OBD2 CAN Logger

Capture, decode, and reconstruct your car's ECU maps from the OBD2 port — using an Adafruit Feather M4 CAN, an SD card, and a browser.

---

Overview

OBD2 CAN Logger is an open-source hardware + firmware + web-app project that:

1. Logs every CAN frame from your car's OBD2 port to an SD card in real-time (CSV + binary)
2. Streams live data to a Betaflight-style web dashboard over USB Serial (Web Serial API)
3. Decodes signals using standard OBD2 PIDs or custom .dbc files from opendbc
4. Reconstructs engine maps — fuel trims, ignition advance, MAF, MAP, throttle, RPM — as a baseline for tuning

The ultimate goal is to reverse-engineer and document the factory calibration maps of any OBD2-compliant car, so they can serve as a starting point for custom ECU tuning.

---

Hardware

Component	Part
Microcontroller	Adafruit Feather M4 CAN Express (ATSAMD51J19A)
SD card + RTC	Adafruit Adalogger FeatherWing
CAN Transceiver	Built-in on Feather M4 CAN — TCAN1051 on CAN1 (CANH/CANL pins exposed)
OBD2 Interface	OBD2 DB9/breakout cable wired to CANH/CANL

Note: The Feather M4 CAN (product 4759) has an integrated CANFD transceiver on CAN1. It is different from the plain Feather M4 Express (product 3857). If you have product 3857 you will need to add an MCP2515 FeatherWing.

Wiring

Feather M4 CAN          OBD2 Connector (J1962)
──────────────          ──────────────────────
CANH           ────────  Pin 6  (CAN High)
CANL           ────────  Pin 14 (CAN Low)
GND            ────────  Pin 4 or 5 (Ground)

The Adalogger FeatherWing stacks directly on top of the Feather M4 CAN — no additional wiring needed for SD or RTC.

---

Project Structure

OBD2CanLogger/
├── firmware/
│   └── OBD2CanLogger/
│       └── OBD2CanLogger.ino      # Arduino sketch for Feather M4 CAN
├── webapp/
│   └── index.html                 # Self-contained web dashboard (no server needed)
├── dbc/
│   └── (place .dbc files here)    # Download from opendbc or add your own
└── README.md

---

Firmware

Requirements

Install the following libraries via the Arduino Library Manager:

Library	Author	Notes
ACANFD_FeatherM4CAN	Pierre Molinaro	CAN/CANFD driver for the ATSAMD51 built-in controller
SD	Arduino	SD card read/write
RTClib	Adafruit	Real-time clock (PCF8523 on Adalogger)
ArduinoJson	Benoit Blanchon	v6+ — JSON serial protocol
Adafruit TinyUSB Library	Adafruit	USB Mass Storage support for SD card access over USB

Board Setup

1. In Arduino IDE, add Adafruit's board package URL:

   https://adafruit.github.io/arduino-board-index/package_adafruit_index.json
   

2. Install Adafruit SAMD Boards via Boards Manager
3. Select: Adafruit Feather M4 CAN (SAMD51)
4. Upload firmware/OBD2CanLogger/OBD2CanLogger.ino

Default Behavior

• On power-up, the firmware reads /config.json from the SD card (if present)
• Creates a timestamped session log: session_20240101_120000.csv and .can
• Logs every CAN frame to SD
• Streams decoded frames as JSON over USB Serial at 115200 baud
• Red LED blinks every 5 seconds as heartbeat

SD Card Setup

The firmware requires a FAT32-formatted SD card. Any standard microSD or SD card up to 32 GB works. Cards larger than 32 GB may be exFAT by default — reformat them to FAT32 before use.

How to format by OS:

OS	Steps
macOS	Open Disk Utility → select your SD card → click Erase → Format: MS-DOS (FAT) → Scheme: Master Boot Record → Erase
Windows	Open File Explorer → right-click the SD drive → Format → File system: FAT32 → Start (for cards >32 GB use guiformat)
Linux	sudo mkfs.fat -F 32 /dev/sdX1 (replace /dev/sdX1 with your card's partition)

Once formatted, insert the card into the Adalogger FeatherWing before powering on the Feather M4 CAN. The firmware will automatically create log files on the next power-up.

No card inserted? The web dashboard's Start Log and Stop Log buttons will be disabled and a red "SD: No Card" indicator will appear in the Live Data toolbar. CAN frames are still streamed live to the browser — only SD logging is unavailable.

Filling up? Use the web app's Configuration → SD Card Management → Format SD Card button to erase all logs and start fresh, or click Enter USB Storage Mode to mount the card as a USB drive on your computer and manage files directly.

SD Card Files

File	Description
/config.json	Device configuration (auto-created on first config command)
/session_YYYYMMDD_HHMMSS.csv	Human-readable log — one row per frame
/session_YYYYMMDD_HHMMSS.can	Binary log — 24 bytes per frame, magic header OBD2CAN\0

CSV Format

timestamp_ms,id_hex,id_dec,extended,fd,dlc,data_hex,data_bytes
0,000007E8,2024,0,0,8,04 41 0C 1A F8 00 00 00,8
4,000007E8,2024,0,0,8,04 41 0D 62 00 00 00 00,8

Serial JSON Protocol

Send commands as JSON terminated by \n:

// Get device status
{"cmd":"status"}

// Start a new logging session
{"cmd":"start"}

// Stop logging
{"cmd":"stop"}

// Update configuration (restart to apply bitrate changes)
{"cmd":"config","bitrate":500000,"logcsv":true,"logbin":true,"stream":true}

// List files on SD card
{"cmd":"ls"}

// Enter USB Mass Storage mode — SD card mounts as a drive on the host PC.
// Press Reset on the Feather to exit MSC mode and resume logging.
{"cmd":"msc"}

// Format SD card — deletes all log files, starts a fresh logging session.
{"cmd":"format"}

The device sends frames as compact JSON:

{"t":1234,"id":"0x7E8","ext":0,"fd":0,"dlc":8,"d":"04 41 0C 1A F8 00 00 00"}

config.json Example

{
  "bitrate": 500000,
  "logcsv": true,
  "logbin": true,
  "stream": true,
  "session": "mycar",
  "filter": []
}

Set "filter": [0x7E8, 0x7DF] to only log those IDs. Leave empty [] to capture everything.

