wortdog
15+ Year Contributor
- 172
- 4
- Jun 23, 2003
-
Champaign,
Illinois
I know most people on DSM forums immediately cringe whenever the subject of twin turbos comes up, but I know some people might at least find my car interesting. Its definitely not a guide on how to make pretty welds 
My basic goals for the project are good power in the wide range of 3k to 9k rpm, quick turbo response, and about 600whp.
The base engine is as follows:
2.3 stroker, Mitsu crank
Ross 9:1 pistons
Manley rods
mildly ported head with Ferrera 1mm oversized valves
the original mediocre Crower 64415 cams
JMF SMIM
Q45 throttlebody
I started with a 2G exhaust manifold and welded a pair of 60mm external wastegates to it. Each wastegate is connected to two of the runners in the manifold. The wastegates output into a Y that feeds the secondary EVO3 16 mounted between the transmission and the radiator. The compressor outlets are kept separate all the way until they Y together(very atrociously) into a single 3.5 pipe feeding a blow through MAF and the Q45 throttlebody.
The secondary turbo IC piping has a one-way swingvalve in it to prevent boost flowing backwards out of the stalled secondary turbo at low RPM.
My initial theory of operation was this:
At low RPM, all of the wastegates in the system are closed and the car behaves as a normal single turbo car. Once a preset boost level is reach by the primary turbo, 20psi for example, the 60mm wastegates begin to open and start feeding more and more exhaust to the secondary turbo, which starts to spin up. At this point, the secondary turbo is not contributing anything to the boosted intake air, but just accelerating up to speed on its own. Once the boost from the secondary turbo matches the primary turbo, the swing valve opens and both turbos are operating in parallel to feed compressed air to the engine. At this point, the internal wastegates take over to regulate final boost, which must be a psi or 2 higher than the pressure the external wastegates opened at.
It doesn't quite work yet, however. Upon trying it out, the primary turbo spooled a little before 3k rpm to 15psi and held until around 5500rpm, when the sound of heavy turbo surging was evident. My problem, as many of you have figured out by now, is that the secondary turbo went way off into surge land, because it was trying to build boost into a closed volume. Eventually around 7500 or so the surging becomes violent enough to push the one way valve open, but its definitely not a good thing to be doing to a turbo.
My solution is to mount an external wastegate with very little spring pressure on the secondary turbo's IC piping before the one-way valve. The valve will bleed off air from the IC pipe back into the turbo inlet pipe(ie between the air filter and the turbo). The bottom side of the WG's diaphragm will be hooked up to manifold pressure, while the top side of the diaphragm will see the same pressure as the secondary IC piping. This way the valve will be the open while the secondary turbo is trying to spool, allowing air to flow and keeping the turbo out of surge. Once the pressure in the secondary IC piping is close to manifold pressure, the valve will shut, causing a pressure spike and a very minute surge until the one-way valve opens and allows secondary boost to reach the engine.
Unfortunately, I won't get a chance to test my fix, as I managed to somehow crack my transmission casing 3/4 of the way around the front diff housing. I guess that's the car's way of telling my to spend some money on the transmission too.
And what project thread would be complete without pictures. You'll have to excuse the the general 'roughness' (rusty pipes, wiring, etc) as I don't feel like making anything pretty until I've gotten it all working.
My basic goals for the project are good power in the wide range of 3k to 9k rpm, quick turbo response, and about 600whp.
The base engine is as follows:
2.3 stroker, Mitsu crank
Ross 9:1 pistons
Manley rods
mildly ported head with Ferrera 1mm oversized valves
the original mediocre Crower 64415 cams
JMF SMIM
Q45 throttlebody
I started with a 2G exhaust manifold and welded a pair of 60mm external wastegates to it. Each wastegate is connected to two of the runners in the manifold. The wastegates output into a Y that feeds the secondary EVO3 16 mounted between the transmission and the radiator. The compressor outlets are kept separate all the way until they Y together(very atrociously) into a single 3.5 pipe feeding a blow through MAF and the Q45 throttlebody.
The secondary turbo IC piping has a one-way swingvalve in it to prevent boost flowing backwards out of the stalled secondary turbo at low RPM.
My initial theory of operation was this:
At low RPM, all of the wastegates in the system are closed and the car behaves as a normal single turbo car. Once a preset boost level is reach by the primary turbo, 20psi for example, the 60mm wastegates begin to open and start feeding more and more exhaust to the secondary turbo, which starts to spin up. At this point, the secondary turbo is not contributing anything to the boosted intake air, but just accelerating up to speed on its own. Once the boost from the secondary turbo matches the primary turbo, the swing valve opens and both turbos are operating in parallel to feed compressed air to the engine. At this point, the internal wastegates take over to regulate final boost, which must be a psi or 2 higher than the pressure the external wastegates opened at.
It doesn't quite work yet, however. Upon trying it out, the primary turbo spooled a little before 3k rpm to 15psi and held until around 5500rpm, when the sound of heavy turbo surging was evident. My problem, as many of you have figured out by now, is that the secondary turbo went way off into surge land, because it was trying to build boost into a closed volume. Eventually around 7500 or so the surging becomes violent enough to push the one way valve open, but its definitely not a good thing to be doing to a turbo.
My solution is to mount an external wastegate with very little spring pressure on the secondary turbo's IC piping before the one-way valve. The valve will bleed off air from the IC pipe back into the turbo inlet pipe(ie between the air filter and the turbo). The bottom side of the WG's diaphragm will be hooked up to manifold pressure, while the top side of the diaphragm will see the same pressure as the secondary IC piping. This way the valve will be the open while the secondary turbo is trying to spool, allowing air to flow and keeping the turbo out of surge. Once the pressure in the secondary IC piping is close to manifold pressure, the valve will shut, causing a pressure spike and a very minute surge until the one-way valve opens and allows secondary boost to reach the engine.
Unfortunately, I won't get a chance to test my fix, as I managed to somehow crack my transmission casing 3/4 of the way around the front diff housing. I guess that's the car's way of telling my to spend some money on the transmission too.
And what project thread would be complete without pictures. You'll have to excuse the the general 'roughness' (rusty pipes, wiring, etc) as I don't feel like making anything pretty until I've gotten it all working.
.
One more thumbs up for the road. 
)
So now I'm running a fairly basic single spring, 1mm oversized valve head. The bores were pretty straight with the head torqued down, and the rings seems to have broken in just fine, so the machine work was good in my book.
