Prioritising the power budget for military sensors
Power is one of the most unforgiving constraints in tactical sensor design. A sensor that runs out of energy in the field is, from an operational standpoint, worse than no sensor at all, as it creates a gap in situational awareness precisely when it is needed most. Yet across many aerospace, remote intelligence, and defence electronics projects, power budgeting is often treated as an afterthought, i.e., something to revisit once the sensor architecture is locked and the PCB is already in layout. That sequencing is expensive, and in defence programmes operating under tight procurement timelines and MIL-SPEC validation requirements, late-stage power redesigns carry costs that go well beyond the engineering hours involved. A device’s tactical sensor power design is what ensures it remains operational in harsh, battery-constrained, or contested environments without compromising weight, thermal limits, or mission duration.
This is the fourth post in Ignitec’s series on embedded electronics for the UK defence sector. Previous posts have addressed LoRa network design for low-probability-of-intercept sensor connectivity, MAVLink protocol integration for defence UAV platforms, and secure over-the-air firmware update architecture for field-deployed defence IoT systems.
All three topics intersect directly with power budgeting: a LoRa node running continuous receive cycles will drain a LiSOCl2 primary cell in weeks rather than months, a MAVLink-connected payload drawing unregulated current from a UAV bus can destabilise the entire aircraft power rail, and a secure OTA update cycle over a constrained radio link carries an energy cost that must be explicitly accounted for in the mission power budget. Tactical sensor power design cannot be treated in isolation from the broader system architecture.
Why is tactical sensor power design different from consumer IoT?
Broadly speaking, tactical sensor power design differs fundamentally from consumer IoT, prioritising survivability and secure operation over convenience and cost-efficiency. While tactical sensors focus on guaranteed performance and cryptographic isolation in high-threat or austere environments, consumer devices are designed for mass markets where affordability and ease of use are the driving forces, thus making them impractical – and unviable – to use interchangeably.
Most of the power optimisation literature for embedded systems targets consumer IoT applications such as smart home devices, wearable health monitors, and asset trackers. The engineering principles are broadly applicable to military-grade sensors, but the operational context could not be more different.
In a consumer IoT deployment, a flat-running sensor can be retrieved and recharged. The consequences are inconvenience and, potentially, a service-level agreement breach. In a tactical deployment, the consequences of power failure range from mission compromise to physical danger for the personnel relying on that sensor data.
The differences that matter most from an engineering standpoint are:
- Recharging infrastructure is unavailable or severely constrained in forward-deployed positions.
- Mission duration is unpredictable, i.e., sensors must sustain full performance across worst-case scenarios, not average-case ones.
- Environmental conditions (temperature extremes, altitude, humidity) degrade battery capacity in ways that consumer-grade models do not capture.
- Emissions constraints (RF, acoustic, thermal) limit which power management strategies are viable in contested environments.
- The sensor must maintain operational integrity under vibration, shock, and physical tampering.
These factors collectively mean that the power budget for a tactical sensor must be built around worst-case operational envelopes rather than typical-use scenarios. This requires a more systematic approach than most embedded development teams default to.
| 🛠️ DESIGN CONSIDERATION
Size your power budget to the worst-case operational scenario, not the nominal mission profile. UK defence procurement evaluators, particularly under DEFCON and DEF STAN 00-35 environmental testing clauses, will expect evidence that power reserves are maintained across the full temperature and vibration envelope. A sensor demonstrating 12-month battery life at 20 degrees Celsius that fails at -20 degrees Celsius is not field-qualified. |
Building the power budget: Energy profiling comes before architecture definition
The most common mistake in tactical sensor power design is selecting a battery chemistry and capacity after the rest of the architecture has been defined. The correct sequence runs in the opposite direction. Before any silicon selection is finalised, the engineering team should produce a detailed energy profile that maps every operational state the sensor will enter, the current drawn in each state, and the time the sensor is expected to spend in each state across a representative mission cycle.
For a ground-deployed perimeter sensor as an example, this profile might include:
- Deep sleep with RTC running: sub-10 microamp draw, 95% of mission time
- Sensor wake and signal processing: 5-50 milliamps for 50-200 milliseconds per detection event
- LoRa packet transmission: 40-120 milliamps for 50-500 milliseconds, depending on spreading factor and payload
- Secure OTA firmware update receive window: elevated current for extended periods during planned maintenance cycles.
