Let’s face it, you’re probably here thinking there is no way two injectors spraying into the snout of a supercharger can provide proper cylinder distribution.
When developing this system, I had the same concern. I tuned two cars back-to-back in the same week using a meth injection kit as a fuel source and knew I could at least do something better than that.
Fuel distribution was not a major concern to me in the beginning. I assumed if people were willing to hose their supercharger down with meth injection, they would choose an option controlled by the stock ECU doing something similar.
As the system progressed and the software developed, it became clear this was turning into something better than I originally anticipated. During initial AuxJection testing using a crude prototype, wideband data already showed almost no delta from bank to bank when the system was active.
I still wasn’t fully convinced, even with the data I had available. To properly validate the results, I reached out to Brian Tooley Racing to conduct controlled engine dyno testing.
AuxJection testing at Brian Tooley Racing engine dyno setup
Engine Dyno Equipment
Before we get into the test setup and results, it’s important to understand the equipment used.
The engine used during testing was a healthy built LT4 equipped with a Harrop 2650 supercharger. Each cylinder was monitored with a wideband O2 sensor and EGT probe in every header tube.
In addition, cylinder pressure transducers were installed in each cylinder to allow real-time combustion analysis.
Direct injection control was handled by a Holley Terminator X system, while the AuxJection plate was controlled using a Haltech ECU. Power was measured using a Superflow PowerMark II engine dyno.
Test Setup
At the time of testing, BTR was running the engine on C16 fuel with XDI +30% injectors, producing over 1050 horsepower. This combination did not have sufficient fuel capacity to run E85.
We began by swapping pulley combinations to reduce boost to a level where the engine could run full direct injection on ethanol, which resulted in approximately 8 psi of boost.
Even when running DI only, cylinder-to-cylinder variation is possible due to airflow distribution in the supercharger. To account for this, we adjusted per-cylinder injector trims in the Terminator X to normalize distribution before introducing AuxJection.
DI Only Baseline - Cylinder Distribution (Pre-AuxJection)
As shown in the wideband data at the marked point, cylinder trims were dialed in accurately.
Zoomed Wideband Data - Balanced Cylinder Trims
With a solid baseline established, we moved on to blending in AuxJection fuel. To simulate real-world operation with DI and AuxJection working together, injection timing in the Holley system was reduced to 160 degrees (instead of 180), resulting in approximately 30% AuxJection injector duty cycle.
AuxJection Active - ~30% Duty Cycle
Zoomed Wideband Data - AuxJection Active - ~30% Duty Cycle
As shown in the data, there was minimal impact on cylinder-to-cylinder variation with AuxJection introduced at this level.
However, a significant drop in supercharger rotor discharge temperature was observed, over 40°F at peak, which helps explain the increase in power seen in the dyno results.
Corrected Power - DI vs. AuxJection
For the final test, we returned to the previous pulley combination that produced around 1050 horsepower on C16 fuel.
We adjusted the Holley settings to use a DI injection duration of 180 degrees (360 SOI / 180 EOI), allowing AuxJection to supply the remaining fuel demand.
This resulted in approximately 80% duty cycle from the AuxJection injectors, supporting 1130 horsepower and 830 lb-ft at around 16 psi of boost.
AuxJection High Load - ~80% Duty Cycle
Zoomed Wideband Data - AuxJection High Load - ~80% Duty Cycle
This final test confirmed that AuxJection can support significant power while maintaining consistent cylinder distribution.
Conclusion
I live by the idea that any claim anyone makes, including myself, is not factual unless there’s data to back it up. Until there’s data, it’s just an opinion.
I understand why people would assume this system could have distribution issues, and that’s exactly why I made it a priority to collect real data so I wasn’t releasing a product based on assumptions.
Hopefully this gives you enough data to see the facts, whether it lines up with your opinion or not.
In the words of Vice Grip Garage Derek, “You don’t gotta believe it, but you’re lookin right at it.”
If you have any doubts, questions, or concerns about the data above, don’t hesitate to reach out: caleb@wisherperformance.com
I’ll always entertain a good, honest discussion. Thanks for reading.
– Caleb Wisher