Electronics Design AUHow Does an Oscilloscope Work and What Can It Measure?
Last updated 27 June 2026 · 7 min read
Direct Answer
An oscilloscope plots voltage against time on a screen, letting you observe waveforms, measure voltage levels, timing, frequency, rise times, and signal integrity. Key settings are: timebase (time per division — controls how much of the signal fits on screen), voltage scale (volts per division — sets amplitude resolution), trigger (the condition that causes the sweep to start, synchronising a repetitive waveform), and coupling (AC removes the DC offset, DC shows the full signal including its DC level). A 10× probe increases measurement bandwidth and reduces probe loading on the circuit.
Detailed Explanation
An oscilloscope is the most versatile instrument in electronics debugging. A multimeter gives you a number; a scope gives you the waveform — how voltage changes over time. That time dimension is what reveals switching transients, clock jitter, I2C timing violations, power supply noise, and countless other problems that a multimeter can never observe.
The Display: Grid and Divisions
An oscilloscope display is divided into a grid of 8–10 vertical divisions and 10–12 horizontal divisions. The two key settings control what each division represents:
- Voltage scale (V/div): How many volts each vertical division represents. A 1 V/div setting means the full vertical scale (8 divisions) spans 8 V.
- Timebase (s/div or time/div): How much time each horizontal division represents. A 1 ms/div setting means the full horizontal span (10 divisions) shows 10 ms of signal.
Choosing these settings is the first step in measuring any signal: set V/div so the waveform occupies most of the vertical screen without clipping, and set time/div so 2–3 full cycles of the waveform are visible.
Trigger
The trigger is the condition that starts (or "triggers") the oscilloscope's sweep. Without triggering, a repetitive waveform scrolls continuously across the screen and appears as a blurred mess. With a correctly set trigger, the oscilloscope starts its sweep at the same point in each cycle, stacking identical sweeps on top of each other and producing a stable image.
Edge trigger (most common): The sweep starts when the signal crosses a set voltage level (the trigger level) in the selected direction (rising or falling edge). For most digital signals, set the trigger level to half the signal amplitude on the rising edge.
Trigger coupling: Usually set to DC — the trigger circuit sees the full signal. AC trigger coupling removes the DC component of the signal for triggering purposes, useful when the trigger signal has a large DC offset combined with a small AC event.
Trigger mode:
- Normal: The oscilloscope only updates the display when a trigger event occurs. Good for rarely-occurring events.
- Auto: The oscilloscope updates the display periodically even without a trigger — useful for initial setup when you aren't sure what the signal looks like yet.
- Single: Captures one trigger event and holds it — essential for capturing one-shot events (power-on glitches, fault conditions).
Probe Selection and Compensation
An oscilloscope probe is not a passive wire — it affects the circuit being measured. The probe tip presents a load (typically 1 MΩ resistance and 10–15 pF capacitance for a 10× probe) at the measurement point, and the probe cable adds more capacitance.
Probe compensation: Most oscilloscope probes have a small trimmer capacitor inside the probe head that must be adjusted so the probe's frequency response is flat. Attach the probe to the scope's calibration output (a 1 kHz square wave reference present on almost all oscilloscopes) and adjust the trimmer until the square wave has flat tops with sharp edges:
- Flat top: correctly compensated
- Drooping top: under-compensated (too much capacitance)
- Overshoot spike on top: over-compensated (too little capacitance)
A poorly compensated probe distorts every measurement — compensation takes 30 seconds and should be done every time a probe is connected to a new scope.
Ground connection: The probe's ground clip must be connected to a circuit ground near the measurement point. A long ground lead adds inductance that creates ringing on fast edges. For high-frequency measurements, use a short ground spring accessory (a coiled wire that clips directly to the probe tip's ground ring) instead of the standard alligator-clip ground lead.
AC and DC Coupling
DC coupling: The oscilloscope's input is directly connected — the signal passes to the ADC with no filtering. Shows the complete signal including DC offset.
AC coupling: A capacitor in series with the input removes the DC component. Useful for measuring small AC signals on top of large DC levels.
Example: Power supply noise measurement. A buck converter's 5 V output with 50 mV ripple at 100 kHz. With DC coupling at 1 V/div, the 5 V DC level takes up 5 divisions, leaving little screen space to observe the 50 mV (0.05 V) ripple — it looks like a flat line. Switch to AC coupling, the 5 V DC is blocked, and now you can zoom to 10 mV/div to see the ripple waveform clearly.
Key Measurements
Period and frequency: Measure period (time from one rising edge to the next) and calculate frequency = 1/period. Most oscilloscopes include automatic frequency and period measurements in their "measure" menu — but always verify the automatic measurement makes sense against what you see on screen.
Rise time: Time for a signal to transition from 10% to 90% of its final amplitude. A proxy for signal bandwidth: bandwidth ≈ 0.35 / rise time (seconds). If a 3.3 V digital output shows a 5 ns rise time, the signal has content to ~70 MHz — relevant for PCB routing length and series termination considerations.
Voltage measurements: DC average, RMS, peak-to-peak, overshoot. The automatic measurement menu provides these; for manual measurement, count divisions from the waveform baseline to the peak and multiply by V/div.
Propagation delay: Time between an event on channel 1 (e.g. a GPIO toggle) and a corresponding event on channel 2 (e.g. an interrupt response). Use two channels, align the trigger to the channel 1 event, and measure the time to the channel 2 event using cursors.
