While tuning my naturally aspirated 2.5L Subaru on the dyno i spent an extra $150 and did some part throttle pulls to find BSFC sweetspot. Since dyno pulls are expensive I only did 4 runs of roughly 15, 20, 25 and 40% throttle. I cannot vouch for the dyno's BSFC estimation accuracy but it used torque at wheels, injector pulse width, rpm, mas air flow and other data. the efficient region seemed to be at a very narrow engine load and speed of 1800rpm @5"Hg vacuum. Any more vacuum (or smaller throttle opening) or engine speed resulted in the BSFC suffering in an exponential manner. even lean burn mixtures and high ignition timing could not make up for pumping losses at high vacuum.
- 40% TPS run reported ~13.9 AFR w/full manifold pressure (no vaccum)
- lower throttle positions were in locked Torque Converter mode with ~14.3-14.7 AFR
here it is in a graph form:
