What do dynamometer BSFC #'s tell me?

Brake Specific Fuel Consumption (or BSFC) is the ratio between the engine's fuel mass consumption and the crankshaft power it is producing. This makes it both a valuable fuel efficiency indicator and one more useful tool in gauging an engine's state-of-tune.

In the USA, the fuel flow for BSFC calculations is normally expressed in pounds per hour (lb/hr) while the output units, of course, are in horsepower (Hp). So, our standard formula for BSFC becomes: lb/Hp-hr . For an engine producing 200 horsepower, while guzzling 100 pounds of gasoline per hour, the equation would yield 100/200 = 0.50 BSFC. Unlike AFR (Air/Fuel ratio) readings which only reveal the mixture, BSFC data represents the power provided per fuel unit. AFR and BSFC are not equivalent!

The beauty of BSFC #'s are that they remain similar over a wide range of engine sizes (assuming both are of similar mechanical design and compression ratio). For example, a tiny one cylinder 50cc four-stroke and a 454ci V8 might both have a BSFC ratio of 0.45 lb/Hp-hr (when optimally tuned) at their respective peak torque points. Thus, if either of these engine's were overly richened, its BSFC might climb into the 0.55 to 0.65 range (because the fuel flow in our equation will be going up as the power is going down.

Consider the above example on a dyno that only displays horsepower. Without the BSFC data, it is harder to know that a rich mixture engine is the reason for sub-par output. But, when you also have the BSFC numbers in front of you, it aims you at the cause.

BSFC values all follow a hooked curve. At idle they run much higher - due primarily to the closed throttle pumping losses and excessive camshaft overlap. Minimum BSFC #'s occur at about the peak torque operating rage - the most fuel efficient (per Hp) operating point for an engine. As RPM increases towards peak power, the BSFC rises again, since more fuel energy is consumed just overcoming the speed induced friction and breathing restrictions.

Below are sample BSFC ranges for several typical engine types. Note, these are provided to illustrate relative BSFC behavior only. Realize that any change that improves the mechanical efficiency of the engine (e.g. a dry sump oil pan, electric water pump, low tension rings, lighter oil, etc.) will also reduce its BSFC values!
 

                     Engine                             @ 1,000          @ Peak          @ Peak
                      Type                                 RPM              Torque               Hp      
4-stroke (low compression, carburetor)       0.62                0.47                0.52
4-stroke (high compression, carburetor)      0.60                0.42                0.47
4-stroke (high compression, closed loop)    0.52                0.35                0.42
4-stroke (supercharged, carburetor)            0.75                0.50                0.55
4-stroke (turbocharged, closed loop)           0.57                0.45                0.50
2-stroke (low compression, carburetor)       0.85                0.55                0.60
2-stroke (high compression, carburetor)      0.80                0.50                0.55
Diesel 4-stroke                                         0.35                0.25                0.30
Diesel 2-stroke                                         0.40                0.29                0.34

To find the optimum fuel curve for an engine running on the dyno, you should experiment with richer and leaner air fuel mixtures at incremental steady state test points (e.g. every 250 RPM). Starting with a known safe AFR value (rich fuel map, jet, or mixture needle setting) record the Hp, temperature, and BSFC values. Lean down slightly and retest � you are looking for improved power.

Keep in mind that you can easily go too far and end up sticking a piston, etc. This is why you must listen for knocking while monitoring the EGTs, AFR, and BSFC values. Experienced tuners will push the envelope a bit to get a flash Hp reading � but then back off for a cooling period - before the thermal heat sink protection is lost.

Once the peak safe power mixture is determined for each RPM point you can plot a target BSFC curve for the engine combination. Actually duplicating that shape is easy with a mapped fuel injection system. For carburetor equipped applications you normally need to work out air emulsion tube air bleed sizes, tweak boost venturis, etc. and even then some compromises must be accepted.

Back on the road, dynamometer optimized curves often prove to be a bit too lean for a good driving feel. This is because during transient conditions, like moving the throttle or accelerating, mixture requirements change. Since a slightly richer mixture only depresses power a tiny amount, whereas a slightly lean mixture can cause a very noticeable miss-fire and bog, the best drivability compromise is usually to richen up a tiny bit. That way, objectionable lean bogs are avoided at only a small expense of power and fuel economy.

The overall dynamometer verified fuel curve is still valid; it�s just been made more tolerant with a little fattening. As you gain tuning experience with various engine and induction system combinations, you will be able to hit an acceptable derivability mixture with little or no road test time. Compare that with blindly flogging the engine to death on the track � trying to optimize the best power curve mixture.