Common Use Cases in Electronics Debugging
| What you're investigating | Scope setup |
|---|---|
| Digital signal presence and logic levels | 1–2 V/div, 1–10 µs/div, edge trigger on the signal |
| UART/SPI/I2C bit timing | 0.5–1 V/div, 10–100 µs/div, edge trigger on clock or first edge |
| Power supply ripple | AC coupling, 10–50 mV/div, 10–100 µs/div |
| Power-on transient | DC coupling, 1–2 V/div, 1–10 ms/div, single trigger on rising edge |
| Interrupt response latency | Two channels: GPIO on Ch1, ISR output on Ch2, rising edge trigger Ch1 |
| Clock jitter | 0.5–1 V/div, very narrow timebase (1–10 ns/div), infinite persistence mode |
For protocol decoding (I2C, SPI, UART frames), a logic analyser is usually more practical than a scope — it shows full decoded transactions where a scope shows raw voltage waveforms. The two tools are complementary, not interchangeable.
Design Considerations
- Bandwidth margin: Measure with a probe and oscilloscope whose bandwidth is at least 3–5× the highest frequency present in the signal. Attempting to measure a 50 MHz signal edge with a 50 MHz oscilloscope produces a significantly rounded result — the instrument's own bandwidth is limiting the measurement.
- Input impedance and probe loading: High-impedance oscilloscope inputs (1 MΩ) load the circuit at DC and low frequency. For measuring high-impedance nodes (op-amp outputs into multi-megaohm loads, ADC reference circuits), the probe's capacitive load at high frequency can alter the circuit behaviour — measure from a low-impedance node or use active probes with lower input capacitance.
Common Mistakes
- Leaving the probe on 1× setting in the oscilloscope's probe configuration while using a 10× probe — the scope displays voltages 10× higher than actual.
- Using the auto-trigger mode to observe a one-shot event, then wondering why the display shows noise after the event — use single trigger mode for transients.
- Not connecting the probe ground, or using a long ground lead, when measuring fast edges — the long ground lead inductance rings and appears as signal ringing that is not present in the actual circuit.
- Interpreting automatic measurements without checking they match what is visible on screen — automatic peak-to-peak and frequency measurements can produce nonsensical values when the trigger is not correctly set or the waveform clips the screen.
Frequently Asked Questions
- What bandwidth oscilloscope do I need for embedded systems work?
- A rule of thumb is that oscilloscope bandwidth should be at least 5× the highest frequency you need to measure accurately. For digital signals, the frequency that matters is the rise time, not the clock frequency: a signal with a 10 ns rise time has significant harmonic content to ~35 MHz (0.35 / 10 ns). For most 3.3 V / 5 V embedded systems work at clock speeds below 100 MHz, a 100–200 MHz oscilloscope is sufficient. For high-speed interfaces (USB, Ethernet, DDR memory) or RF applications, 500 MHz–1 GHz bandwidth is needed. The widely used Rigol DS1054Z (50 MHz, hackable to 100 MHz) and Siglent SDS1104X-E (100 MHz) are good entry-level choices for embedded work.
- When should I use AC coupling versus DC coupling?
- DC coupling shows the complete signal: its DC level plus any AC component. Use DC coupling by default. AC coupling blocks the DC component and shows only the AC variation — useful when you want to see small ripple or noise superimposed on a large DC supply voltage (e.g. 20 mV ripple on a 3.3 V rail — DC coupling at a scale that shows the 3.3 V level makes the 20 mV ripple invisible; AC coupling removes the 3.3 V offset and lets you zoom in on the ripple). Be aware that AC coupling introduces a high-pass filter (typically at 10 Hz) that distorts signals at low frequencies — never use AC coupling when measuring waveforms below ~100 Hz.
- What is the difference between a 1× and 10× probe?
- A 1× (1:1) probe passes the measured signal directly to the oscilloscope input with minimal attenuation. A 10× (10:1) probe has a 9 MΩ resistor in the probe tip that, combined with the oscilloscope's 1 MΩ input impedance, forms a 10:1 voltage divider — so the oscilloscope sees 1/10 of the actual signal voltage. The benefit of the 10× probe is much higher bandwidth and much lower capacitive loading on the circuit (typically 10–15 pF vs 100+ pF for a 1× probe). For most measurement work above DC, use the 10× probe; the oscilloscope's software compensates for the 10× attenuation automatically once you select '×10' in the probe settings. Use 1× probes only for very low-amplitude signals where the 10× attenuation makes the signal too small to measure.
References
Related Questions
Logic Analyser vs Oscilloscope: Which Should You Use?
A logic analyser decodes protocols (I2C, SPI, UART). An oscilloscope shows analog waveforms and signal integrity. Covers when to use each for debugging.
How Do You Debug Embedded Firmware?
Covers JTAG/SWD hardware debugging, printf over UART or SWO trace, and logic analyser use for embedded firmware on STM32, ESP32, and other MCU platforms.
PCB Bring-Up Checklist: First Power-On for a New Board
Covers PCB bring-up: pre-power inspection, short-circuit checks, current-limited power-on, rail verification, firmware loading, and peripheral validation.
How Do You Use a Multimeter for Electronics Work?
Covers DC/AC voltage, resistance, continuity, and current measurement — probe placement, range selection, series connection, and common mistakes.
Sensor Signal Conditioning Basics
Covers signal conditioning for sensors: op-amp gain, offset, 4-20mA interface, anti-aliasing filter design, ADC input protection, and two-point calibration.
What Is JTAG?
JTAG (IEEE 1149.1) is a 4-pin interface for testing and debugging ICs via boundary scan. Learn how the TAP state machine, JTAG chains, and debug probes work.