7Semi MLX90640 IR Thermal Camera Breakout - 110 Degree Wide FOV
Add affordable thermal imaging to your projects with the 7Semi MLX90640 IR Thermal Camera Breakout. Featuring a high-resolution 32x24 pixel infrared sensor array with a wide 110 degree field of view, this compact thermal camera provides real-time heat detection and thermal imaging for industrial, research, and embedded applications. Built around the Melexis MLX90640 far-infrared sensor, it delivers professional-grade performance with easy I2C integration.
768 Pixel Thermal Imaging Array
With 32 x 24 pixels (768 individual IR sensors), this thermal camera creates detailed heat maps for presence detection, equipment monitoring, thermal analysis, and more - all in a compact, easy-to-integrate breakout board.
Key Features
High-Resolution Thermal Imaging
- 32 x 24 Pixel Array - 768 individual IR temperature sensors
- Detailed Heat Maps - Visualize temperature distribution
- Low NETD - 0.1K RMS detects tiny temperature differences
- Factory Calibrated - Ready to use, no calibration needed
Wide 110 Degree Field of View
- 110 x 75 Degree FOV - Covers expansive areas
- Room Monitoring - Single sensor covers entire spaces
- Up to 7m Range - Human detection at distance
- Wide Angle Coverage - Minimizes repositioning
Accurate Temperature Sensing
- -40C to 300C Range - Industrial temperature coverage
- Plus/Minus 2C Accuracy - Reliable measurements
- Non-Contact - Safe remote temperature sensing
- Real-Time - Instant temperature readings
Easy Integration
- I2C Interface - Up to 1MHz Fast Mode Plus
- Qwiic Connector - Solder-free plug-and-play
- 3.3V/5V Compatible - Works with most microcontrollers
- Programmable Refresh - 0.5Hz to 64Hz adjustable
Technical Specifications
| Sensor |
| Sensor Type |
Melexis MLX90640 Far-Infrared Array |
| Resolution |
32 x 24 pixels (768 IR sensors) |
| Field of View |
110 x 75 degrees (wide angle) |
| Technology |
Thermopile array |
| Temperature |
| Target Range |
-40C to 300C |
| Accuracy |
Plus/Minus 2C (in 0-100C range) |
| NETD |
0.1K RMS @ 1Hz refresh rate |
| Operating Temperature |
-40C to 85C |
| Performance |
| Refresh Rate |
0.5Hz to 64Hz (programmable) |
| Recommended Rate |
8-16Hz for optimal imaging |
| I2C Speed |
Up to 1MHz (Fast Mode Plus) |
| I2C Address |
0x33 (default, programmable) |
| Electrical |
| Supply Voltage |
3.3V to 5V |
| Current Consumption |
Less than 23mA |
| Interface |
I2C with Qwiic connector |
| Features |
| Calibration |
Factory calibrated (no user calibration) |
| Data Storage |
EEPROM for calibration constants |
| Operating Mode |
Chess pattern (default, recommended) |
Platform Compatibility
Important: RAM Requirement
The MLX90640 requires at least 20KB RAM on the host microcontroller.
Arduino Uno/Nano WILL NOT WORK (only 2KB RAM). Use platforms like Raspberry Pi, ESP32, Teensy 3.1+, or Arduino Due.
| Platform |
Compatible |
Notes |
| Raspberry Pi 5/4/3 |
YES |
Excellent - Python and C libraries available |
| ESP32 |
YES |
Good - 520KB SRAM sufficient |
| Teensy 3.1+ |
YES |
Excellent - Recommended for Arduino ecosystem |
| Arduino Due |
YES |
Good - 96KB SRAM available |
| STM32F4 |
YES |
Good performance |
| Arduino Uno/Nano |
NO |
Insufficient RAM (only 2KB) |
Perfect For Many Applications
Presence Detection
Detect people without cameras, privacy-preserving occupancy sensing
Industrial Monitoring
Equipment temperature, predictive maintenance, quality control
Smart Building
HVAC optimization, thermal leak detection, energy efficiency
Security Systems
Intrusion detection in darkness, fire early warning
Research
Thermal analysis, scientific experiments, product testing
Electronics
PCB hot spot detection, CPU/GPU monitoring, heat dissipation
110 Degree vs 55 Degree FOV
| Feature |
110 Degree FOV (This Product) |
55 Degree FOV |
| Coverage |
Wide area - entire rooms |
Narrow - focused targets |
| Best For |
Presence detection, HVAC, security |
Targeting, inspection, distant objects |
| Human Detection |
Up to ~7 meters |
Up to ~11 meters |
| Detail Level |
Lower per-pixel (wider spread) |
Higher per-pixel (concentrated) |
How It Works
The MLX90640 uses thermopile technology to detect infrared radiation emitted by objects. Each of the 768 IR sensors measures the temperature of the area it "sees," creating a grid of temperature values that form a thermal image.