- Tamper detection circuitry: constant low-level draw, often overlooked in initial budgets
Each of these states must be quantified using real silicon datasheet values, not marketing figures. Sleep currents in particular are frequently misrepresented. A microcontroller rated at 1.5 microamps deep sleep draws that figure from the core alone; the full system, including peripheral leakage, pull-up resistors, crystal oscillators, and any always-on sensor circuitry, will typically sit considerably higher.
|
🛠️ DESIGN CONSIDERATION Build a spreadsheet power model before selecting any silicon. Populate it with measured values from evaluation boards rather than datasheet minimums. For battery capacity planning, apply a 0.7 derating factor to nameplate capacity to account for temperature effects, end-of-life degradation, and self-discharge over a 3-5 year deployment. For systems transmitting over LoRa (covered in detail in our earlier post on LoRa for Defence Sectors) model transmission current at both minimum and maximum spreading factor to bracket the range. |
Primary battery selection for long-life tactical sensor power
For sensors that cannot be recharged in the field and must sustain multi-year deployments, primary lithium chemistry is the dominant technology choice. Within that category, the selection between lithium thionyl chloride (LiSOCl2), lithium manganese dioxide (LiMnO2), and lithium carbon monofluoride (LiCFx) involves trade-offs specific to the tactical operating profile.
Lithium Thionyl Chloride (LiSOCl2)
LiSOCl2 offers the highest energy density of any commercially available primary cell (typically 500-700 Wh/kg) and an exceptionally flat discharge curve. Operating temperature range extends to -60 degrees Celsius at the low end, making it the default choice for sensors deployed in Arctic or high-altitude environments, including those supporting UK operations in northern theatres.
The engineering concern with LiSOCl2 is the passivation effect. After a period of storage or low-current draw, a passive film forms on the lithium anode, leading to a transient voltage depression when a significant current pulse is applied. For sensors that transmit in short bursts, such as LoRa uplinks, this passivation voltage dip can cause the system to brown out and reset, even when substantial energy remains in the cell.
Mitigation strategies include: pulsed-load conditioning circuits, hybrid cell designs that pair LiSOCl2 with a supercapacitor or a high-pulse lithium capacitor to absorb current spikes, and careful firmware design that sequences radio transmissions to allow recovery time.
Lithium Manganese Dioxide (LiMnO2)
LiMnO2 is better suited to applications that require moderate pulse capability, without the passivation concerns of LiSOCl2. Energy density is lower (approximately 270-300 Wh/kg), but the cells tolerate higher continuous and pulse currents with less voltage sag, and they perform predictably across a wide temperature range. For a sensor with frequent transmission intervals (e.g., a real-time tracking payload on a deployed vehicle), LiMnO2 may offer better field reliability, even at the cost of reduced total capacity.
Hybrid Architectures for Defence-Grade Sensor Design: Primary Cell Plus Supercapacitor
For sensors with extreme pulse current requirements and multi-year standby, a hybrid architecture using a primary cell as the long-term energy reservoir and a supercapacitor as the pulse buffer is increasingly common in defence-grade sensor design. The primary cell maintains the supercapacitor charge during quiescent periods; the supercapacitor sources burst current during radio transmissions, GPS lock acquisition, or rapid ADC sampling sequences. This approach decouples energy storage optimisation from power delivery optimisation, allowing each to be selected for its specific role.
|
⚠️ CRITICAL ALERT LiSOCl2 cells require specific disposal and transport handling under UK Dangerous Goods regulations. If your sensor programme involves airfreight (including UAV-carried payloads), confirm compliance with IATA Dangerous Goods Regulations Section II for lithium batteries before finalising cell selection. This is a procurement and logistics constraint, not just an engineering one, and late-stage changes to battery chemistry can delay export licensing under UK Strategic Export Control frameworks. |
Rechargeable architectures for tactical sensors: Energy harvesting and managed charging
Not all tactical sensor deployments are incompatible with rechargeable systems. Sensors mounted on vehicles, UAVs, or fixed infrastructure with access to solar or other harvested energy sources represent a growing segment of the UK defence sensor market, particularly as programmes supporting persistent surveillance and border monitoring expand under the UK’s Integrated Review commitments.
Energy harvesting introduces its own engineering constraints that interact directly with the power budget model:
Solar Harvesting in UK Tactical Contexts
Solar harvesting for outdoor sensors in UK conditions is complicated by the relatively low peak irradiance, short winter day length, and frequent cloud cover. A harvesting circuit sized for central European summer isolation will operate at substantially reduced effective input for significant portions of the UK deployment year.
The design implication is that the secondary battery must be sized to bridge extended harvesting deficits (typically modelled on 7-14 day sequences of overcast conditions at northern UK latitudes) rather than the average daily energy balance. Maximum Power Point Tracking (MPPT) circuitry is standard practice for extracting available energy efficiently from degraded solar input, but the energy yield from UK-deployed solar panels in winter should be modelled conservatively.
