Systems / 04

Electronics & data

Sensors, acquisition and validation so that every trace can be trusted, aligned and explained.

Illustrative · simulated demo dataOUTPUT: validated channels at stated sample rates (illustrative)

Signal path and channels

Signal path · channel “Wheel speed”

01 SENSOR
Wheel speed · km/h
→
02 DATA LOGGER
500 Hz · 2 ms period
→
03 VALIDATION
No samples flagged
→
04 ENGINEER
Raw + validated + log

Hall or magnetic-encoder pickup on a toothed ring; pulse timing converted to speed.

Select sensor

Clock offset, logger B0 ms

At 100 km/h, 0 ms ≈ 0 mm of travel.

Inject a connector fault (spike + dropout)

Validation report

Range check: 0
Rate-of-change check: 0
Flat-line dropout check: 0
Timestamps monotonic: yes

Flagged samples are logged, not deleted.

Simulated · generated locally500 samples / 1 s window

Channel: Wheel speed

Logger ALogger B (+0 ms)

Hover, tap or focus the chart and use ← → to read values.

Approach

Every claim in the other three disciplines rests on a signal. We specify sensors, sample rates and clocks, then pass each channel through checks for range, rate of change and timestamp consistency before an engineer sees it.

Methodology

Choose sample rates from the bandwidth of the phenomenon, with anti-alias filtering before sampling.
Timestamp as close to the sensor as the hardware allows, and record how clocks are synchronised.
Validate in a pipeline: range, rate, dropout and cross-channel plausibility, with every rejection logged.

Deliverables

Channel list with rates, ranges and filters
Clock and synchronisation design
Validation pipeline with event log

Scope and constraints

Bus arbitration and logger buffering add jitter to arrival times.
Filters delay signals; the delay differs per channel.
More channels mean more bandwidth, more wiring and more things that can fail.

Next: 01 Aerodynamics →