Betavolt Nuclear Battery Reality Check: Sizing, Limits, and 2026 Hybrid Design

When Betavolt Technology announced the BV100 betavoltaic cell, the maker and off-grid communities lost their minds. The promise of a 50-year lifespan powered by nickel-63 beta decay sounds like the holy grail for remote power systems. But as with any new energy tech, the leap from press release to workbench reality is governed by unforgiving physics. A Betavolt nuclear battery outputs microwatts, not kilowatts. If you are designing a remote power system in 2026, you need to know exactly where this technology fits and where it fails.

This guide strips away the hype. We will build the system block from source to load, run the sizing math with Peukert and efficiency factors, and establish a hard decision tree that terminates in a concrete part pick for your specific application.

System Block Architecture: Betavolt Source to Load

You cannot wire a Betavolt BV100 directly to a load and expect stable operation. The BV100 outputs a nominal 3V at roughly 100µW (microwatts). Because beta decay is constant, the internal impedance is relatively high, and the current is microscopic (~33µA). To make this usable, you must design a hybrid system block:

1. Source: Betavolt BV100 array (provides continuous micro-trickle charge).
2. Harvester/Charge Controller: An ultra-low voltage boost converter with cold-start capabilities (e.g., e-peas AEM10941 or TI bq25570). Standard solar MPPTs will not work here; their quiescent draw exceeds the Betavolt output.
3. Buffer Storage: A secondary energy reservoir (LiFePO4 cell or supercapacitor) to handle peak load transients that the Betavolt cannot supply directly.
4. Power Conditioning: A micro-inverter or DC-DC buck/boost converter to match the buffer voltage to the load.
5. Load: The end device (e.g., LoRaWAN sensor, medical implant, or aerospace telemetry).

Array Wiring: Series vs. Parallel Consequences

Because a single BV100 only produces 3V, you will likely need to build an array to reach the cold-start voltage of your harvester IC. The rules of series and parallel wiring apply to betavoltaics exactly as they do to chemical cells, but the consequences of getting it wrong are different.

Series Wiring (Voltage Adds, Capacity Stays the Same)

If you wire four BV100 cells in series, your array outputs 12V, but your current remains capped at ~33µA (100µW total). Consequence: This is the preferred method for Betavolt arrays because ultra-low-power harvesters like the e-peas AEM10941 require a minimum input voltage (typically 3V to 5V) to initiate cold-start. Wiring in series gets you past the harvester's voltage threshold without increasing parasitic losses.

Parallel Wiring (Current Adds, Voltage Stays the Same)

If you wire four BV100 cells in parallel, your array outputs 3V at ~132µA (400µW total). Consequence: While this increases total wattage, 3V is dangerously close to the dropout voltage of many boost converters under load. Furthermore, if one Betavolt cell degrades or is shaded (though light doesn't affect nuclear cells, physical damage to the diamond semiconductor does), it can create a reverse-bias parasitic drain on the parallel cells.

Sizing Math: Peukert, Efficiency, and the Stated Loads

To prove where this tech belongs, let us run the sizing math for two distinct stated loads. We must account for boost converter efficiency ($\eta$) and, for the buffer battery, Peukert's Law and Depth of Discharge (DoD).

Load A: 50mW Remote IoT Sensor (The Betavolt Use Case)

Assume a sensor drawing 50mW average. The BV100 outputs 0.1mW. Factoring in an 80% ($\eta = 0.8$) boost converter efficiency:

• Required raw power = 50mW / 0.8 = 62.5mW.
• BV100 cells needed = 62.5mW / 0.1mW = 625 cells.

At an estimated early-adopter price of $50+ per cell, this is a $31,000 power source for a 50mW load. The math dictates that Betavolt is currently viable only for sub-10µW continuous loads (like pacemakers or deep-space micro-telemetry).

Load B: 500W Remote Telecom Station (The Conventional Fallback)

For a 500W continuous load, Betavolt is mathematically impossible (you would need millions of cells). We default to a 24V LiFePO4 buffer bank. Let us size the buffer using Peukert's equation: $t = C / I^k$.

• Load Current: 500W / 24V = 20.8A.
• Battery Bank: 24V 200Ah LiFePO4 (Nominal Capacity $C = 200Ah$).
• Peukert Exponent ($k$): 1.05 (typical for high-quality LiFePO4).
• Effective Runtime ($t$): $200 / (20.8^{1.05}) = 200 / 23.1 = 8.65$ hours.
• Apply 80% DoD Limit: $8.65 \times 0.8 = 6.9 hours$ of autonomous runtime before the buffer is depleted.

Buffer Battery Limits and Lithium Fire Safety

Because the Betavolt source cannot handle transient loads, the buffer battery takes the abuse. The Betavolt array has no C-rate limits—it simply decays at a constant rate. However, your LiFePO4 buffer is strictly bound by its C-rate (usually 1C max discharge, 0.5C max charge) and its Depth of Discharge (DoD) limits to preserve cycle life.

Inverter and Harvester Sizing

Sizing the 'inverter' or charge controller depends entirely on which side of the decision tree your load falls on.

For Sub-1W Loads (Betavolt Path)

You do not use a standard AC inverter. You use an ultra-low-power energy harvesting PMIC (Power Management IC). The e-peas AEM10941 is the benchmark here. It features a cold-start voltage of just 380mV and a quiescent current of 50nA. Size the inductor on the PCB to handle the peak transient current of your sensor (e.g., 500mA for a LoRaWAN transmit burst), relying on the buffer capacitor or micro-battery to supply the burst while the AEM10941 replenishes it.

For >10W Loads (Conventional Path)

If your load exceeds 10W, abandon betavoltaics. For our 500W stated load (Load B), you need a pure sine wave inverter sized for the continuous load plus a 20% surge margin.

• Inverter Sizing: 500W $\times$ 1.2 = 600W minimum. Pick a Victron Phoenix 24/800 (800VA / 700W continuous) to handle inductive startup surges.
• Charger Sizing: To recharge the 200Ah buffer at a safe 0.2C rate, you need a 40A charge controller. A Victron SmartSolar MPPT 150/40 paired with a 600W solar array will replenish the daily draw.

Decision Tree: Betavolt vs. Conventional Storage

Use this decision matrix to terminate your design process and select your power architecture. Do not force Betavolt into a high-draw application; the physics will not bend to your will.

The Final Default Recommendation

If you are building a remote environmental sensor node drawing an average of 2mW, the math terminates here: Build a series array of ten Betavolt BV100 cells to achieve 30V, feed that into an e-peas AEM10941 evaluation board, and buffer it with a single 3.2V 10Ah LiFePO4 cell. This guarantees 50 years of baseline trickle charge without chemical degradation.

However, if your stated load is anything requiring an AC inverter, motor control, or continuous draw above 5W, the Betavolt nuclear battery is a distraction. Default immediately to a 24V LiFePO4 bank paired with a Victron MPPT and Phoenix inverter ecosystem. The Betavolt revolution is real, but in 2026, it is strictly a microwatt revolution.