Thermal and Kinetic Harvesting
For vehicle- or personnel-mounted sensors, thermoelectric generators (TEGs) and piezoelectric harvesting elements provide supplementary energy sources that are less dependent on ambient light. Harvested power levels are typically low (microwatts to low milliwatts) but can be sufficient to offset the quiescent current draw of a well-optimised sensor node, extending primary battery life or reducing charging frequency for secondary systems.
|
🛠️ DESIGN CONSIDERATION When designing solar-harvested sensor systems for UK deployment, model harvest yield using UK Met Office solar radiation datasets for the relevant latitude and deployment season. Do not use generic European solar models as they will overestimate available energy by 20-40% for northern UK deployment zones. For fixed infrastructure sensors, a 72-hour blackout tolerance is a reasonable minimum design target. |
Firmware power state management for tactical sensor longevity
Hardware selection establishes the theoretical power budget ceiling. Firmware determines how close the system actually operates to that ceiling. In our experience designing embedded systems for defence applications, poorly structured firmware is responsible for a substantial proportion of power budget overruns, often consuming two to five times more energy than the hardware should require.
The core discipline is rigorous management of power states across every system component. This extends beyond the processor sleep modes and encompasses every peripheral on the design.
Peripheral Power Gating
Sensors, analogue front ends, RF modules, GPS receivers, and memory devices all draw standby current that, in aggregate, can accumulate to significant levels. Each peripheral should be power-gated independently where possible, either via load switches controlled by GPIO lines or via device-level power-down modes. The firmware must track the state of every peripheral and ensure that devices not required in a given operational phase are fully powered down, not simply idle.
For LoRa-connected sensor nodes (as detailed in our earlier post on LoRa for Defence Sectors) the SX1276 and equivalent radio transceivers draw approximately 1.4-2 milliamps in sleep mode with the datasheet configuration. With careful register configuration, this can be reduced to the sub-microamp range. Across a thousand-node sensor network, this difference compounds significantly.
Interrupt-Driven Wake Architecture
A tactical sensor that polls for events wastes energy between detection events. The correct architecture is interrupt-driven: the processor remains in deep sleep until a hardware interrupt (e.g., from an accelerometer, a PIR detector, a timer, or an external trigger) wakes it with a specific, bounded task. On completion of that task, the processor returns to the deepest available sleep state.
This architecture requires careful handling of the MAVLink communication stack in UAV-mounted payloads, where the protocol expects periodic heartbeat messages and status updates. The firmware must maintain protocol compliance without forcing the processor into continuous active mode. Our earlier post on MAVLink Integration in Defence UAVs covers scheduling strategies that balance protocol requirements against power constraints.
Secure OTA Update Power Budgeting
Secure over-the-air firmware updates represent a significant and often underestimated power load in long-life tactical sensor deployments. A full firmware image transfer over LoRa at SF12 (the highest spreading factor, used for maximum link budget at the cost of time on air) can take 20-40 minutes and consume a meaningful fraction of a low-power node’s weekly energy budget.
The power budget must explicitly allocate energy for OTA update cycles, and the update schedule should be planned around peak energy availability (i.e., post-harvest for solar systems, or during known low-activity periods for battery-primary nodes). Differential update techniques, which transmit only the changed portions of the firmware image rather than a complete reflash, can reduce OTA energy consumption by 60-80% for minor firmware revisions. As discussed in our post on secure OTA for defence IoT, the cryptographic overhead of authenticated update verification also carries a measurable energy cost that should be included in the model.
|
⚠️ CRITICAL ALERT Failed or interrupted OTA updates can leave a sensor in an inoperable state requiring physical recovery. For sensors deployed in inaccessible locations, the power budget must guarantee sufficient energy to complete a full OTA cycle including rollback verification. Build a minimum 30% OTA energy reserve into the design. This is a field reliability requirement, not a nice-to-have margin. |
Thermal management and battery performance in deployed conditions
Battery capacity is not a fixed quantity. It varies with temperature in ways that can fundamentally alter mission duration calculations if not properly modelled.
LiSOCl2 cells, despite their excellent low-temperature floor rating, deliver measurably less usable capacity at -20 degrees Celsius than at room temperature. Depending on the discharge current profile, effective capacity at -20 degrees Celsius may be 60-75% of the room-temperature nameplate figure. For a sensor deployed in northern Scotland, Norway, or other cold-weather environments within the UK’s operational scope, this derating must be factored into the initial energy budget rather than treated as a worst-case contingency.
