Aerial view of New Smyrna raceway. The red lines indicate the recorded position of the car
Once all laps were run, the team analyzed the data to determine both the performance and emissions benefits. Graph 1 shows the average time per lap. Upon examination, you'll see that the slowest average lap-time was recorded using the industry standard carbureted, race gas configuration. Over the half-mile oval, the average lap time was 19.2 seconds. The fastest average lap-time was recorded using E85 and EFI without catalytic converters, at 0.5-second quicker per lap. Even with converters bolted on, the E85 EFI configuration's average lap time was reduced 0.2-second, relative to the race-fueled and carbureted counterpart using no catalysts.
Referencing the May '10 issue of CT, dynamometer testing revealed that the E85-powered, EFI configuration generated approximately 7 percent greater torque across most of the engine speed range. Even with catalysts, the E85-powered, EFI version still showed greater torque benefits. At the time (given specific rpm ranges where torque was improved), it was projected that the greater torque would allow for quicker corner-exit acceleration, resulting in lower lap times. Coupled with proper gearing, the team felt that even further lap time improvements could be achieved.
If you compare the torque curves generated during engine dyno testing (see Graph 2) to the reduced average lap time shown, the results are conclusive. Quicker lap times were observed for all EFI configurations, with the greatest improvement coming from the E85/EFI configuration. Even with the use of catalytic converters, average lap-times were reduced from the carbureted, race fuel, non-catalytic converter baseline tests.
Graph 2: Engine dynamometer, full-load testing results at Mast Motorsports, November 2009
As mentioned earlier, one of our objectives was to produce the most consistent laps possible to make certain all data gathered shared common test conditions. The times shown were recorded with the on-board emissions measurement equipment (which added nearly 100 pounds to the Camaro while increasing its center of gravity height).
At the conclusion of on-board emissions testing, the equipment was removed, and the team conducted a series of maximum effort qualifying laps with the EFI/E85 combination for comparison with the carbureted/race fuel configuration. Lap times for the E85/EFI package were recorded at 18.20 seconds. By comparison, the carbureted configuration was again 0.5-second per lap slower, an improvement of more than 2 seconds over our five-lap run.
Graph 3: Relative A/F ratio measured for five laps and shown in Lambda units. Lambda is t
Several factors played a role supporting improved performance. Among them was an increase in air/fuel charge density based on the superior vaporization characteristics of ethanol. Another was the more favorable burn curve characteristics of ethanol. In addition were the benefits of individual cylinder fuel metering derived from EFI, combined with improved cylinder-to-cylinder air/fuel charge distribution and balancing. It was the integration of these factors that netted the overall gains. In addition, the wide-range O2 sensing that we used allowed for overall fueling rate correction with EFI, thereby closely maintaining ideal peak power A/F ratios. Recording real-time exhaust emissions enabled us to analyze the transient A/F ratio response of the car (on the track) and coincidentally analyze the excursion (transient period or spike as shown in Graph 3).
Comparing the various A/F ratio excursions, you can see that the carbureted/race fuel configuration shows significantly leaner spikes of duration longer than for the EFI system. Past the venturi, the metered fuel flow to each intake runner is a complex non-ideal flow pattern. Compared to intake air, fuel droplets exhibit greater momentum and have greater difficultly navigating bends and curves of the intake manifold and its runner. This mechanical separation of air and fuel results in a portion of the fuel wetting the intake manifold floor and walls. This fuel pooling or impingement inside and against the manifold results in transient delays getting air/fuel charges into the cylinders. In combination with unequal cylinder-to-cylinder air distribution (based on manifold design) this contributes to poor charge distribution among the cylinders. These effects result in the variations measured and contributed to the reduction in performance of the carbureted package.
Graph 4: HC emissions based on a five-lap average for each engine/fuel configuration. Uni
Also, the severity and earlier occurrence of the lean transient spikes decreases during lap accumulation as you can see in Graph 3. However, the carbureted spikes still remained significantly higher than those of their EFI counterpart. This is most likely due to the intake manifold becoming warmer, resulting in a larger fraction of the intake manifold fuel film being vaporized and delivered to the cylinders. When the manifold was cooler, less fuel vaporized and the spikes were less pronounced.
The other piece of the puzzle was to assemble so-called "engine out" (pre-catalytic converter) emissions and the effectiveness of utilizing two different versions of catalytic converter configurations on regulated emissions (unburned fuel [hydrocarbons, HC], carbon monoxide [CO], and oxides of nitrogen [NOx]). Graphs 4-6 display five-lap averaged emissions for each powertrain configuration for data comparisons. The units shown are in grams of emission per mile traveled, the standard units of measure. This data was collected at the same time the performance data were measured.