Measuring µA-Level Standby Current: Methods, Fixtures and Pitfalls

In low power MCUs, standby current is the tiny draw when the processor sleeps in deep-sleep or shutdown modes. For battery powered designs, minimizing this standby current is crucial. Leading vendors have driven standby currents into the nanoamp range. For example, STMicroelectronics’ STM32U0 series (based on an Arm® Cortex®-M0+) consumes only about 160nA in standby and down to 16nA in shutdown. Likewise, NXP Semiconductors’ latest MCX L-series (Cortex-M33) advertises deep-sleep currents “down to sub-µA”.

These μA/nA levels make measurement challenging: one must ensure the MCU truly entered its lowest-power mode (and disable clocks, pull-ups, debug interfaces, etc.) before measurement.

Accurate standby current measurement is the first step in any standby current reduction effort, as it validates that low power modes are configured correctly.

Methods of Standby Current Measurement

Measuring microamp-level currents generally requires converting the current into a more easily measured voltage. A common method is low-side current sensing: place a small precision resistor in the ground return path of the MCU and measure the voltage drop across this shunt with a high-resolution amplifier or ADC. As Analog Devices notes, “the simplest way to measure current…is with a small sense resistor between the load and system ground”. In practice, you can amplify that tiny drop with an instrumentation amplifier, a dedicated current-sense amplifier, or even the MCU’s own differential ADC inputs.

A source-measure unit (SMU) can bias the MCU at a fixed voltage and directly report the current. High end bench SMUs can measure down to picoamps, but a high-quality DC supply with built-in current readout (range into µA) may suffice. However, Tektronix warns that IoT devices span “from tens of nanoamps to hundreds of milliamps in a matter of microseconds” and that conventional instruments rarely cover this entire range. Auto-ranging can take time and introduce glitches.

One convenient tool is the Nordic Semiconductor Power Profiler Kit II (PPK2), a USB-powered instrument designed for MCU current profiling. The PPK2 can act as either an ammeter or as a supply (providing from 0.8V to 5.0V and up to 1A). It can measure currents from around 0.2µA up to 1A with a resolution of between 0.1µA and 0.2µA. In ammeter mode, connect the MCU’s V_cc and ground leads through the PPK2 and an external supply. In source mode, the PPK2 itself powers the MCU. The PPK2 also provides eight logic inputs to allow for the insertion of markers in code, and for correlation of power measurements with program execution. Such integrated tools simplify long-term monitoring and averaging of sub-µA currents.

In the classic low-side current-sense setup, a precision shunt resistor (R_sense) is placed in series with the MCU’s ground. An instrumentation amplifier then measures the tiny drop across the resistor. The amplifier references the drop to the MCU ground, producing a voltage proportional to current. This method is straightforward but requires careful attention to the amplifier’s input leakage and offset.

Measurement Fixtures and Setup

Fixtures for µA-level measurement aim to minimize leakage, noise and unwanted drops. Use short, thick wires or traces to connect the shunt resistor, and consider a four-terminal (Kelvin) wiring scheme: two wires carry the current through R_sense and two separate sense wires connect the amplifier across it. This eliminates lead resistance error. Keep the shunt as close to the MCU as possible to avoid stray PCB leakage.

If using a development kit or custom board, isolate the MCU’s power by rerouting V_cc and GND through the measurement device. For example, Nordic DK boards have jumpers to insert the PPK2 inline. Avoid powering the MCU from USB or regulator pins that are not measured – the source must go through the meter.

If using the PPK2 in source mode, set it to the desired supply voltage and connect the device under test (DUT) to its + and – terminals. In ammeter mode, supply the MCU from a bench supply, but route the return through the PPK2. In either case, set filtering or sampling to capture steady state or transient events as needed.

Common Pitfalls and Best Practices

Measuring nanoamp-level currents is extremely error-prone. Tektronix cautions that sleep-mode currents can be “as low as tens of nanoamps,” so even the smallest error source dominates. Key pitfalls include:

• Instrument-sourced voltage distortion: Many ammeters insert some series resistance. On a 200µA range this could be 0.5V or more, which alters the DUT’s conditions. Always choose a meter with minimal voltage overhead, or use a 4-wire sense set-up.
• Leakage and bias currents: Input bias currents on multimeters or amplifiers, moisture on cables or dielectric absorption can create offsets. Tektronix lists cable leakage and ammeter input bias as error sources. Even walking near the setup can induce triboelectric charges. Use a shielding guard or baking to reduce moisture, and keep cables short.
• Ground reference shift: In a low-side sensing set-up, the MCU’s ground is lifted by the shunt drop. Analog Devices notes that this means “the load floats above system ground by the sense voltage – a moving ground reference is no reference at all”. Any noise or fluctuation across the shunt thus directly perturbs the MCU. Mitigate by keeping R_sense small (see next point) and by filtering if needed.
• Resistor trade-offs: A large R_sense (eg 10Ω) makes the drop easy to measure but wastes power and raises the MCU ground significantly. A very small R value, such as 0.1Ω, reduces loading but pushes the reading closer to the amplifier’s noise floor. Tektronix warns that using too small a shunt “will degrade the signal-to-noise ratio” and slow the measurement due to smaller voltage swings. Choose R_sense to give a reasonable voltage (millivolts) at your expected current, and average multiple readings if noisy.
• Device configuration: Disable all peripherals, oscillators and debug interfaces. Some MCUs have multiple sleep modes (stop, standby, shutdown): verify that the deepest mode has been selected. For example, ST’s 16nA shutdown mode calls for a special register setting, and for SRAM retention to be enabled. Otherwise, standby current might be 160nA.
• Temperature and environment: Leakage currents often increase with temperature. Perform measurements in a stable environment. Keep probes and hands away from sensitive nodes.

Measurement of µA-level standby current often requires averaging over many samples and controlling the test setup as stringently as the design itself. Double-check connections, account for any voltage drop, and if possible validate your set-up by measuring a known current source. Integrating the results (by coulomb counting) or using a dedicated power profiler (like the PPK2) can help confirm the long-term average current. By carefully combining the right instrumentation, fixture and configuration, designers can accurately gauge and minimize their MCU’s standby power draw.

Sources: This article draws data from recent manufacturer announcements and technical data sheets to illustrate typical standby currents, measurement tools and error mechanisms. It is good practice to verify the data from the original sources before using it in a design project.