Measure hardware-trigger delay and jitter

What the test resolves

Measure trigger-to-exposure timing independently of image delivery. Host receipt time includes readout, transfer and scheduling and is not the exposure-start time.

This is a vendor-neutral commissioning method. Check the exact camera, lens, sensor or controller manual before changing settings. The numerical worksheet is hypothetical, not a Deyi Vision measurement or a product specification.

Operating prerequisites

The camera must expose a documented timing output or an optically verified alternative. Check I/O voltage, polarity and output propagation delay; a timing output is a proxy, not a perfect sensor timestamp.

Separate trigger acceptance, exposure-start delay and the time when an image reaches software. This experiment observes electrical edges and, where documented, a timing-output proxy for exposure. It cannot directly establish sensor timing unless the manual specifies that relationship. Use a trigger rate the camera can accept, and exclude busy-camera events from a nominal latency distribution only if they are separately counted and reported as failed acquisitions.

Bench and records

Use an oscilloscope with suitable probes, a safe pulse source and camera timing documentation. Save trigger mode, debounce, delay, ROI, exposure and the definition of each measured edge.

Check input voltage, polarity, electrical reference and output loading before connecting an oscilloscope. Select a documented exposure-related output when the model supports it; otherwise report that the proposed electrical proxy is unavailable. Record probe points, cable routing, scope threshold, sample capture settings, trigger activation, debounce and any intentional delay. Save a configuration export so changes to those settings can be matched to changes in the measured distribution.

Run the test

1. Configure a supported input edge and timing output using the exact model manual.
2. Probe the signal at the camera input and its exposure output; compensate known probe skew.
3. Acquire 1000 paired edges at low rate, then repeat at maximum intended trigger rate.
4. Calculate mean, minimum, maximum and missing events; repeat after restart and under host load.

Capture the input and output on the same time base. Mark the chosen active edge on each trace and calculate their difference with the same edge threshold throughout the run. Keep representative traces for the minimum, maximum and unusual values, not only an averaged screenshot. Repeat with the expected controller and wiring after a bench generator test; differences between the two configurations reveal contributions that a camera-only test would miss.

Hypothetical worksheet

Hypothetical output-edge delays of 39–47 µs average 42 µs. Jitter is 8 µs peak to peak. At 500 mm/s the largest delay corresponds to 23.5 µm travel; output propagation must be considered before calling this sensor latency.

L_i = exposure_output_edge_i - trigger_input_edge_i
mean(L) = 42 us; min(L) = 39 us; max(L) = 47 us
peak_to_peak_jitter = 47 - 39 = 8 us

For the hypothetical distribution, the peak-to-peak value describes observed spread, while the maximum delay estimates travel before the output proxy changes state. Neither is a guaranteed worst-case specification. A constant delay may be compensable in a position model, but compensation does not remove jitter. Record the reference point from which position is calculated and verify any compensation independently using a moving target, keeping the original uncompensated measurements in the report.

Test sequence

Interpret the evidence

Interpret the timeline in order: controller edge, camera input acceptance, exposure event and observable output. An output propagation delay can shift the measured mean even when sensor timing is unchanged. Compare like-for-like probe locations and output functions. If timing documentation does not quantify an intermediate stage, leave it as an unresolved contribution; do not subtract an assumed value to create an apparently precise sensor latency number.

Troubleshooting

• Extra fixed delay may come from configured trigger delay or debounce.
• Sporadic missing events at high rate can indicate triggering while the camera is not ready; reducing debounce blindly can introduce false triggers.

Missing output pulses require checking trigger-ready or busy behavior before computing latency statistics. A second cluster of delays can indicate a controller sequence, debounce interaction or measurement threshold problem; inspect the corresponding traces. If changing the probe threshold substantially changes measured delay, investigate signal rise time and noise. When testing illumination as well, use a photodiode in a separate synchronized check because the lamp's optical response is not an electrical camera-output edge.

Acceptance and handoff

Set limits for both worst delay and jitter from allowable position error. Require expected event counts and preserve scope traces, output-delay assumptions and settings; report proxy timing explicitly.

Archive the raw files, settings, sample identifiers and decision limits together. Discuss the documented requirements with Deyi Vision using the related product and preparation guides below.

Accept against limits defined for the observed quantity: trigger-to-output delay, acquisition success and position error have separate criteria. Run fresh paired captures with production wiring, controller logic and the highest intended acquisition demand. Document unaccepted triggers and gaps alongside the timing distribution. The handoff must name the measured output, explain its relationship to exposure and state the conditions under which any positional compensation is valid.