BSFC challenge

Brake specific fuel consumption (BSFC) measures how efficiently fuel energy is converted into engine shaft power. Wiki does a fairly good job of defining the BSFC but one common practice is to measure this with the engine on a dynamo, full-throttle, and adjusting the load. This typically results in a graph like this:

Each of the enclosed circles is the range of how much fuel is burned per unit of power. Typically, there is a line, the operating line, that shows how an engine is expected to perform in a given application (aka., the vehicle.) Change the vehicle and a different curve shows up.

The seminal Toyota SAE paper, SAE: 2004-01-0064, "Development of New-Generation Hybrid System THS II - Drastic Improvement of Power Performance and Fuel Economy," Koichiro Muta, Makoto Yamazaki and Junji Tokieda Toyota Motor Corp. 2004 SAE World Congress, Detroit, Michigan, March 8-11, 2004, has two, BSFC charts:

This shows the nominal operating range expected for the NHW20 Prius engine.


This second chart shows some operational data. It differs in showing that sometimes the engine load changes, primarily "heretical mode." But it also collects transients, the bane of sampling.

A couple of years ago, I used a Graham scanner to monitor engine performance over a full tank of gas, over 500 miles. I then reduced the huge file to a smaller set to plot my best approximation:

Now Excel is not very good about drawing areas so as a quick hack, I used 'bubbles" of different sizes to stack the relative BSFC. What is interesting is the small 'red dots', the highest efficiency metrics, overlap some of the formal BSFC data but also extends quite is outside on the low power range.

One hypothesis, speculation on my part, I suspect ordinary vehicle operation involves partial throttle operation that is not evident in the standard BSFC charts. It may be that partial throttle operation shifts the high efficiency, engine operations into lower speed ranges.

A second hypothsis is that "heretical mode," or as Toyota calls it energy recirculation mode, may change the load characteristics. This may also change the peak BSFC regions.

Now I have a long trip coming up over some hilly terrain. I'm starting to think about how to gather a new set of data to clarify what is going on. Right now, I'm thinking of using my Graham scanner to record:

• mass air flow - for fuel consumption
• ICE rpm - for half of power production
• MG1 NM - for other half of power production
• throttle position - for BSFC mapping
• MG2 rpm - for speed and transaxle mode
• MG2 NM - for power, regeneration and battery assist

On the route, I am also planning to record my GPS ephemeris. Within the limits, I will get altitude data to help map the load.

I'll have to manually record the temperature but a simple log will work. The only challenge is the 'heat island' around each city.

The only other element I'd like to measure is E30 performance versus winter blend, Shell. What I'd like to do is find an E30 fuel station along I-75 between Cincinnati and Detroit. Any recommendations?

I'm open to suggestions on other approaches. Mostly, I'm out to get a better understanding of BSFC as we find in the field versus the SAE papers.

I will have a fresh oil change to 0W-30 and transaxle oil change, Type WS. The tires will be at maxium cold pressure, rotated, and alignment checked. The engine air filter was changed in the October.

Thanks,
Bob Wilson