Butch -
while a larger turbo has the ability to move more cfm efficiently, it doesn't automatically do so. It only moves what the engine can take.
Take two idnetical engines. Assume that the rpm, boost, IC outlet temperature, and exhaust backpressure are the same. If engine #1, with a stock turbo, is pulling 500 cfm through the turbo, then engine #2 with a TE-63 must also be pulling 500 cfm. That TE-63 may be capable of moving 900 cfm, but just because it CAN do it doesn't mean that it WILL do it. To get use of that 900 cfm flow capacity, you have to rev the engine higher or run more boost.
So, why could there be some gains when bolting on a bigger turbo at the same boost level? Charge temperature is one thing, but I don't think it is as big a player as most people seem to think.
Say you go from a 65% compressor to a 75% compressor at 20 psig discharge. The temperature at the IC inlet may drop from 335 F to 300 F. Which is nice, but then it goes through the intercooler. That 35 F improvement at the IC inlet does NOT make for a 35 F improvement on the outlet. It will be less. Maybe a 10 F improvement, maybe 20 F, something like that. How much power is that worth? The old saw about 10F = 1% hp holds true here at the IC outlet. Actually I think 1.5% is closer. So on a 350 hp GN, that 20 F (if it really is that much) would be worth about 10 hp (a little over 5 hp per 10 F drop).
So, temperature effect alone, a 10% increase in efficiency might be worth only 10 hp. That doesn't explain much.
I think the biggest gains are exhaust side. A better turbine wheel and turbine housing are just like putting on a better exhaust system, losing your converter, going with a THDP, etc... All these things reduce the exhaust backpressure, which leaves less exhaust in the cylinders, which gives more room for fresh charge, which increases power. This will increase the air flow through the compressor and into the engine even though the boost level, rpm, temperature, etc stay the same.
A more efficient compressor wheel takes less hp to turn it, so less exhaust back pressure is required by the turbine wheel to give that power to turn the compressor wheel, which helps out more. A more efficient turbine wheel and housing also require less backpressure to provide the driver hp, which helps too. This is the reason why a 0.82 AR turbine housing would make more hp than a 0.63 housing at the same boost level/rpm/etc. This is the reason why a PTE-52 might be expected to outperform a TE-60 (they both have identical compressor sides by my understanding, it is the exhaust sides that are different). Think about the system as a whole, don't just focus on the compressor side. A big compressor attached to a little turbine isn't going to run mcuh (if any) any better than a small compressor attached to that same turbine.
John