Tuning Wiki

This is a historical revision. Read the current page or return to its history.

Page organisation in this revision
Page type
Technical article
Subject
Tuning
Parent
Section level
Applies to
No ECU-specific applicability
Tool groups
None
Manufacturers
Dodge

A 2011 Charger R/T 5.7 owner asked how to use fuel-trim information with the car's neural-network airflow model and related injector tables. A separate High Performance Academy worked example uses HP Tuners on a 2020 Challenger R/T Shaker 5.7. These are individual examples: confirm the actual controller, calibration and available functions before applying the workflow. 2011 Charger discussion; 2020 Challenger worked-example introduction.

Why cam position matters

In the Challenger example, the trainer explains that a neural network models volumetric efficiency with changing cam position. This differs from assuming one fixed VE table describes every operating condition. HP Tuners says the VE Neural Network Trainer is available for select 2010-on FCA and 2019-on GM vehicles. Open the exact calibration and check whether the trainer is offered; eligibility is not implied by engine family alone. HP Tuners trainer documentation.

Logging and training approach

  1. Save the original vehicle read. Verify mechanical condition, fuel pressure, relevant sensors and injector characterisation before treating fuel-trim error as an airflow-model error.
  2. Configure VCM Scanner for the available engine-speed, pressure/load, cam-angle and bank-specific fuel-correction channels. In the Challenger worked example, data is grouped by RPM, pressure ratio and the applicable cam position. Confirm that the histogram axes and channels correspond to this particular calibration.
  3. At warmed, suitable steady operating conditions, check that closed-loop control is active and collect enough repeatable samples. Review counts, outliers and temperature changes before using the data. A lightly sampled cell or a transient is not a reliable correction.
  4. For a supported calibration, create the neural-network training file in VCM Editor, process it in Tuner Tools, edit the applicable VE tables and export them for training. Import the processed result into the tune as described by HP Tuners. Some calibrations have multiple networks or many cam-angle tables; avoid applying a single cam table to all conditions.
  5. Recheck with new logs. Assess commanded versus independently measured mixture under any conditions where closed-loop trims are inactive, using suitable instrumentation and controlled testing.

The Challenger worked example discusses the logging and trainer process. It warns that changing the fuel-mass/injector-pulse-width relationship merely to mask an airflow-model error can affect other ECU behaviour. That is the trainer's assessment of this approach, not proof that every injector-table change is wrong. Use injector data appropriate to the installed hardware.

The 2011 Charger forum answer discusses inverse injector tables but its author says he was unfamiliar with that exact vehicle; it is not a verified universal procedure. Do not copy values or disable protection strategies from forum files. Check current ECU service, licensing and legal requirements in the HP Tuners tool listing.