Arduino Nano 33 BLE Sense Rev2
Arduino Nano 33 BLE Sense Rev2 — 9-Axis IMU — TinyML Edge AI — Bluetooth 5 BLE The Arduino Nano 33 BLE Sense Rev2 packs a Nordic nRF52840 Cortex-M4F processor,...
Read MoreSKU:ARD-007
- Regular price
- Rs. 3,863.90
- Sale price
- Rs. 3,863.90
- Regular price
-
Express Shipping
Shipping Info
Fast delivery options for your convenience
Pick-Up in Store
Free
Select this option to pick up your order directly from our store. Fast and convenient!
Arduino Nano 33 BLE Sense Rev2 — 9-Axis IMU — TinyML Edge AI — Bluetooth 5 BLE
The Arduino Nano 33 BLE Sense Rev2 packs a Nordic nRF52840 Cortex-M4F processor, ten onboard sensors, and native Bluetooth 5 BLE into the iconic Nano form factor — smaller than a stick of gum. Built for TinyML edge inference, it ships sensor-ready for gesture recognition, wake-word detection, and environmental monitoring with zero external hardware required.
Key Highlights
- nRF52840 Cortex-M4F @ 64 MHz — ARM's most capable Cortex-M4 core with a hardware FPU and DSP extensions, delivering enough compute for real-time neural-network inference at the edge without cloud dependency.
- 9-Axis IMU: BMI270 + BMM150 — A 6-axis accelerometer/gyroscope paired with a 3-axis magnetometer for precise motion tracking, gesture classification, and full-attitude orientation sensing — all onboard, no extra wiring.
- MEMS Microphone (MP34DT06JTR) — Integrated digital microphone captures live audio for keyword spotting, sound classification, and voice-triggered applications entirely on-device.
- Gesture, Proximity, Colour & Light (APDS9960) — Four sensing modalities in one IC: detect hand gestures, measure proximity up to ~10 cm, read full RGB + clear intensity, and respond to ambient light changes.
- Barometric Pressure + Temperature (LPS22HB) — High-accuracy absolute pressure sensor (260–1260 hPa) with a built-in temperature probe, ideal for altitude estimation and indoor weather monitoring.
- Humidity & Temperature (HS3003) — ±2% relative humidity accuracy and ±0.2 °C temperature precision for reliable environmental sensing in wearables, enclosures, and greenhouse controllers.
- Bluetooth 5 BLE + NFC Pairing — Stream sensor data or inference results to smartphones and BLE peripherals; NFC tag enables effortless device pairing without a button or code entry.
- 1 MB Flash / 256 KB SRAM — Substantially more memory than classic Nano boards, providing comfortable headroom to run TensorFlow Lite Micro models alongside full application logic.
- Native USB Controller — The nRF52840's built-in USB eliminates a separate USB-serial chip, enabling HID, CDC, and WebUSB use cases directly and reducing board complexity.
- Wide Input Voltage: 4.5V–21V via VIN — Accepts single-cell LiPo packs, USB power banks, and benchtop supplies through the VIN pin, while 3.3V logic integrates cleanly with modern sensor modules and shields.
Technical Specifications
| Specification | Details |
| Microcontroller | Nordic nRF52840 via u-blox NINA-B306, Arm Cortex-M4F |
| Clock Speed | 64 MHz |
| Operating Voltage | 3.3V (I/O not 5V tolerant) |
| Input Voltage (VIN) | 4.5V – 21V |
| Flash Memory | 1 MB (nRF52840) |
| SRAM | 256 KB (nRF52840) |
| EEPROM | None |
| Digital I/O Pins | 14 |
| PWM Pins | All digital pins |
| Analog Input Pins | 8 (12-bit ADC, 200 k samples/s) |
| Analog Output | PWM only (no DAC) |
| External Interrupts | All digital pins |
| DC Current per I/O Pin | 15 mA |
| UART / SPI / I2C | 1 / 1 / 1 |
| USB | Native (nRF52840 built-in) |
| Wireless | Bluetooth 5 BLE (u-blox NINA-B306) + NFC |
| IMU | BMI270 (6-axis accel + gyro) + BMM150 (3-axis magnetometer) |
| Microphone | MP34DT06JTR (MEMS digital) |
| Gesture / Light / Proximity / Colour | APDS9960 |
| Barometric Pressure | LPS22HB (260–1260 hPa) |
| Temperature & Humidity | HS3003 (±0.2 °C / ±2% RH) |
| LED_BUILTIN | Pin 13 |
| Dimensions | 18 mm × 45 mm (Nano form factor) |
Common Applications & Use Cases
- Gesture-Controlled Interfaces — Use the APDS9960 and BMI270 together to build touchless controls for displays, smart-home devices, and industrial HMIs; deploy a TinyML gesture classifier and get zero-latency responses with no internet connection.
