Setting Triggers & Timebase for Automotive Signals

Why capture settings matter more than the scope

An inexpensive scope set up correctly beats an expensive one set up wrong. Four settings decide whether a capture can be measured at all:

• Timebase — how much time the capture holds.
• Voltage range — how much amplitude fits before the signal clips.
• Trigger — which moment the capture is anchored to.
• Sample rate — how finely the signal is sliced in time.

A capture that's clipped, undersampled, or too short can't be fixed afterwards, by you or by any analyzer.

The two numbers behind every timebase

Screen time = time/div × number of horizontal divisions. Most scopes show 10 divisions, so 20 ms/div holds 200 ms.

Engine cycle time = 120 / RPM seconds. One four-stroke cycle is two crank revolutions (720°).

Every per-cylinder test needs at least one full 720° cycle on screen, because every cylinder fires once per cycle. Whole-engine analysis needs several.

Timebase quick reference

Timebases assume a 10-division screen.

Why whole-engine tests need 10 seconds

The analyzer measures complete sync-to-sync engine cycles and needs at least three of them to compare cylinders. At 700 RPM one cycle takes 171 ms, so a 200 ms capture can hold at most one complete cycle. Ten seconds holds about 58 cycles, which is enough to average out cycle-to-cycle variation and still have margin if a snap throttle is in the capture.

Voltage range — leave headroom

Screen range = volts/div × number of vertical divisions. On an 8-division screen, 5 V/div shows 40 V. On a ±range scope, the range setting is the limit.

If the trace touches the top or bottom of the range, the peak you see is the scope's limit, not the signal. Go up one range and capture again.

Trigger modes

Auto, Normal, Single

• Auto draws even without a valid trigger. Use it to find an unknown signal, not to analyze one.
• Normal redraws only when the trigger condition is met. Use it once you know the signal.
• Single captures one triggered event and stops. Use it with a glitch or pulse-width trigger for intermittents.

Edge trigger

Choose the cleanest, most repeatable transition, and set the level above the noise band:

• Port injector: falling edge of the driver pull-down.
• COP primary: the driver turn-off, or the rising edge of the primary current.
• Crank sensor: one edge of the tooth pattern, or the gap with a pulse-width trigger.
• High-speed CAN: rising edge of CAN-H (recessive to dominant, which is how every frame begins).
• Engine-cycle tests: a coil or injector signal on one known cylinder, so the capture starts at the same point in the cycle every time.

Pulse-width (glitch) trigger

Triggers only when a pulse is shorter or longer than a limit you set. Examples:

• "Low pulse longer than 5 ms" catches an injector held on too long.
• On a 60-2 crank wheel the normal gap already spans 3 tooth periods, so a plain "longer than" trigger fires every revolution. Use a pulse-width range instead: trigger on a period between 1.5 and 2.5 tooth periods. That catches a single dropped tooth and ignores the normal gap.

Trigger holdoff

After the scope triggers, holdoff blocks new triggers for a time you set. On a signal with many identical edges (crank teeth, CAN bits), set holdoff just under one cycle so every capture anchors to the same event instead of a random edge.

Sample rate and memory depth

Sample rate = samples captured ÷ capture time. When you stretch the timebase, a scope with limited memory lowers its sample rate to fit.

Example: 10,000 samples across a 5 s capture is 2 kS/s, one sample every 0.5 ms. That's enough for starter-current humps at about 8–11 Hz, but a 3 ms injector pulse would get only six samples, and its opening notch would be invisible.

Rules:

• Read the actual sample rate on screen after setting the timebase, and record it with the capture.
• Prioritize memory depth for long captures. 10 s at 100 kS/s needs 1 million samples per channel.
• Resolution-enhancement modes average adjacent samples. They reduce noise on slow sensors but also reduce bandwidth. Don't use them for ignition, injectors, or network signals.

Probing — where readings go wrong

• Know your scope's grounds. On most USB scopes every channel ground is common and tied to the laptop. A ground clip on anything other than ground is a short circuit through the scope.
• Reference what the module reads. A sensor signal is measured by the module against its sensor ground. Referencing battery negative adds any ground-path drop to your reading. To measure that drop, put a second channel on the sensor ground.
• Back-probe at the connector. Never pierce insulation; the hole corrodes and becomes the next intermittent fault.
• Keep leads short. A long clip-lead loop picks up ignition noise and shows ringing that isn't there.
• Current clamps: zero the clamp with no current flowing through the conductor (or before clamping on), then clamp around one conductor only.
• CAN: two channels plus a math channel (CAN-H − CAN-L) show the differential signal. A differential probe is not required.

A repeatable capture checklist

1. Verify the battery and charging system first.
2. Choose the probe and connection point; confirm the ground.
3. Set the voltage range so the expected peak fills about 60% of the range.
4. Set the timebase from the table above: one event, one cycle, or 10 seconds for whole-engine tests.
5. Set the trigger source, edge, and level; confirm it locks in Normal mode.
6. Read the actual sample rate and confirm it meets the minimum.
7. Capture, then check for clipping before you save.
8. Save the full capture with its settings, and record the operating state: RPM, temperature, load, and supply voltage.

Frequently asked questions

Related guides

• Automotive oscilloscope guide →
• Current clamps for engine diagnostics →
• Intermittent fault capture →
• Relative compression test →
• Battery & charging system health →