---
title: MoleNet IoT Sensor Board
url: https://www.emergentmind.com/topics/molenet-iot-sensor-board
type: topic
---

# MoleNet IoT Sensor Board

MoleNet IoT Sensor Board is a low-power, small-form-factor hardware platform and associated methodology supporting high-precision environmental monitoring and low-current profiling for battery-operated Internet of Things (IoT) deployments. Its design integrates multi-modal sensor support, robust power management, and advanced techniques for characterizing supply current in ultra-low-power states. The platform is engineered to enable long-term field deployments for atmospheric and environmental data logging, with application-specific modules including CO$_2$ sensing, soil moisture measurement, and wireless data transmission [2601.16727, 2308.03780].

## 1. Hardware Architecture

MoleNet’s core architecture comprises a regulated 3.3 V power rail sourced from a Li-ion battery or USB supply through a TI TLV731 LDO (up to 1 A, ±1%) [2601.16727]. The board supports differentiated subsystem powering for:

- ESP32-S3 MCU (dual-core Xtensa®, Wi-Fi 2.4 GHz, Bluetooth LE).
- SX1276-class LoRa transceiver (SPI; 125 kHz BW; ±17 dBm).
- Bosch Sensortec BME280 sensor (integrated pressure, humidity, temperature; ±1 °C, ±3 %RH, ±1 hPa).
- microSD card socket (SPI, up to 4 MHz typical).
- Expandable external sensor headers (UART, SDI-12 environmental, I²C bus).

User I/O includes a tactile pushbutton for test-sequence initiation, status LEDs, and a dedicated GPIO pin for hardware triggering with measurement equipment.

Input voltage tolerance is 4.5–5.5 V, supporting battery or USB operation. The ESP32-S3 operates in multiple power modes: Active (all systems), Idle (CPU on, radio off), Light Sleep (RTC on, peripherals off), and Deep Sleep (RTC only, ULP optional), with measured sleep-mode currents reduced to single-digit microamperes.

For CO$_2$-specific deployments, the MoleNet CO$_2$ Sensor Board substitutes the MCU subsystem with NodeMCU-ESP8266, powers the MQ135 analog gas sensor, DHT-11 temperature/humidity sensor, and optional LCD-I2C status display; all components operate on a 3.3 V rail [2308.03780].

## 2. Sensor Suite and Data Acquisition

The MoleNet board natively supports:

- Local atmospheric monitoring via BME280 for weather/pressure trends.
- SDI-12 5TM soil moisture module input for volumetric water content.
- UART connectivity for generic 3.3 V sensors (particulate, CO$_2$).
- I²C interfacing for expansion (light, gas sensors).

The microSD card implements SPI-based data logging, supporting up to 4 MHz communication rates.

MoleNet CO$_2$ variant leverages:

- MQ-135 CO$_2$ sensor, with a detection range of ~10–10 000 ppm, typical accuracy ±(50 ppm + 5%) in CO$_2$ mode, and ~15 s step response. Sensor output is calibrated and digitized via onboard ADC (0–1 V, scaled as needed).
- DHT-11 for temperature (±2 °C; 1 °C resolution) and humidity (±5 %RH; 1 % resolution).

Sensor data routing follows standard digital interfaces: UART (115,200 baud), SDI-12 with level-shifting and 4.7 kΩ pull-up, SPI/MOSI-MISO-SCLK-CS to microSD, and I²C SCL/SDA (up to 400 kHz, with 10 kHz typical for environmental modules).

## 3. Low-Current Measurement Methodology

Precise current characterization uses dedicated SMUs (e.g., tinyCurrent, JouleScope JS220):

- SMU is inserted in series between regulator output and board input, with common ground referencing.
- Test-point GPIO triggers SMU data capture for automated measurement sequences.
- Board and SMU are powered down before series integration; current ranges and sampling rates are set per test phase (JouleScope at 1 MSa/s; tinyCurrent at oscilloscope rate).
- ESP32-S3 firmware cycles peripheral activation and sleep periods, with GPIO assertion aligning SMU measurements.

SMU front-end shunt design is critical. tinyCurrent offers nanoampere (±1.25 µA full-scale, R$_{shunt}$ ≈10 MΩ, burden ≈10 µV/nA) and microampere ranges (±1.25 mA, R$_{shunt}$ ≈10 Ω, burden ≈10 µV/µA). JouleScope’s integrated shunts achieve best resolutions ≈15 nA across ADC-acquired 1 MSa/s USB output.

Measurement equations:
- $I = \Delta V_{shunt} / R_{shunt}$
- $\Delta V_{shunt} = V_{measured} - V_{offset}$
- Combined uncertainty: $\delta I = \sqrt{(\delta V_{amp})^2 / R^2 + (I \cdot \delta R / R)^2}$, with $\delta V_{amp}$ amplifier error and $\delta R$ shunt tolerance.