- Works in Complete Darkness - Detects heat, not visible light
- Non-Contact Measurement - No need to touch the object
- Real-Time Imaging - Continuous temperature monitoring
- No Recalibration - Factory calibration maintains accuracy over time
Connection
Pinout:
| VCC |
Power (3.3V to 5V) |
| GND |
Ground |
| SDA |
I2C Data |
| SCL |
I2C Clock |
Qwiic Connection:
Simply plug a Qwiic cable between the breakout and any Qwiic-enabled board. No soldering required - the polarized connector prevents incorrect connection.
What's Included
- 1x 7Semi MLX90640 Thermal Camera Breakout - 110 degree FOV version
- Documentation - User guide and datasheets (downloadable)
What You Need (Sold Separately)
Required Items - NOT Included
- Qwiic/I2C Cable - 4-pin JST SH connector cable
- Microcontroller - Raspberry Pi, ESP32, Teensy, or Arduino Due (20KB+ RAM required)
- Display (optional) - TFT/OLED for thermal image visualization
- Power Supply - For your microcontroller
Important Notes
Do Not Hot-Plug
The MLX90640 has EEPROM that can be damaged by hot-plugging. Do not connect or disconnect cables while the device is powered. Avoid excessive vibration during operation.
Dead Pixels
Per Melexis specification, up to 4 dead pixels are allowed and within normal tolerance. Dead pixels are marked in EEPROM and can be interpolated from neighboring pixels in software. This is not a defect.
IR and Glass
Standard glass blocks infrared radiation. Do not place glass or regular plastic in front of the sensor. Use IR-transparent materials if an enclosure window is needed, or leave the sensor exposed.
Software Support
Libraries Available:
- Raspberry Pi Python - Adafruit CircuitPython MLX90640 library
- Arduino - Adafruit MLX90640 Library, SparkFun MLX90640 Library
- ESP32 - Compatible with Arduino libraries
- C/C++ - Melexis and community drivers
Resources:
- Melexis MLX90640 Datasheet
- 7Semi User Guide
- Raspberry Pi Getting Started Manual
- 3D STEP File for mechanical design
- DXF File for enclosure design
Frequently Asked Questions
Q: Can I use this with Arduino Uno?
A: No. Arduino Uno only has 2KB RAM, but the MLX90640 requires at least 20KB RAM. Use Raspberry Pi, ESP32, Teensy 3.1+, or Arduino Due.
Q: Does this work in complete darkness?
A: Yes! Thermal cameras detect infrared radiation (heat), not visible light. They work perfectly in total darkness.
Q: What's the detection range?
A: With the 110 degree FOV, human detection is possible up to approximately 7 meters. Body contour recognition works up to 2-5 meters.
Q: Is calibration required?
A: No. The sensor is factory calibrated and ready to use out of the box.
Q: Why are there dead pixels?
A: Up to 4 dead pixels are within Melexis specification and normal. Software can interpolate values from neighboring pixels.
Q: What refresh rate should I use?
A: 8-16Hz is recommended for smooth thermal imaging. Higher rates (32-64Hz) increase noise.
Why Buy from CrazyPi.com?
- Genuine Products - Authentic 7Semi and Melexis components
- Expert Support - Technical assistance for your projects
- Fast Shipping - Quick delivery across India
- Complete Ecosystem - Controllers, cables, displays available
- Bundle Deals - Save on complete thermal imaging kits
Thermal Imaging Kits Available!
Get everything you need to start thermal imaging. We offer Starter Kits with Raspberry Pi, ESP32 Thermal Camera Kits with displays, and Industrial Monitoring Kits - all at discounted bundle prices!
What You Can Build
- DIY Thermal Camera - Portable heat vision device
- People Counter - Privacy-preserving occupancy sensor
- Smart Thermostat - Temperature-aware HVAC control
- Security System - Heat-based motion detection
- Equipment Monitor - Predictive maintenance alerts
- Fire Detector - Early thermal anomaly warning
- PCB Analyzer - Find hot spots on circuit boards
- Science Projects - Visualize heat transfer
See Heat Like Never Before - 32x24 Pixel Thermal Imaging
Buy 7Semi MLX90640 Thermal Camera at CrazyPi.com - India's source for thermal imaging and sensing solutions.