- Wake-Word Detection — Run TensorFlow Lite Micro speech models on the nRF52840 to detect trigger words locally — no cloud, no latency, and full audio privacy since raw audio never leaves the device.
- Wearable Health & Fitness Trackers — Combine IMU data with BLE streaming to build step counters, activity classifiers, fall-detection alerts, or posture monitors compact enough to embed in a wristband or clip-on device.
- Indoor Air Quality & Weather Stations — Log and transmit temperature, humidity, and barometric pressure over BLE for real-time environmental dashboards, grow-tent automation, or local e-paper display projects.
- Predictive Maintenance & Anomaly Detection — Train a vibration anomaly model on IMU data from healthy machinery, then deploy it on the board to flag deviations locally — reducing downtime without cloud infrastructure.
- Sound Classification — Capture audio with the MEMS microphone and run edge inference to classify sounds — glass breaks, coughs, machine faults, or wildlife calls — without ever uploading raw audio to a server.
- Orientation & Heading Sensing — The 9-axis IMU enables full sensor fusion for precise attitude estimation in drone controllers, handheld instruments, gimbal stabilisers, and orientation-aware robotics.
- Smart Lighting & Proximity Triggers — Use the APDS9960 proximity and colour channels to build adaptive lighting that brightens on approach, adjusts white balance to ambient colour temperature, or triggers actions on presence detection.
- TinyML Education & Prototyping — Official Edge Impulse integration provides a beginner-friendly pipeline from live data capture to deployed model in under an hour, making this the standard board for machine-learning-at-the-edge courses and workshops.
- BLE Peripheral & Sensor Node Development — Program the nRF52840 as a BLE HID peripheral (keyboard, mouse, gamepad) or as a low-power BLE sensor beacon broadcasting to a central hub, gateway, or smartphone app.
What's in the Box
- 1 × Arduino Nano 33 BLE Sense Rev2 with pre-soldered pin headers
Note: accessories such as USB cables, power supplies, breadboards, cases, and SD cards are sold separately and not included unless stated above.
Frequently Asked Questions
What operating systems and IDEs does the Arduino Nano 33 BLE Sense Rev2 work with?
The board is compatible with Windows 10 / 10 IoT, macOS, and Linux through the Arduino IDE 2.x and Arduino CLI. The Arduino Web Editor supports browser-based development with no local installation required. For machine-learning workflows, Edge Impulse Studio provides first-class support with a dedicated data-acquisition pipeline and one-click model deployment. The board also runs CircuitPython for rapid scripting without a compile step, and supports Android 7+ via USB OTG for mobile integration.
What voltage does the Nano 33 BLE Sense Rev2 need, and how do I power it?
The board is powered via USB (5V) or through the VIN pin, which accepts 4.5V to 21V DC. The onboard regulator steps the supply down to the 3.3V rail used by the processor and all sensors. All I/O pins operate at 3.3V only — the board is strictly not 5V tolerant, so connecting 5V signals directly to any GPIO risks permanent damage to the nRF52840. For portable builds, a single-cell LiPo (3.7V–4.2V nominal) connected via VIN is a common and reliable choice. A USB power bank or regulated wall adapter through VIN works equally well for bench or deployed use.
Does the Nano 33 BLE Sense Rev2 support CircuitPython or MicroPython?
CircuitPython is officially supported and provides a beginner-friendly environment where code runs directly from a virtual USB drive — no compilation step required. The Arduino Mbed OS core is the primary firmware target and enables access to all onboard sensors via the official Arduino sensor libraries. TensorFlow Lite for Microcontrollers and Edge Impulse deployment libraries are available through the Arduino Library Manager for deploying quantised neural-network models. MicroPython support exists via community ports but receives less maintenance than the CircuitPython or Mbed core paths. For production TinyML projects, the Mbed + TFLite Micro combination is the most thoroughly tested stack.
