Okay, I read some of this thread (but not all... I'm sick with the herp... I mean, cold--I got through page 2 or so). Brian, I think what you're goal is--this is an okay way to go, but there's a better way and you can easily make the uber flat torque curve, but you're not going to like what I have to say.
You should make your own sequential turbo manifold/system. If you do a compound turbo you're going to run into a few problems, especially when you want to get a specific torque curve. Since you're flowing the exhaust flow into a small turbo, then into a large turbo (st, lt, etc), you're actually removing a certain amount of energy from the exhaust flow in the st which will cause the lt to be under spooled. If you can by-pass the small turbo such that it's on pep the entire time the large turbo has matched the pressure output of the lt you will have a much flater torque curve than a compound.
The way I imagine the sequential manifold set up is by two wastegates. When the pressure of the small one hits the target value, the waste gate activates and dumps the wasted gas into the lt. lt reaches peak pressure and will be more efficient in it. For instance (and just as an example), here are some compressor maps from turboneticsinc.
As you can see from these two turbos the 60-1 will run out of steam just as the HP76 is kicking in. You then get more mass flow of air compared to the 60-1 and no bad will come of this. Since the 60-1 is just getting enough exhaust to stay on pep it will always be ready for action when you let up the throttle and the large turbo drops back out of its efficiency range, and you wouldn't have the energy losses from compounding the exhaust stream.
Put it another way: There are three different scenarios that the above performs.
1) The small turbo spools quickly providing the torque you want early on the large turbo dumps its slightly less compressed air into the intake stream of the small turbo to help the large turbo spool quicker.
2) The large turbo spools to the same pressure as the small turbo and you run with an increased mass flow of air
3) The large turbo exceeds the small turbo pressure and as a result the small turbo dumps its charge into the intake stream of the large turbo which is better at the higher rpm.
This and both turbos receive the full benefit of the temperature difference from the exhaust.
In fact, I have an idea about how to do both the intake and exhaust manifold.... I believe this very well is the only way to get the torque curve you want without the downsides of the compound turbo set up. There is another way you may want to look into:
Basically instead of pressurized air being pushed back into the intake, the turbo directly puts the energy on the e-shaft (through a series of gears to increase torque).
I found a pic to what I'm talking about
:
http://www.heat2power.net/en__benchmark.php