Advantages to manual boost controllers
As a driver, it’s nice to get full boost as quickly as the motor and turbo can produce it when you’re at full throttle. With an MBC this is what you get. It’s not a night-and-day different, but you’ll see it in data logs. Especially in higher gears, if you go WOT at low RPM (below the RPM at which the hardware is able to make peak boost). Note that an electronic boost controller that supports per-gear boost control can do this as well - but the stock Subaru ECUs use one set of target boost and wastegate control tables for all gears.
As a tuner, it’s nice to be able to set your boost level just by turning a knob. When I was tuning y car, I got AFR dialed for idle, cruise and low boost (MBC set to minimum) with conservative timing, and increased boost little by little, tweaking MAF scaling as I went. When I was happy with AFR I went back to wastegate boost and then worked my way up again, setting timing at each new boost level, basically working my way to the right on the fueling and timing tables.
I think it’s simpler to tune one thing at a time. First fuel, because it’s a little bit unpredictable in the beginning. Then timing - you pretty much get the timing you ask for, so I like the idea of using conservative timing at first, and then raising it to the knock threshold when I know that AFR is where it should be. During those two stages, boost tuning just meant twisting a knob, which made it simple to go back and re-test different cells in the fueling and timing tables.
When those were done, I started tuning boost electronically, knowing that it doesn’t really matter if I overshoot (within reason) or undershoot, because whatever RPM and load combinations I hit have already been tuned.
But I’ve got the BCS and the MBC hooked up in parallel, so I can still turn the boost down to wastegate-spring levels if I want to re-test different curves. There’s a thread about the hybrid setup at NASIOC:
Drawbacks to manual boost controllers
1) You can get full boost with way less than full-throttle. Contrary to what half of NASIOC seems to believe, this is not harmful to the motor. However it can be a little bit annoying. When I want to accelerate gradually, I push the pedal down part-way, and hold my foot still, and the car accelerates… and boost builds, so over the next couple/few seconds it accelerates progressively harder. So I end up having to lift to keep acceleration in check. It’s no big deal with a 3076-sized turbo, but it’s there. I hear it can be a real nuisance with stock-sized turbos.
2) If I go WOT, get full boost, and then lift slowly, sometimes the turbo will surge. There’s a loud “chuff” noise and power drops more abruptly than you’d expect from just lifting the throttle. This problem stopped when I switched to electronic boost control. With electronic boost control, the wastegate opens when I lift off the throttle, so the boost tapers quickly enough to avoid the surge.
Keep in mind that while surge is dangerous at full throttle, it’s no big deal when you’re off the throttle. The horror stories about surge destroying turbo are true but they happen at full throttle, when the exhaust output is still driving the compressor at full strength and the motor can’t inhale quickly enough. When you get off the throttle, manifold pressure drops rapidly, exhaust pressure drops with it, so there’s not enough force on the rotating assembly to tear the blades off.
3) You can’t change the boost curve much. You can set peak boost, but the shape of the curve is mostly decided by mechanical factors. I devised a hybrid setup with the stock BCS (not the same as the hybrid setup described above) which let me adjust how much taper I got, but it’s not very flexible - if you really want to define the shape of the boost curve for yourself, electronic boost control is the way to go.
4) Boost controllers tend to leave you with more boost in very cold weather and less boost in very hot weather. With a good controller like the Hallaman Pro Rx that I was using, it’s only a 1–2psi difference. Other folks have reported 4–5psi difference.