Calibration procedures include zero-offset correction, burden voltage checks ($V_{burden} \ll 10$ mV), and EMI mitigation via shielded microvolt outputs.

## 4. Firmware and Network Management

For CO$_2$ monitoring, firmware implements a periodic measurement, display, and network transmission cycle:

- Sensor readout and calibration: MQ135 baseline (R$_{0}$) acquired by exposing to fresh air, then Rs determined per cycle, and CO$_2$ ppm computed via:
  - $ppm_{CO_2} = A (V_{out} - V_{0})^{B}$, with $A = 116.6$, $B = -2.769$, and $V_{0} = 1.98$ V [2308.03780].
- DHT-11 data is acquired with error checking; LCD updates expound environmental values and trigger programmable alerts (LED/buzzer) for CO$_2$ levels exceeding 1000 ppm.
- Data is published as a JSON payload via MQTT (TLS 1.2, broker mqtt.molenet.org) or HTTP, with transmission intervals configurable (default 60 s), exponential back-off for retry, and local storage via circular SPIFFS buffer in case of loss of connectivity.
- Security options include TLS verification and hardware secure elements (ATECC608A).

ESP8266 MCU deep-sleep is invoked (ESP8266.deepSleep(60e6)), with average current draw calculated per 60 s measurement cycle ($I_{avg} \approx 3.0$ mA).

## 5. Experimental Performance and Analysis

Measured supply currents with SMU profiling:
- LoRa TX (active, +17 dBm): peak ≈100–125 mA, mean ≈90 mA.
- Full sensor readout (UART + SDI-12 + I²C): 15–25 mA.
- microSD SPI operations: 25–35 mA at 100 kHz.
- MCU idle: 50–60 mA.
- Light Sleep: 0.7–1.0 mA.
- Deep Sleep (RTC + ULP): 2–8 µA, confirming effective ultra-low-power regime [2601.16727].

Measurement setup achieves sub-µA accuracy with JouleScope (≈15 nA noise floor) and tinyCurrent (≈200 nA rms noise at 10 MSa/s, settling time ≈1 µs).

Accuracy for the CO$_2$ sensing path is validated:
- Baseline: 400 ppm at 1.98 V, Rs/R$_{0} \approx 1.0$.
- Span: 1000 ppm at 1.05 V, error $<$±5%.
- Repeatability: ±2 ppm over 1 h.

Response latency for cycle completion is ~200 ms, with system reliability at 99% uptime over 48 h indoors. Sensor resolution: 0.1 °C, 1 %RH, 1 ppm CO$_2$ [2308.03780].

## 6. Design Optimizations and Trade-offs

Quiescent current minimization is paramount:
- LDOs selected for sub-µA standby draw.
- Firmware disables unused buses and peripherals; batch sensor reads maximize deep-sleep residency.
- High-value pull-ups (>100 kΩ) on seldom-used communications buses further reduce leakage.

Measurement precision versus cost and complexity reveals:
- tinyCurrent plus oscilloscope (≈€20): optimal for manual bench profiling, high time resolution, lacking auto-range/trigger.
- JouleScope JS220 (≈€500): builtin ADC, auto-range, hardware trigger, sub-10 nA resolution, preferred for automated, high-speed diagnostics.
- High-end bench SMUs (≫€2,000) afford advanced capabilities but excess for routine current profiling of MoleNet boards.

Shunt resistor design balances resolution and burden voltage, impacting regulator stability and low-current fidelity.

## 7. Integration and Extension within IoT Platforms

MoleNet board integration leverages standardized MQTT topics, device tagging with location/floor metadata, and time-series data aggregation into MoleDB for large-scale analytics. The MoleAnalytics stack enables anomaly detection, e.g., automated flagging of CO$_2$ excursions above 1000 ppm [2308.03780].

Potential enhancements include:
- I²C gas-sensor arrays for additional molecular species (CO, NO$_{2}$, VOC).
- ESP-NOW mesh networking for coverage without Wi-Fi.
- Embedded Edge-AI for calibration drift correction and cross-sensitivity compensation.

Deployment recommendations endorse Joulescope-class analyzers for development with auto-range and hardware triggers. In production, sub-µA sleep-mode budgets are attainable through careful hardware selection and firmware strategies [2601.16727].

**References:**  
- "Precise Low-Current Measurement Techniques for IoT Devices: A Case Study on MoleNet" [2601.16727]  
- "Exploring IoT for real-time CO2 monitoring and analysis" [2308.03780]

Source: https://www.emergentmind.com/topics/molenet-iot-sensor-board