Energy & Power · Study deck

Energy Measurement and Profiling

Picture a battery sensor whose average current looks low while short radio bursts drain it early.

Battery Bruno is your guide for this deck.

awaremeasurement

Battery Bruno, the module guide, in a scene from this chapter.

After studying this chapter

Learning objectives

You will be able to:

Choose a measurement method for sleep current, active current, radio bursts, and startup transients.
Explain burden voltage, shunt power, dynamic range, and why one fixed shunt often cannot cover the whole profile.
Build a repeatable measurement setup that isolates the target device from USB, LEDs, debug adapters, and regulator leakage.
Capture a full duty-cycle trace and connect current spikes to firmware states.

A shunt resistor like this is the known impedance this chain measures the voltage across -- the 100 ohm example just above is sensitive enough to resolve a 10 uA sleep current but would burn 5 V of headroom at a 50 mA wake current, which is exactly the burden-voltage trade-off this section is describing. Photo: wdwd, CC BY-SA 4.0

Major section

Start With the Trace, Not the Datasheet

One average can hide the states that decide real life.

Duty cycle means the share of time a device spends active instead of asleep.
Firmware means the software stored on the device.
This runway does not prove lifetime in every field condition.
Both matter, but only the trace can prove the budget.

An analog oscilloscope shows a rounded triangular trace above a square-wave trace on the same time grid, allowing the two signal shapes to be compared.

Major section

Measurement Evidence Loop

Good energy profiling is an iterative loop.

The first trace rarely confirms the original model; it usually reveals hidden loads, extra wake time, radio retries, or a board path that was not in the spreadsheet.
The visible relationship between state budget and and duty cycle identifies what must remain true in the real system.
This explains why no-panel energy measurement workflow showing budget prediction, instrument setup, duty-cycle capture, anomaly explanation, fix, and revised estimate.

Energy measurement evidence loop from predicted budget through instrument setup, trace capture, anomaly explanation, and revised battery-life estimate.

Major section

Recalculate

Sleep validation, state-cost diagnosis, battery survival, and before-and-after fixes require different measurement points.

This keeps the lifetime estimate tied to observed behavior instead of one attractive current value.

The shunt must sit in series with the measured rail, and its value must be chosen so the voltage drop is both measurable and small enough not to change device behavior.

Major section

Current Measurement Chain

The instrument may be a multimeter, oscilloscope, current-sense amplifier, auto-ranging power profiler, or source-measure unit.

At this point, current measurement chain needs an inspectable chain rather than another rule of thumb.
The first establishes the starting condition, the second adds a distinct review condition, and the third adds a distinct review condition.
If the burden voltage is too high, the device sees a lower supply voltage and may behave differently.

Current measurement chain showing supply, series shunt, sense leads, instrument, device under test, and optional firmware marker.

Major section

Tool Selection

The first screen shows one repeating trace; the second has two different traces, one above the other.

The third photo shows the knobs and input sockets used to select and measure a signal.
A trace makes timing and short changes visible, but it is not a current reading by itself.
Improves small-signal measurement while keeping the shunt low.

Analog oscilloscope displaying a sine wave with a frequency of 10 kHz, showing periodic oscillations on a grid background.

Major section

Tool Selection (continued)

For the wake and radio pulses in the tool table below, measure the voltage across a suitable shunt, choose the time scale, and check that the probe and shunt do not change the device's behaviour.

Tool choice is about current range, time resolution, burden voltage, and repeatability.
Steady current checks, simple shunt voltage readings, and quick sanity tests.
Too slow for short duty-cycle pulses and often adds noticeable burden in current mode.

Oscilloscope control panel with voltage and time knobs, buttons, indicator lights, and input connectors for setting up measurements.

Major section

Tool Selection (continued)

Low-side or high-side sensing when the shunt voltage is too small for direct measurement.

Bandwidth, offset, gain error, and common-mode range must match the device.
Captures a wide dynamic range with fewer manual range changes.
Range-switching behavior and supply-mode limits still need to be understood.

An oscilloscope like this is the "oscilloscope plus shunt" entry in the table above: it is the tool that actually shows current versus time, catching the short radio bursts and wakeup pulses a multimeter blinks straight past. Photo: BekundaOff, CC BY 4.0

Major section

Capturing a Power Profile

The trace is only useful if every region can be explained.

10 µA changes the question by adds a distinct review condition, and 1 mA completes it by adds a distinct review condition.
The map at the Current-by-state sequence compares radio transmission charge with deep sleep charge to show how current and duration determine energy use diagram exposes that gap so the next design step can use a bounded conclusion rather than a slogan.
The first concrete marker in the Current-by-state sequence compares radio transmission charge with deep sleep charge to show how current and duration determine energy use diagram is CURRENT: it exposes the energy consequence.

Current-by-state sequence compares radio transmission charge with deep sleep charge to show how current and duration determine energy use.

Major section

From Trace to Battery-Life Estimate

The most useful output of profiling is charge per useful cycle.

Total charge per cycle is about 52 uAh.
A 2400 mAh cell with 80% usable capacity would be estimated at roughly 256 days before adding further temperature, ageing, self-discharge, and reserve effects.
The point is not the exact number.

Major section

Common Findings and Fixes

Sleep current is far above budget.

Active window is longer than expected.
Batch work, shorten timeouts, remove debug logging, or change the reporting policy.
Peripheral shutdown sequence, capacitor discharge, regulator mode change, or radio not fully off.
Life: a slow return-to-sleep tail often means a shutdown sequence never actually finished.