At the high-temperature end, primary cells operating above 60 degrees Celsius accelerate self-discharge and can develop internal pressure in poorly designed enclosures. Vehicle-mounted sensors and sensors integrated into UAV airframes, where proximity to motors, electronics, and solar-heated structural panels can elevate local temperatures well above ambient, require thermal analysis as part of the power budget process, not just as a mechanical design exercise.
|
🛠️ DESIGN CONSIDERATION Run power budget calculations at three temperature points: -20 degrees Celsius, +20 degrees Celsius, and +60 degrees Celsius. The -20 degrees Celsius case governs battery capacity derating. The +60 degrees Celsius case governs enclosure thermal design and self-discharge rates. If your system cannot pass all three cases with margin, the battery selection or thermal management approach requires revision before moving to hardware build. |
Power budget validation: From a workbench to the field
A power budget model is only as useful as its accuracy against real hardware. Validation requires a measurement infrastructure that many development teams underestimate. A multimeter is not a sufficient tool for characterising the current profile of a system with microsecond-scale transients: the averaging behaviour of most bench multimeters will significantly underread peak current and provide false confidence in the budget model.
Nordic Semiconductor’s PPK2 (Power Profiler Kit) and similar energy monitoring tools provide the time resolution and dynamic range needed to characterise the full current profile of a tactical sensor across all operational states. These measurements should be taken on final hardware builds, not evaluation boards, as PCB layout, connector resistance, and component tolerances all affect real-world current draw.
For defence procurement programmes, the power budget validation data forms part of the technical file that procurement teams will review during design acceptance. Demonstrating traceability from the design model to measured hardware performance, particularly under the temperature and vibration conditions specified by DEF STAN 00-35 or equivalent NATO STANAG testing requirements, is a programme management requirement, not merely good engineering practice.
|
⚠️ CRITICAL ALERT Do not rely on software simulation alone for power budget validation in defence programmes. UK MOD design acceptance and Integrated Test and Evaluation (IT&E) processes require empirical evidence of performance under the environmental envelope specified in the system requirements. Modelled figures without supporting measured data will not satisfy procurement assurance requirements. |
Cross-cutting considerations: Power budgeting within the wider sensor system
Tactical sensor power design does not exist in a vacuum. The power budget interacts with every other system-level design decision, and changes to connectivity architecture, physical design, or firmware strategy will affect the energy model.
In LoRa-connected sensor networks (the subject of our earlier post in this series), the choice of spreading factor and transmission frequency directly and significantly affects energy consumption. A sensor transmitting at SF7 with a 30-second interval consumes an order of magnitude less radio energy than the same sensor at SF12 with a 10-second interval, but it also affects range and link reliability. The power budget and the connectivity architecture must be co-optimised rather than designed independently.
For UAV-mounted payloads communicating over MAVLink, the sensor’s power draw appears as a load on the aircraft’s power bus. UAV bus voltage regulation typically tolerates a defined current range, and sensor payloads that generate high transient loads, particularly during transmission events, can introduce noise onto the bus, affecting flight controller performance. The sensor power design must include appropriate bus filtering and current limiting, and the peak load must be declared to the UAV integration team as a formal interface requirement.
For systems that employ secure OTA updates over constrained radio links, the energy cost of cryptographic operations used to verify update authenticity adds to the computational energy budget. This is a non-trivial consideration for microcontrollers without hardware cryptographic accelerators: a software AES-256 implementation on a Cortex-M0 will draw more power meaningfully and take considerably longer than the same operation on a Cortex-M33 with TrustZone and hardware crypto. Processor selection should reflect this if the OTA update frequency is high.
Final Thoughts: Power Budget as a First-Class Design Discipline
Power budgeting for tactical sensors is not a final check before manufacture. It is a design discipline that begins at the architecture stage and runs continuously through to field validation. Teams that treat it as such will build sensors that operate reliably across the full mission envelope. Teams that treat it as a box-ticking exercise will discover the consequences in the field, where remediation options are limited.
For UK defence programmes, where procurement timelines, safety cases, and through-life support obligations all create pressure to get the design right the first time, the investment in systematic power budget engineering pays dividends that extend well beyond the initial programme phase.
Ignitec’s electronic design and firmware engineering teams work with UK defence contractors, prime integrators, and MOD direct procurement programmes on exactly these challenges. Whether the requirement is for an initial power architecture review, detailed energy modelling, or full sensor system development from requirements through to production, we bring the embedded systems expertise and defence-sector understanding these programmes demand.