---

Web App

Open webapp/index.html in Google Chrome or Microsoft Edge (Web Serial API required).

No server, no Node.js, no installation — it's a single self-contained HTML file.

Tabs

Tab	Description
📡 Live Data	Real-time gauge dashboard + scrolling frame table. Filter by CAN ID. OBD2 PIDs auto-decoded.
🗺️ Map Reconstruction	Live scatter plots: TPS/RPM, Fuel Trim/RPM, Ignition Advance/RPM, MAF/RPM, MAP/RPM, Speed/RPM. Export as JSON.
📋 DBC Decoder	Load any .dbc file (drag & drop) or use built-in OBD2 PID table. Shows live decoded signal values.
💾 Log Viewer	Open CSV logs from the SD card. Searchable/filterable.
⚙️ Configuration	Configure bitrate, session name, log format, ID filters. Apply directly to device.
🖥️ Serial Console	Raw JSON command terminal for direct device communication.

Connecting

1. Plug Feather M4 CAN into your computer via USB
2. Open webapp/index.html in Chrome
3. Click Connect USB
4. Select the correct COM port
5. Live data begins streaming immediately

---

DBC Files & Signal Decoding

This project uses the opendbc database to decode manufacturer-specific CAN messages.

Quick Start with opendbc

# Clone opendbc
git clone https://github.com/commaai/opendbc.git

# Find your car's DBC file
ls opendbc/opendbc/dbc/ | grep -i toyota
ls opendbc/opendbc/dbc/ | grep -i honda

# Copy to dbc/ folder in this project
cp opendbc/opendbc/dbc/toyota_nodsu_pt_generated.dbc ./dbc/

Then drag the .dbc file into the DBC Decoder tab of the web app.

Supported Signal Types

• Standard OBD2 PIDs (Mode 01): Built-in — RPM, speed, TPS, ECT, MAF, MAP, fuel trims, O2 sensors, ignition advance
• Manufacturer-specific CAN (via DBC): Engine maps, transmission data, ADAS data, EV battery data
• CANFD frames: Logged and streamed; DBC decoding uses the first 8 data bytes

---

Map Reconstruction Methodology

The web app reconstructs factory maps by correlating live sensor readings:

Maps Being Built

Map	X-Axis	Y-Axis	What It Tells You
Throttle Map	RPM	TPS %	Throttle body characterization
Fuel Trim Map	RPM	STFT/LTFT %	How much the ECU is correcting fueling
Ignition Map	RPM	Ignition Advance °	Spark timing at different RPM
MAF Map	RPM	MAF g/s	Airflow at different RPM (volumetric efficiency proxy)
MAP Pressure	RPM	Intake MAP kPa	Manifold vacuum/boost curve
Speed/RPM	RPM	Speed km/h	Gear ratio estimation

Recommended Drive Cycle

To capture representative maps:

1. Cold start and idle warmup (captures idle fuel trims)
2. Gentle acceleration runs 1000–6000 RPM in each gear
3. Cruise at steady speeds (30, 60, 100 km/h)
4. Deceleration / overrun (captures decel fuel cut)
5. Wide Open Throttle run (optional — captures WOT fueling)

Export the map data JSON after the drive for offline analysis.

---

Binary Log Format (.can)

The binary log is optimized for post-processing. Each file starts with an 8-byte magic header followed by 24-byte records:

Magic: "OBD2CAN\0" (8 bytes)

Per-frame record (24 bytes):
  Offset  Size  Field
  0       4     timestamp_ms (uint32, ms since session start)
  4       4     id (uint32, CAN arbitration ID)
  8       1     dlc (data length code)
  9       1     flags (bit0=FD, bit1=BRS, bit2=Extended ID)
  10      2     padding
  12      8     data (first 8 bytes; FD frames truncated to 8)

Parse with Python:

import struct

with open('session.can', 'rb') as f:
    magic = f.read(8)
    assert magic == b'OBD2CAN\x00'
    while True:
        rec = f.read(24)
        if len(rec) < 24: break
        ts, id_, dlc, flags, _, _, *data = struct.unpack('<IIBB2B8B', rec)
        print(f"t={ts}ms id=0x{id_:X} dlc={dlc} data={bytes(data[:dlc]).hex()}")

---

Roadmap

• OBD2 PID poller (auto-query Mode 01 PIDs at configurable rate)
• 3D surface map visualization (RPM × Load → value)
• Compare two log sessions side-by-side
• Export maps to CSV / MegaTune format
• Automatic gear detection from speed/RPM ratio
• CANFD full 64-byte data logging
• Wi-Fi streaming (with ESP32 co-processor or Feather variant)
• opendbc fingerprinting (auto-identify car model from CAN traffic)

---

Credits & References

• ACANFD_FeatherM4CAN — Pierre Molinaro — CAN driver
• opendbc — comma.ai — DBC database & car CAN decoding
• Adafruit Feather M4 CAN — Hardware documentation
• Betaflight Configurator — UI inspiration

---

License

MIT License — see LICENSE