Wireless temperature sensors are a major step forward for monitoring critical environments. Among the available technologies, NB-IoT (Narrowband Internet of Things) stands out as particularly suited to large-scale deployments: wide coverage, low energy consumption and cellular communication with no local infrastructure. However, the advertised battery life of these devices, up to 10 years for Efento sensors, is only achievable if you understand what drives their energy consumption and configure the parameters properly. This article explains the basic principles and the configuration strategies that maximise the life of your sensors while guaranteeing the quality of the monitoring.
Measuring and Transmitting: Two Separate Clocks
A common mistake is to confuse the measurement frequency with the transmission frequency. The two parameters are independent, and keeping them apart is fundamental to optimising battery life.
The measurement interval
Defines how often the sensor takes a temperature reading. This operation, done locally, uses relatively little energy. The readings are time-stamped and stored in the sensor's internal memory, which can hold up to 40,000 measurements.
The transmission interval
Determines how often the sensor connects to the NB-IoT cellular network to send the accumulated data to the monitoring platform. This is the phase that draws most on the battery.
Every transmission cycle is a full sequence: the radio module has to wake up, synchronise with the network, negotiate the connection, send the data and wait for the acknowledgement.
In practice: a sensor can be set to measure every 5 minutes while only transmitting every 2 hours. That keeps fine-grained data while drastically reducing the number of transmission cycles, and therefore the energy consumption.
What Determines Energy Consumption
Radio signal quality: the dominant factor
NB-IoT signal quality is the parameter with the biggest impact on battery life. The NB-IoT protocol defines three coverage levels (ECL, Extended Coverage Level) that reflect the reception conditions:
| Level | Conditions | Theoretical battery life* | Realistic battery life** |
|---|---|---|---|
| ECL 0 | Optimal signal | ~11 years | 5-7 years |
| ECL 1 | Moderate signal | ~4.7 years | 2-4 years |
| ECL 2 | Weak signal | ~2.35 years | 1-2 years |
*Ideal laboratory conditions, Bluetooth off, room temperature.
**Estimate taking real operating conditions into account (Bluetooth on, temperature variations, occasional retransmissions).
Caution: these theoretical estimates, supplied by the manufacturer, assume ideal conditions that are rarely met in real operation. In practice the Bluetooth interface usually stays active to allow maintenance work, the signal fluctuates with building activity and weather, and refrigerated environments penalise battery performance. It is therefore prudent to count on an effective battery life of 3 to 5 years for a typical deployment in favourable conditions, and 1 to 3 years in demanding environments (freezer rooms, areas with limited coverage).
This large variation comes from two mechanisms. In poor signal conditions the sensor has to raise its transmission power, up to 23 dBm (200 mW) against 0 dBm (1 mW) in optimal conditions. In addition, packet losses lead to energy-hungry retransmissions, and the time spent in active mode grows significantly.
At installation, it is essential to check the signal quality at the planned location. The Efento mobile app shows a signal indicator: always favour locations where it reads "Good" or "Moderate". If several NB-IoT operators are available in your area, switching operator can sometimes improve reception considerably.
Transmission frequency
Every transmission cycle is a full sequence: wake the radio module, attach to the network, send the data, receive the acknowledgement, go back to sleep. This sequence uses a fairly fixed amount of energy, regardless of how much data is sent.
Reducing the transmission frequency from 15 minutes to 1 hour can therefore extend battery life by several years. For applications where responsiveness is not critical, storage of stable products, archives, some industrial processes, a transmission every 4 to 6 hours may be enough.
Operating temperature
The thermal environment in which the sensor operates directly affects the performance of its lithium thionyl chloride (Li-SOCl₂) battery. Below 0°C, the effective capacity of the battery can drop by 50%. Conversely, prolonged temperatures above 40°C accelerate the chemical degradation of the cells and shorten their total life.
For freezer rooms, a shorter battery life than the nominal specifications should therefore be expected. A sensor installed in a room at -20°C will have a significantly shorter life than the same sensor operating at room temperature.
The Bluetooth Low Energy interface
Efento NB-IoT sensors have a Bluetooth Low Energy (BLE) interface used for initial configuration and local reading of the data through the mobile app. Even in passive mode, this interface uses some energy.
Efento lets you set a delay after which the Bluetooth interface switches off automatically. This option can be set during initial deployment through the mobile app. For installations where physical access to the sensors is rare, reducing the Bluetooth activity window helps preserve battery life.
Threshold-Triggered Transmissions: a Smart Strategy
Beyond periodic transmissions, Efento sensors include an on-board analysis feature (edge analytics) that triggers an immediate transmission when certain conditions are met.
Absolute threshold
Immediate transmission if the temperature exceeds a set value (e.g. > -15°C for a freezer room).
Differential threshold
Transmission if the gap between the current reading and the last transmitted value exceeds a set delta.
Moving average
Comparison based on the average of the last n readings, which filters out transient variations.
This approach turns the sensor into an autonomous sentinel: in normal conditions, transmissions stay spaced out to preserve the battery. But as soon as an anomaly is detected, the alert is sent to the monitoring platform immediately.
This strategy is particularly relevant in regulated environments (pharmaceutical, food) where responsiveness to a temperature drift is critical, while keeping maximum battery life in normal operation.
Simulate Before Deciding: the Modelling Approach
Efento provides an online calculator to estimate battery life according to the configuration parameters and the signal conditions. It lets you simulate different scenarios before deployment.
Comparative example
Theoretical battery life: > 10 years — Estimated realistic battery life: 5 to 7 years
Theoretical battery life: ~4 years — Estimated realistic battery life: 1.5 to 2.5 years
The difference shows how important it is to adapt the configuration to the context of use. Proper sizing avoids both premature battery replacements and unplanned monitoring interruptions.
Conclusion
Configuring an NB-IoT sensor is not just a technical setting: it is a trade-off between data granularity, responsiveness to anomalies and battery life. Understanding the underlying mechanisms, measurement versus transmission, the impact of the radio signal, the thermal behaviour of batteries, makes it possible to make informed choices for each context.
For critical deployments, a prior analysis of the signal conditions, a simulation of the expected battery life and the configuration of alert thresholds are the essential steps of a successful project. Battery replacement should also be planned into the maintenance cycle, counting on a life shorter than the maximum advertised specifications.