistic phase to cause some degree of inaccuracy.
In the other ballistic phase called "external ballistics" from the bullets exit of the barrel to target impact is where the centrifugal force comes into play acting upon those imbalances in the bullet. As we can see the external ballistic phase is completely different. For a given cast bullet at a given RPM the centrifugal force will be there and there is nothing that a perfectly made rifle can do about it or change it. However, as mentioned the more balanced we can launch the bullet the less imbalance that centrifugal force has to act upon.
In all my pressure testing of equal loads in different twist 223s and .308Ws I have yet to see real evidence (velocity variation, psi variation or any difference in the time/pressure curves) that a cast bullet is "slammed" into a fast twist barrel any "harder" that a slow twist barrel. The bullet is fully engraved and spinning at the rate of the twist at one bearing surface of bullet travel. That happens very early in the time/pressure curve well before the peak psi is reached and well before the full velocity is reached (at the muzzle). Bullets slamming into the fast twist rifling harder sounds good in theory but I've found no proof of it in closed breach rifles.
Why is it that a powder with a gentler energy profile seems to give better accuracy from cast boolits?
Because the slower time/pressure curve causes less obturation, slumping and/or collapsing of the bullet. Thus it causes less damage and imbalance to the bullet. Take an ingot of lead and push it across the bench 26" with a hammer. Then smack it the same 26" across the floor in one whack with the hammer.....see the damage difference to the ingot.....same principle applies. A softer slower time pressure curve will always cause less damage to the cast bullet during acceleration than a faster burning powder even though both may hit the same peak psi. BTW; even given the same peak psi the slower time/pressure curve will also produce the higher velocity.
Could it be that the very fact that faster twist rifling has a more aggressive engraving characteristics is the real reason the RPM appears to be a limiter?
As already discussed; nice in theory but it doesn't prove out........If it were the case then the slower twist barrels would maintain accuracy at a higher RPM since the bullets didn't "slam" into the rifling as hard and were thus less damaged/imbalanced. But they don't, accuracy in the slower twist barrels goes south at the same or even a bit lower RPM than in the faster twist barrels. Yes the velocity is higher in the slower twist barrels but the accuracy loss at pretty much the same RPM. That is the proof.
Have we just been hammering nails into trees?
Probably, at least until we separate internal ballistics from external ballistics and understand the causes and effects during both. BTW; there are 3 spate phases of ballistics; internal, external and terminal ballistics. All 3 are separate and have different physical forces affecting them. That is why when we combine 2 of them we can just be hammering nails into trees...........
Larry Gibson