[texan]

These are all good questions, though we've been over some of them in one way or another before I'll try to hit the topics again here for your specific question.

As far as normal production car turbo engines go (at least gasoline turbocharger setups), there are three basic designs. The single turbo we see so commonly, the twin parallel setup, and the twin sequential alternative. Which works best is a long standing argument depending largely on engine size and layout, efficiency of exhaust routing, and intended use. A drag car for instance does not need a very flexible engine power curve, one can primarily focus on a small RPM range and setup the car accordingly to operate in said range. A street car contrastingly needs a very flexible engine with not to much emphasis placed on any one RPM point's power, but I suspect these realizations are something you already understand. So that leaves us with engine configuration and exhaust layout to cover, and exactly why one turbo setup may work better vs. another in a given application.

When you look at engine layout and exhaust routing, the problem of single turbos is that they don't work very well on V engine configurations. Here you have two banks of cylinders which almost certainly exhaust not into the center of the V, but towards the outer edges, meaning you have two exhaust routes separated by quite a bit of space and solid metal. Routing an exhaust system to tie the two together is not very thermally efficient, in that quite a bit of heat (energy) will be radiated into the engine compartment and away from the turbo's turbine blades. Neither is a good thing, and as such a single turbo system is from the manufacturer's point of view likely of out the question. So now you have two options, either a parallel or sequential twin turbo setup. The sequential actuall represents an even worse option here, since not only will both banks' exhaust output need to be merged, but they'll need to be merged into two turbos and not one. Hence the only real option is a parallel setup, whereby each turbo is fed by one bank of cylinders and in turn feeds it's compressed air back into the system. And if you look at what production systems are out there, this is exactly what you'll find. Its efficient in terms of exhaust routing and thermal losses, its fairly simple in overall complexity for this type of engine, and its very effective at increasing power output.

Alternatively, let's look at an inline 6 setup. You still have all three options, but again at least one can be looked down upon as a good setup for a production car with street use as it's primary function. The single turbo setup has a serious problem here... it ties each cylinder into a high pressure exhaust path. With inline 6's, the best possible exhaust manifold design breaks the engine down into two three cylinder setups. This is due to the overlap of ehxaust strokes which occur in an inline 6 configuration, where coupling all cylinders into a single high pressure exhaust tract means every cylinder's exhaust stroke is subjected to the highest possible pressures, and therefore doesn't function as efficiently. A better system would be one that functions as two three cylinders, where street oriented cam timing would nearly completely separate the pressure surge from one cylinder's exhaust stroke from any other. So it's not hard to see that some sort of twin turbo system is required here, but which is the best?

Well, that largely depends upon what the engine's power requirements are and what technology is available. Sequential twin setups were conceived in the early 90's, when the trickle down efects of Formula 1 turbo setups were still coming of age. In such a state, there were clear benefits to having one small turbo which spooled quickly and another to keep providing boost on the high end, but that benefit is quickly being reduced by further advances in technology. When the Toyota Supra TT and the 3rd generation Mazda Twin Turbo were being developed, turbo technology clearly showed an inability to produce a compressor which worked efficiently at both low and high RPM flow ranges. Therefore complex exhaust manifolds were designed where the overall engine flow would progressivley spool up one turbo and then another to provide sustainable boost amounts at virtually any engine RPM. Nowadays, however, parallel turbo setups can mimmick these low RPM boost characteristics while providing superior top end power, all in an overall simpler package with more efficient exhaust routing. Simplicity is the mother of reliability, so from a manufacturer's perspective the current choice is clear cut.

Lastly, take a look at an inline 4 cylinder. It is small enough not to get much advantage from 2 turbos, has all exhaust ports on on side and is often packaged into tight engine bays. What this calls for is a single turbo setup, anything else takes up too much space and the airflow levels required are easily within the single turbo range.

However, the single most important thing to remember here is that basic design is not nearly so important as tuning and execution. Any setup properly designed should work well, and all of them will have a positive impact on power output. Engine management is also a key issue, most Honda people haven't figured out yet that the stock ECU is completely incapable of handling a turbocharged motor. In fact, it's slowly becoming evident to many of us that turbocharged Hondas just don't make sense as a street car, they are too expensive and difficult to properly setup.