Tag

hardware-metrics

3 articles

An electronics workbench with an oscilloscope showing several waveforms, a multimeter, a circuit board under test with probe wires, and a board layout on a monitor.

Fault Injection: Proving Diagnostic Coverage Instead of Claiming It

300 injected faults, 300 detected, and the voltage supervisor still covers only about 80 % of the failure rate it was credited with. The 99 % in an FMEDA row usually comes from a handbook table. Fault injection tests the mechanism as built, and the result only means something when each failure mode is weighted by its share of the failure rate. New article, with a core supply supervisor and its deglitch filter as the worked example.

6 minNov 1, 2026

Close-up of a silicon wafer with a grid of iridescent chip dies, shallow depth of field.

Technical

When a Bit Flips: Soft Errors and Automotive Safety

A particle from space can change a value inside a working chip and leave nothing behind for a test to find. Soft errors are random hardware faults with no defect behind them. Testing cannot remove them, so ISO 26262 expects the design to handle them and the hardware metrics to count them. New article, with a crash threshold in an airbag control unit as the example.

6 minNov 1, 2026

AI-generated illustration of an automotive ECU on an electronics test bench, with oscilloscope probes, a schematic and an analysis screen.

Why FMEDA Matters for ISO 26262: Five Practical Reasons

Hardware can fail systematically, through errors made during development, or randomly, as parts wear or break in service. FMEDA addresses the random failures, connecting failure analysis with stronger evidence and better engineering decisions. Explore five reasons it matters for ISO 26262 projects, from revealing diagnostic gaps and examining supplier assumptions to guiding improvements and keeping the safety argument aligned with a changing design.

6 minSep 25, 2026

Keep learning

Go from article to full guide.

Each guide is a full course with diagrams and worked examples. Open one to see what it covers.

Browse all guides

HardwareHardware MetricsLearn how to calculate SPFM, LFM, and PMHF with correct failure rate data and diagnostic coverage values. Covers ASIL-dependent targets, data sources (IEC 61709, FIDES), common calculation pitfalls, and compliance strategies.Pro plan pricing Expert plan pricing16 chapters
Verification and assessmentFault Injection & Safety-Mechanism VerificationProve that safety mechanisms detect and react to faults within the Fault-Tolerant Time Interval, turning FMEDA Diagnostic Coverage claims into demonstrated evidence across hardware, software, simulation, and system-level ISO 26262 campaigns.Expert plan pricing10 chapters
HardwareSoft Errors & Transient FaultsA cosmic ray hits a memory cell and your ASIL D function computes with a wrong value. This concept walks from the physics of radiation-induced bit flips through single event effects, ECC and hardened design, all the way to Part 11 quantification.Expert plan pricing12 chapters