Learn › Topic 12

How to Use an Oscilloscope: Scales, Triggering and Measurements

7 min read · bitGateLab Team

A multimeter gives you a number; an oscilloscope shows you the shape of the signal over time. With it you can see whether a sine wave is clipped, how much ripple a power supply has, or how far a filter's output lags its input. At first the number of knobs is intimidating, but it all comes down to four ideas: the vertical scale, the horizontal scale, the trigger and the coupling. Here you'll learn to set them and to measure amplitude, period, frequency and phase shift with concrete numbers.

The screen: a graph of voltage versus time

The oscilloscope screen is divided by a graticule, usually with 10 horizontal divisions and 8 vertical ones. The vertical axis represents voltage and the horizontal axis, time. Every measurement consists of counting divisions and multiplying them by the corresponding scale:

voltage = vertical divisions × (V/div)
time    = horizontal divisions × (s/div)

Each division usually has five subdivisions marked along the center axes, which lets you read with a resolution of 0.2 divisions.

Vertical scale (V/div) and position

The vertical scale control sets how many volts each division represents. The values follow the 1-2-5 sequence: 10 mV, 20 mV, 50 mV, 100 mV, 200 mV, 500 mV, 1 V, 2 V, 5 V… The position control moves the trace up or down without changing the scale; use it to put the 0 V reference wherever suits you.

The rule of thumb is to choose the most sensitive scale at which the whole signal still fits on the screen. A signal that spans 6 divisions is measured far more accurately than one that spans half a division.

If you use a ×10 attenuating probe, the actual voltage is ten times what you would see without compensating for it. Set the channel to the probe's attenuation, or multiply the reading yourself.

Horizontal scale (s/div)

The time base sets how much time each horizontal division represents. The whole screen spans 10 × (s/div). To measure the period accurately it is best to see between two and five complete cycles: with less than one you cannot measure the period, and with dozens of cycles the details get crowded together.

Two 500 Hz sinusoids on a graticule of 10 × 8 divisions. CH1 measures 6 divisions peak to peak and its period spans 4 divisions. CH2 crosses zero going up 0.4 divisions after CH1.

Example 1: amplitude, period and frequency. In the figure, the vertical scale is 2 V/div and the horizontal scale is 0.5 ms/div.

CH1 goes from +3 to −3 divisions: 6 div × 2 V/div = 12 V peak to peak.
Amplitude: 12/2 = 6 V peak. If it is a sinusoid, its RMS value is 6/√2 = 4.24 V.
One complete cycle spans 4 divisions: T = 4 × 0.5 ms = 2 ms.
Frequency: f = 1/T = 1/0.002 s = 500 Hz.
CH2 measures 4 divisions peak to peak: 8 V peak to peak, or 4 V peak.

The trigger

The oscilloscope draws the signal by sweeping across the screen over and over. If each sweep started at a different point on the waveform, you would see the traces piled on top of one another and "running". The trigger solves this: the sweep starts only when the signal crosses a certain voltage in a certain direction, so all the sweeps line up and the picture holds still. Its main settings are:

Source: which channel the trigger watches. When comparing two signals, trigger on the reference one.
Level: the voltage the signal has to cross. It must be within the signal's range; if you set it above the signal's peak, the oscilloscope never triggers.
Slope: rising edge or falling edge.
Mode: in auto the oscilloscope sweeps even when there is no trigger, useful for finding the signal; in normal it draws only when there is a trigger, better for irregular signals; in single it captures one event and stops, ideal for transients such as a power supply turning on.

Phase shift between two channels

With two signals of the same frequency on screen, measure the time separation Δt between two equivalent points, for example the rising zero crossings, and convert it to degrees:

φ = 360° × Δt / T

If the second signal arrives after the reference, it is lagging; if it arrives before, it is leading. For better accuracy, expand the time base around the crossings, or use the same vertical scale on both channels and center the traces on zero.

Example 2: phase shift. In the figure, CH2 crosses zero going up 0.4 divisions after CH1.

Δt = 0.4 div × 0.5 ms/div = 0.2 ms.
φ = 360° × 0.2 ms/2 ms = 36°.
CH2 lags CH1 by 36°.

Application: in the series RC circuit from alternating current and phasors (1 kΩ, 1 µF, 100 Hz), the voltage across the resistor, which has the same shape as the current, leads the source by 57.9°. With T = 10 ms you would expect to see Δt = 57.9/360 × 10 ms = 1.61 ms. This is how the phase of a current is measured with an oscilloscope: through the voltage across a resistor.

DC and AC coupling

Each channel has a coupling selector:

DC: shows the complete signal, including its average level. It is the default mode and the right one for measuring DC voltages, logic levels or low-frequency signals.
AC: inserts a capacitor that blocks the DC level, so the trace ends up centered on zero. It is equivalent to a high-pass filter with a cutoff frequency of a few hertz.
Ground: disconnects the signal and shows a horizontal line at 0 V, useful for locating the reference.

Example 3: measuring ripple. A power supply delivers 5 V DC with 0.2 V peak to peak of ripple.

With DC coupling at 2 V/div the trace appears 2.5 divisions above center, and the ripple spans barely 0.1 division: you cannot make it out.
If you go down to 50 mV/div without changing the coupling, the 5 V sends the trace off the screen.
With AC coupling the 5 V level is removed. At 50 mV/div the ripple spans 0.2/0.05 = 4 divisions and you can measure its shape and frequency.

Careful: because AC coupling is a high-pass filter, it distorts very low-frequency signals. A slow square wave shows up with its flat tops tilted.

Common mistakes

Setting the trigger level outside the signal's range and concluding that "there is no signal".
Forgetting the probe's attenuation factor and reporting a voltage ten times too small.
Measuring a DC voltage with AC coupling: the trace stays at zero.
Measuring the phase shift between crossings of opposite slope (one rising and one falling), which adds 180°.
Choosing a time base that shows less than one cycle or dozens of them, which makes the period reading imprecise.

Try it in the simulator

The simulator's virtual oscilloscope has conceptually the same controls as a lab oscilloscope, so it is a good place to practice without risk.

Place a sine-wave source with 6 V amplitude and 500 Hz and connect it to channel 1.
Set the vertical scale to 2 V/div and the horizontal scale to 0.5 ms/div, and check that you see 6 divisions peak to peak and 4 divisions per cycle.
Move the trigger level above 6 V and see what happens to the picture; then return it to 0 V.
Build an RC filter and connect channel 2 to its output; measure Δt between the zero crossings and calculate the phase shift.
Add a DC level to the source and switch between DC and AC coupling to see the difference.

Exercises

A sinusoid spans 3.2 divisions peak to peak at 5 V/div. Calculate its peak-to-peak, peak and RMS voltage.

Show answer
3.2 × 5 = 16 V peak to peak; 8 V peak; 8/√2 = 5.66 V RMS.
One cycle spans 2.5 divisions with the time base at 20 µs/div. What is the frequency?

Show answer
T = 2.5 × 20 µs = 50 µs. f = 1/50 µs = 20 kHz.
Two 1 kHz signals are 1.5 divisions apart at 0.1 ms/div. What is the phase shift?

Show answer
Δt = 1.5 × 0.1 ms = 0.15 ms and T = 1 ms. φ = 360° × 0.15/1 = 54°.
You want to see at least three complete cycles of a 1 kHz signal on a 10-division screen. Which time base from the 1-2-5 sequence do you choose?

Show answer
Three cycles last 3 ms, so you need at least 0.3 ms/div. The first value in the sequence that meets this is 0.5 ms/div, which shows 5 ms, that is, 5 cycles. With 0.2 ms/div you would see only 2.