mp-ai-20/mechanical-engineering-skills/accelerated-life-testing

Claude Code and agent skill by mp-ai-20/mechanical-engineering-skills

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Accelerated life testing (ALT) — Arrhenius/Coffin-Manson/power-law models, acceleration factors, HALT/HASS, Weibull analysis, test planning, MIL-HDBK-781, IEC 60068.

Install

npx vskill@latest install mp-ai-20/mechanical-engineering-skills/accelerated-life-testing

SKILL.md

The same SKILL.md was published earlier as mp-ai-20/mechanical-engineering-skills/accelerated-life-testing.

Accelerated Life Testing (ALT) — Complete Skill

Purpose and Approach

Goal: estimate product life at use conditions by testing at elevated stress (T, voltage, humidity, vibration, cycling) to precipitate failures faster Key assumption: failure mechanism is the same at accelerated and use conditions (no new mechanisms introduced) Output: acceleration factor AF → extrapolate test life to field life

Acceleration Models

Arrhenius Model (Temperature-Driven)

Applies to: chemical degradation, oxidation, corrosion, polymer aging, semiconductor wear-out

Failure rate at temperature T: λ(T) = A × exp(-E_a / (k_B × T))

Acceleration factor (T_use vs. T_test): AF = exp[(E_a / k_B) × (1/T_use - 1/T_test)] [T in Kelvin; k_B = 8.617 × 10⁻⁵ eV/K]

Activation energy E_a:

Failure mechanismE_a [eV]
Semiconductor electromigration0.7–0.9
Corrosion (general)0.5–0.7
Polymer oxidation0.8–1.2
Solder joint fatigue0.3–0.5
General electronics0.7 (JEDEC standard)

Example: E_a = 0.7 eV; T_use = 40°C (313 K); T_test = 125°C (398 K) AF = exp[(0.7/8.617×10⁻⁵) × (1/313 - 1/398)] = exp[8121 × 6.83×10⁻⁴] = exp[5.55] ≈ 256 → 1000 hr at 125°C ≈ 256,000 hr at 40°C field use

Coffin-Manson Model (Thermal Cycling)

Applies to: solder joint fatigue, thermal fatigue of metals

Cycles to failure ratio: AF = (ΔT_test / ΔT_use)^m × (f_test / f_use)^n

ΔT = temperature cycle range [°C]; f = cycling frequency [cyc/hr] m = 1.9–2.5 (solder); n ≈ 0.33 (frequency exponent; minor factor)

Modified Coffin-Manson (JEDEC JESD22-A104): AF = (ΔT_acc / ΔT_use)^m × exp[E_a/k_B × (1/T_use_max - 1/T_acc_max)] Combined temperature range + Arrhenius for time-dependent mechanisms

Power Law (Voltage/Electrical Stress)

Applies to: dielectric breakdown, electrostatic discharge wear-out

AF = (V_test / V_use)^n n = 2–4 (oxide wear-out); n = 8–12 (ESD sensitive)

Example: V_use = 3.3 V; V_test = 6.6 V; n = 3 AF = (6.6/3.3)³ = 2³ = 8

Inverse Power Law (Mechanical Fatigue, Vibration)

Applies to: vibration-induced fatigue, wear

N₁ × S₁ᵐ = N₂ × S₂ᵐ (Basquin) AF = (S_test / S_use)^m [S = stress amplitude; m = S-N slope; typically 3–10]

Humidity-Temperature (Peck Model)

AF = (RH_test / RH_use)^n × exp[E_a/k_B × (1/T_use - 1/T_test)] n ≈ 2.7–3.0 (humidity exponent; empirical from corrosion data) Combined Arrhenius + humidity power law

JEDEC JESD22-A101 (moisture sensitivity): 85°C/85% RH standard test; 168 hr or 240 hr depending on MSL

Combined Stress

Multiple stresses: multiply individual AFs (if mechanisms independent): AF_total = AF_temp × AF_voltage × AF_humidity

Weibull Analysis for ALT Data

Weibull life distribution: F(t) = 1 - exp[-(t/η)^β] β = shape parameter (< 1: infant mortality; = 1: random; > 1: wear-out) η = characteristic life (63.2% cumulative failures) [hours]

Accelerated analysis: η(S) = η₀ × exp[E_a/(k_B × T)] (Arrhenius-Weibull model) Fit using Maximum Likelihood Estimation (MLE) or least squares

Software: ReliaSoft ALTA, Minitab, JMP, R (ALTmodelR package)

Test Planning

Number of test units: n ≥ [-ln(1 - C)] / [1 - (1 - p)^(1)] per stress level Simplified: n ≥ 10–30 per stress level for adequate Weibull fit

Stress levels: Minimum 3 levels to fit acceleration model Level 1: T_use × 1.5; Level 2: T_use × 2.0; Level 3: T_use × 2.5 Upper bound: never exceed material capability (new failure modes)

Test duration: t_test ≈ t_requirement / AF [hr; AF from acceleration model]

Confidence bound on life: η_lower (90% confidence, 1-sided) from MLE ± 1.28 σ_ln(η)

HALT (Highly Accelerated Life Test)

Purpose: discover design weaknesses, not measure life Protocol: step-stress (increase T, then vibration, then combined) until failure Environment: rapid thermal cycling ΔT rate 60°C/min; random vibration 2–100 g_rms

HALT process:

  1. Cold step-stress: -10, -30, -50, -70°C (step until fail)
  2. Hot step-stress: +50, +70, +90, +110°C
  3. Rapid thermal cycling: ΔT = 60, 90, 120°C range at 60°C/min
  4. Vibration step-stress: 5, 10, 20, 40 g_rms (6-DOF vibration)
  5. Combined: all stresses simultaneously

Operating limit (OL): stress where DUT fails to function (reversibly) Destruct limit (DL): stress where permanent damage occurs Design margin: (DL - OL) / use_condition_stress; want margin > 2

HASS (Highly Accelerated Stress Screening)

Production screen (100% of units): brief exposure to stresses within HALT limits Goal: precipitate infant-mortality failures before product reaches customer Protocol (product-specific): typically ±60°C thermal cycling + 5–10 g_rms vibration for 2–10 cycles

Standards

StandardScope
MIL-HDBK-781Reliability test plans; sample sizes
IEC 60068Environmental testing; temperature/humidity/vibration
JEDEC JESD47Stress-test-driven qualification of new ICs
JEDEC JESD22-A104Temperature cycling
JEDEC JESD22-A101Steady-state humidity/bias testing
MIL-STD-810HMilitary environmental engineering

Output

Provide: acceleration model type (Arrhenius/Coffin-Manson/power-law), activation energy E_a [eV] or model parameter m, acceleration factor AF at test conditions, test conditions (T [°C], voltage [V], humidity [%RH], vibration g_rms), test duration [hours], sample size per stress level, estimated field life at use conditions [hours or years] with 90% confidence lower bound, Weibull β and η parameters from fitted data, HALT operating and destruct limits, and applicable standard (IEC 60068, JEDEC JESD22, MIL-HDBK-781).

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/mp-ai-20/mechanical-engineering-skills/accelerated-life-testing/badge)](https://verified-skill.com/skills/mp-ai-20/mechanical-engineering-skills/accelerated-life-testing)

Something wrong with this skill or its page? Tell us.