Tools › Reliability › Growth and Testing › HALT and HASS Planning › Example
MIL-HDBK-344A screening strength: 4 thermal cycles at 5 °C/min over 100 °C and 5 minutes at 5 Grms with detection efficiency 0.855
| Tool | HALT and HASS Planning |
| Source | U.S. Department of Defense (1993). Environmental Stress Screening (ESS) of Electronic Equipment. MIL-HDBK-344A, 16 August 1993. Section 5.4.3 Procedure C (stress constants, detection-efficiency factors, Example of SS Calculation), Tables 5.14 and 5.15, Procedure A1 step 5, Appendix A equations A-1 to A-7, Procedure A4 fatigue-life example. |
| Level | Professional |
Problem
A production unit is to be screened by 4 thermal cycles between −25 and +75 °C at 5 °C/min followed by 5 minutes of random vibration at 5 Grms, with functional-and-parametric testing under stress (detection factors 0.9, 1.0, 0.95). Find the precipitation efficiency and screening strength of each screen, the defects remaining and the yield for an incoming density of 1.0 latent defect per unit split 80 % thermal / 20 % vibration, the remaining defects allowed by a 2,000-hour MTBF requirement (inherent 10,000 h, safety margin 1.5, 1,000-hour period, field precipitation rate 1/1000 per hour), the fatigue life consumed, and what finding 20 of 20 seeded defects demonstrates.
Inputs to Enter
- Thermal cycling: T high 75 °C, T low −25 °C, rate 5 °C/min, 4 cycles; random vibration 5 Grms, 5 minutes, axis factor 1.0
- Detection factors 0.9 (functional and parametric), 1.0 (under stress), 0.95 (isolation and removal); D in 1.0 per unit, 80 % thermal
- Required MTBF 2,000 h, inherent 10,000 h, safety margin 1.5, field period 1,000 h, field k 0.001 per hour, SAF 1.0; screen 100, rework 300, field failure 5,000 per event
- Life 7,300 cycles of 30 °C and 2 × 10⁶ minutes at 1 Grms (handbook exponents B = 2.5 and 6.4); 20 seeded defects, 20 found, 90 % confidence, target SS 0.90
Answers
- k(TC) = 0.0017 (100.6)^0.6 [ln(5 + e)]^3 = 0.2308 per cycle, PE(TC) = 1 − exp(−4 k) = 0.6027 (handbook .6027, Table 5.15); k(RV) = 0.0046 × 5^1.71 = 0.07211 per minute, PE(RV) = 0.3027 (handbook .303, Table 5.14)
- DE = 0.9 × 1.0 × 0.95 = 0.855; SS(TC) = 0.5153 (handbook 0.5153); SS(RV) = 0.2588 (the handbook prints 0.2, a truncation)
- Defect flow: thermal population 0.8 → 0.3877, vibration 0.2 → 0.1482; remaining 0.5360, removed 0.4640 (overall SS 46.4 %), yield exp(−0.464) = 62.9 %, P(shipped unit defective) 41.5 % against 63.2 % unscreened
- Allowable remaining defects: FR = (1/2000 − 1/10000)/1.5 = 2.667 × 10⁻⁴ per hour, SS_field = 1 − e⁻¹ = 0.6321, D_remaining,max = 0.4219, required SS 57.8 %: not met by the planned screens
- Safety of screen: one screen consumes 4 × 100^2.5 / (7300 × 30^2.5) = 1.11 % of the thermal life and 5 × 5^6.4 / 2 × 10⁶ = 7.44 % of the vibration life (the handbook's own example gives 21.9 % and 23.8 %); 1.67 equivalent minutes at 6 Grms (MIL-HDBK-2164A)
- Proof of screen: 20 of 20 found demonstrates SS > 0.1^(1/20) = 0.891 at 90 %; 22 seeded defects must all be found to show SS > 0.90
Notes
The tool's default inputs are this case; press Submit. The handbook's printed example gives only PE, DE and SS; the flow, requirement, cost, life and proof values are the tool's application of the handbook's equations to stated illustrative inputs (test_halt_hass.php, 172 checks, numpy/scipy script References/HALT-HASS/checks/halt_check.py). Tables 5.14 to 5.17 are reproduced by the equations to the printed digit; the Procedure A4 example (21.9 % / 23.8 %) and the Figure 5.2 flow nodes are checked in the test.