HOW-TO GUIDE
How to Measure Pulse Width With an Oscilloscope
By the OhmSavvy Research Team ·
Check where the ground clip goes
- A bench scope's probe ground is tied to mains earth. Clip it only to the circuit's ground; on any other node it creates a short through the scope.
What pulse width means on an oscilloscope
A pulse is a brief excursion from one level to another and back. Its width is how long that excursion lasts, but because no edge is perfectly vertical you need an agreed point on each edge to measure from. The standard choice is the 50% level: halfway between the pulse's base (low) level and its top (high) level.
Two widths, one signal
1Connect the probe and scale the pulse
Use a 10x probe with the channel's probe setting matched, and make sure it is compensated: a miscompensated probe rounds or spikes the edges and shifts the 50% crossings. Our guide to compensating an oscilloscope probe takes a minute and is worth doing first.
Choose DC coupling so the base level stays where it really is, then set V/div so the pulse fills four to six divisions without clipping off the top. The scope needs the whole top and base to find 50%.
2Choose a timebase that spreads the pulse out
The most common mistake is a timebase that is far too slow. If a pulse covers a fraction of a division, its edges are only a few samples wide. Aim for three to five divisions.
Divide the expected width by four and round to the nearest 1-2-5 timebase step:
| Expected pulse width | Starting time/div | Pulse spans about |
|---|---|---|
| 1 µs | 200 ns/div | 5 divisions |
| 10 µs | 2 µs/div | 5 divisions |
| 200 µs | 50 µs/div | 4 divisions |
| 1 ms | 200 µs/div | 5 divisions |
3Trigger on the leading edge
Set an edge trigger on the channel carrying the pulse. For a positive pulse choose the rising slope; for an active-low pulse choose falling. Put the trigger level near the middle of the swing, which is also the 50% point, so the leading edge lands on the trigger position each sweep and the trailing edge sits a predictable distance to its right. Our oscilloscope triggering guide explains slope, level and holdoff in full.
4Read the width: automatic measurement vs cursors
Open the measure menu and add +Width (sometimes called Positive Pulse Width) and, if relevant, −Width. The scope finds the top and base levels, works out 50%, and reports the time between crossings, often averaged across every pulse in the capture.
Time cursors are the cross-check. Place cursor A where the rising edge crosses mid-swing and cursor B where the falling edge crosses the same height; the readout shows Δt between them.
5Capture a one-off pulse with Single mode
Many pulses you need to time do not repeat: a reset line releasing at power-up, a sensor's one-shot output, a button press. In Auto mode the scope keeps sweeping and the pulse flashes past. Instead:
- Set V/div and time/div for the pulse you expect, using the table above as a guide.
- Set the edge trigger on the leading edge, with the level halfway through the expected swing.
- Press Single. The scope arms and waits.
- Make the event happen. The scope captures one trigger and stops.
- Measure the frozen pulse with the automatic width or cursors.
Using the pulse-width trigger
Many digital scopes offer a pulse-width trigger, which fires only on pulses that are wider than, narrower than, or within a time you set. It turns a measurement into a search. If a sensor pulse should always be 200 µs but occasionally comes out short, set the trigger to fire on pulses narrower than 180 µs and leave it in Normal or Single mode. The screen stays still until a short one arrives, then captures it. Menu names vary, so check your scope's manual.
Very narrow pulses: when bandwidth sets the limit
Every scope has its own rise time, approximately 0.35 ÷ bandwidth for a typical front end. A pulse that is only a few times wider than that gets its corners rounded and its top pulled down, so the 50% level shifts and the width reading drifts. As a rule of thumb, keep the pulse at least three to five times wider than the scope's rise time, and use a sample rate that puts several samples on each edge. Our oscilloscope bandwidth explainer covers why.
Bars use a logarithmic scale: each step is ten times the last.
Pulse width vs duty cycle
Pulse width is an absolute time, and it applies to any pulse, repeating or not. Duty cycle is a ratio: the width divided by the period of a repeating signal. If you are setting a servo position or checking a single strobe, width is the number that matters. If you are checking a motor or LED dimming signal that repeats at a fixed rate, our guide to measuring PWM duty cycle with an oscilloscope adds the period and average-voltage side.
Which oscilloscope you need for pulse width
For microcontroller and logic pulses down to tens of nanoseconds, a 50MHz to 100MHz scope at 1 GSa/s is a comfortable fit. A pocket scope at 200kHz suits only slow pulses in the millisecond range. Compare the full line-up in our best beginner oscilloscopes roundup.
Scopes in our reviews
50MHz, 4 channels, 1 GSa/s, with a large 7-inch display. Our best overall pick.
100MHz, 4 channels, 1 GSa/s, and a well-regarded, intuitive interface.
100MHz, 2 channels, 1 GSa/s, plus a built-in waveform generator in a compact unit.
As an Amazon Associate I earn from qualifying purchases. Prices and availability are shown on Amazon.
Frequently asked questions
Why is pulse width measured at 50% of the amplitude?
Real edges take time to rise and fall, so the base and the top of a pulse are not sharp corners. Measuring at the halfway point gives one repeatable answer that sits in the steepest part of each edge, where noise moves the crossing least. It is the convention most scope measurement menus and datasheets use, so your number will match theirs.
What is the difference between +width and −width?
+Width is the time a pulse spends above the 50% level, from a rising crossing to the next falling crossing. −Width is the time spent below it, from a falling crossing to the next rising one. For an active-low signal, such as many reset and chip-select lines, the −width is the pulse you actually care about.
Why does my scope show the pulse width as invalid or with a question mark?
Usually the scope cannot find both edges on screen. Either the timebase is so slow that the pulse is only a few samples wide, or so fast that one edge is off screen. The vertical scale also matters: if the pulse is clipped at the top of the screen, the scope cannot find its true top level. Adjust both until the whole pulse is visible.
How narrow a pulse can my oscilloscope measure?
As a rule of thumb, a pulse should be at least three to five times wider than the scope's own rise time, which is roughly 0.35 divided by its bandwidth. A 50MHz scope has a rise time of about 7 ns, so pulses narrower than a few tens of nanoseconds will look rounded and lower in amplitude, and their width reading becomes unreliable.
Can a multimeter measure pulse width?
Not directly. Some multimeters show frequency and duty cycle, and for a steady, repeating signal you can work out the high time as duty cycle divided by frequency. That gives only an average, though. It cannot measure a single pulse, show a pulse that varies from one to the next, or reveal slow edges and ringing that distort the timing.