How much memory is available for TinyML models?
The nRF52840 provides 1 MB of flash for program and model storage and 256 KB of SRAM for runtime data and inference buffers. In practice, TensorFlow Lite Micro models for gesture recognition and keyword spotting typically consume 50–200 KB of flash, leaving substantial headroom for application code. Int8 quantisation is strongly recommended to keep models within SRAM constraints — float32 models of equivalent accuracy will often not fit. Edge Impulse's deployment wizard automatically generates quantised C++ libraries optimised for the nRF52840's memory budget, making model sizing straightforward even for beginners.
What accessories do I need to get started?
The Nano 33 BLE Sense Rev2 ships with pin headers already soldered, so it drops straight into a standard breadboard without any soldering. The only hardware you need is a USB-A to Micro-USB cable for programming and power. On the software side, install the free Arduino IDE 2.x and then add the "Arduino Mbed OS Nano Boards" package via the Boards Manager — this brings all required drivers and sensor libraries. No external sensors, shields, or power supplies are needed to start experimenting with the onboard IMU, microphone, and environmental sensors immediately out of the box.
How does the Rev2 differ from the original Nano 33 BLE Sense?
Rev2 replaces three sensors from the original: the LSM9DS1 IMU is replaced by the BMI270 + BMM150 combination (offering improved low-power modes and better noise performance), the HTS221 humidity/temperature sensor is replaced by the more accurate HS3003, and the MP34DT05 microphone is upgraded to the MP34DT06JTR. The nRF52840 processor, Bluetooth 5 BLE, NFC, Nano form factor, and pin layout are identical between Rev1 and Rev2, so existing projects can be migrated with library-driver swaps for the affected sensors only. Rev2 is the current production revision and receives ongoing Arduino board-support-package and library updates.
How many GPIO, SPI, I2C, and UART interfaces are available?
The headers expose 14 digital I/O pins, all of which support external interrupts and PWM output — every pin is interrupt-capable with no exclusions. There is 1 hardware UART, 1 SPI bus, and 1 I2C bus on the headers, plus 8 analog input channels sampled at 12-bit resolution and 200 k samples/s. The onboard sensors share an internal I2C bus, so the external I2C header pins remain fully available for additional peripherals without conflicts. The nRF52840's native USB also exposes CDC serial, HID, and WebUSB endpoints independently of the UART pins.
Is this board suitable for beginners, or is it aimed at advanced users?
The Nano 33 BLE Sense Rev2 suits both levels well. Beginners benefit from the drop-in breadboard form factor, official Arduino IDE support, comprehensive sensor libraries, and step-by-step Edge Impulse tutorials that guide from raw data collection to a deployed model in under an hour. Advanced users gain access to the full nRF52840 feature set — FreeRTOS via Mbed OS, BLE GATT server/client APIs, USB HID, low-power sleep modes, and direct register access. The primary beginner caveat is the 3.3V-only I/O requirement, which demands a logic-level shifter when interfacing with older 5V modules or Arduino Uno shields — something to factor in when planning peripherals.
What is the most common mistake users make with the Nano 33 BLE Sense Rev2?
The single most frequent issue is applying 5V signals directly to the GPIO pins. Unlike the classic Arduino Uno, this board has no 5V tolerance on any I/O — connecting a 5V sensor, shield, or logic signal without a level shifter can permanently damage the nRF52840. A 3.3V ↔ 5V logic-level shifter is required for any 5V peripheral. A second common mistake is attempting to upload sketches before installing the "Arduino Mbed OS Nano Boards" core package in the Boards Manager — the IDE will fail to find the board target until this package is installed. Double-checking both before wiring anything saves most first-session headaches.
Where can I find documentation, community support, and firmware updates?
The official documentation hub at docs.arduino.cc/hardware/nano-33-ble-sense-rev2 covers pin-out diagrams, full technical specifications, and guided tutorials including a dedicated TinyML getting-started guide. The Arduino Forum (forum.arduino.cc) has a dedicated Nano 33 BLE Sense section with extensive community Q&A threads. For TinyML projects, Edge Impulse Studio provides free project workspaces, pre-built example projects for this exact board, and active community forums. Board-support-package updates and sensor library releases arrive through the Arduino Library Manager and Boards Manager — checking for updates every few months ensures access to the latest bug fixes and sensor-calibration improvements from Arduino.