Major section

Validation Conditions

A single room-temperature bench trace is useful, but it is not final validation.

Final hardware: development boards include loads that production hardware may not have, and production boards include leakage paths that prototypes may hide.
Battery voltage range: radio transmit current, regulator efficiency, brownout margin, and sensor behavior can change as voltage falls.
Temperature range: leakage, battery capacity, sensor settling, and oscillator behavior can shift at temperature extremes.
Multiple units: one good board does not prove a manufacturing population.

Deck summary

Key takeaways

One average can hide the states that decide real life.

Good energy profiling is an iterative loop.
Sleep validation, state-cost diagnosis, battery survival, and before-and-after fixes require different measurement points.
The instrument may be a multimeter, oscilloscope, current-sense amplifier, auto-ranging power profiler, or source-measure unit.
The first screen shows one repeating trace; the second has two different traces, one above the other.

Retrieval practice

Recall check 1 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q1How should a team update battery life after capturing a full current trace?

AUse sleep current alone as the cycle average because active-state charge is negligible over a full cycle.

BUse peak radio current alone as the cycle average because the radio peak represents its duty cycle.

CDivide cycle charge by cycle time for mean current; divide usable capacity by that current.

DAverage state currents equally because each state contributes the same share of cycle time.

Show answer

Answer: C Use consistent units: mAh ÷ mA gives hours; account for temperature, age, self-discharge, and reserve.

Retrieval practice

Recall check 2 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q2A battery sensor measures 12 uA sleep current on the isolated target rail, but 6 mA when powered through the development board USB connector. What is the best interpretation?

AIgnore isolated-rail data because USB is common in development.

BTreat target sleep as working; count extra USB-board loads.

CAssume the meter is broken because rail and USB readings differ.

DUse the lower number without verifying final product supply rails.

Show answer

Answer: B Low-power measurements must state exactly which rail is measured.

Retrieval practice

Recall check 3 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q3Why can a 100 ohm shunt be a poor choice for a device that wakes to 50 mA, even if it helps measure 10 uA sleep current?

AIt makes the 10 uA sleep signal too small to see, so only wake current is measured.

BIt converts the DC current into AC ripple, so Ohm's law no longer applies.

CIt drops about 5 V at the wake current, starving the device instead of observing it.

DIt protects the battery by limiting every wake pulse to the planned current budget.

Show answer

Answer: C Shunt selection is a tradeoff between low-current resolution and high-current burden voltage.

Retrieval practice

Recall check 4 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q4Why is a single fixed shunt often a poor measurement choice for both a 10 µA sleep current and a 200 mA transmit current?

ASleep needs a larger shunt for resolution, while transmit needs a smaller shunt for low burden.

BThe battery chemistry changes between sleep and transmit, so each phase needs a different supply model.

CThe meter ADC can sample only one current level per boot, so separate tests are always required.

DThe radio bypasses the sense resistor during transmit, so only the sleep interval is recorded.

Show answer

Answer: A A larger shunt improves the tiny sleep signal but creates more burden voltage at 200 mA; a smaller shunt reduces burden but makes 10 µA harder to resolve.

Retrieval practice

Recall check 5 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q5You need burden under 50 mV at a 200 mA peak but also a 100 uV signal at 10 uA sleep. What do the two bounds tell you?

AA 5 Ω shunt meets both burden and sleep-signal requirements at once.

BOnly the peak-current burden matters; sleep resolution can be ignored.

CBoth bounds support a 0.25 Ω shunt for the 10 µA sleep signal.

DBurden needs at most 0.25 Ω; sleep resolution needs at least 10 Ω.

Show answer

Answer: D see answers page

Retrieval practice

Recall check 6 of 6

Battery Bruno says: answer from memory, then check your reasoning.

Q6A bench multimeter reads 0.4 mA average for a duty-cycled radio node, and an engineer concludes a coin cell will be fine. Why is that conclusion unsafe?

AThe average proves the node never transmits, so the radio duty cycle must be disabled.

BThe average can hide a short transmit peak that may exceed what the coin cell can supply.

CCoin cells respond only to total charge, so the peak current has no effect on voltage sag.

DA multimeter cannot measure milliamp averages, so the 0.4 mA reading is unusable.

Show answer

Answer: B see answers page

Print reference

Answers 1 of 3

Answer key.

C · Use consistent units: mAh ÷ mA gives hours; account for temperature, age, self-discharge, and reserve.
B · Low-power measurements must state exactly which rail is measured.
C · Shunt selection is a tradeoff between low-current resolution and high-current burden voltage.
A · A larger shunt improves the tiny sleep signal but creates more burden voltage at 200 mA; a smaller shunt reduces burden but makes 10 µA harder to resolve.

Print reference

Answers 2 of 3

Answer key.

D · 0.050/0.200 = 0.25 ohm caps the shunt for burden; 100 uV/10 uA = 10 ohm floors it for resolution. The bounds are 40x apart and cannot both be met by one resistor, which is precisely what auto-ranging solves.

Print reference

Answers 3 of 3

Answer key.

B · The 0.4 mA average says nothing about a 200 mA pulse buried within it. Coin-cell feasibility depends on the peak pulse against the cell's internal resistance, which a slow average-only reading cannot reveal. A high-bandwidth capture is needed to see the peak.