The problem is that themodynamics are great. But when you have 400*F drops in exhaust temperatures on just a normal setup, and 800*F+ drops in "remote" setups agaisnt manifold ones. And yet you can still spool the turbo up jsut fine.
It's a 1% differance against 99% of everything else.
Sorry, I'm not understanding what you're trying to say, but I think you're missing an important point; the "total" energy extracted by the turbine from the exhaust stream is proportional "both" to the mass flowrate (deltaP) and to the change in temperature (deltaT) of the exhaust stream. In equation format this is:
Q=MCp(Ti - To)
Where:
Q = Energy in kW
M = Mass flow rate of exhaust gas
Cp = Coefficient of specific heat for the exhaust gas
Ti = Temperature(R) of the inlet gas to turbine
To = Temperature(R) of the outlet gas from turbine
The orifice equation for the mass flow rate is a little more complicated, but for sake of simplicity we can say it's proportional to the pressure drop across the turbine. This is the part that I think you were referring to as the 99% contribution, and the temperature drop was the 1%? Actually, as you can see from the equation, they work together.
We know it has to take "some" energy to compress the inlet air on the compressor side of the turbo, and just how much is defined by the equation above. If the compressor side is working really hard, at a high pressure ratio, there'll be a big deltaT and a big flowrate across the turbine. If it's not working too hard, there'll be a smaller deltaT and smaller flow rate (WG open?).
Sure, the further downstream the turbo is in the exhaust system (like in the STS system), the cooler the temps will be at the inlet, but the same principle applies. It doesn't matter if the inlet temp is 1600 degrees (R) and the outlet is 1200, or the inlet is 1000 and the outlet is 600, the same amount of energy has been extracted. Having said that, it's a lot easier to extract energy from hotter gas than cooler gas, so that's why most conventional turbos are located right at the manifolds.