MIL-STD-690 Failure Rate Sampling Plans And ProceduresThis standard provides procedures for failure rate FR qualification, sampling plans for establishing and maintaining FR levels at selected confidence levels, and lot conformance inspection procedures associated with FR testing for the purpose of direct reference in appropriate military electronic parts established reliability ER specifications. Figures and tables throughout this standard are based on exponential distribution. Weibull distribution will be acceptable in certain components such as capacitors. Use of Weibull distribution for any component must be approved by the qualifying activity. This standard also provides guidance to specification writers in the use of this standard see appendix A and references material for users of ER parts.

Pages 1 - 2 - 3 - 4 - 5 - 6 - 7 - 8 - 9 - 10 - 11 - 12 - 13 - 14 - 15 - 16 - 17 - 18 - 19 - 20 - 21 - 22 - 23 - 24 - 25 - 26 - 27 - 28 - 29 - 30 - 31 - 32 - 33 - 34 - 35 - 36 - 37 - 38 - 39 - 40 - 41 - 42 - 43 - 44 - 45

MIL-STD-690D

w/ CHANGE 1

4. GENERAL REQUIREMENTS

4.1 FR levels.

a. FR levels are related to operation of the part at the stress level specified by their applicable ER specification. b. Provisions are made for FR levels ranging from 1.0 to 0.001 percent per 1,000 component part hours. In the

event the existing FR level of a current product is higher than 1.0 percent, an additional level (level "L") shall

be added which will represent the state-of-the-art for the part.

c. Although the failure is expressed in percent per 1,000 hours (%/1,000 hr) throughout this standard, sampling plans and statistical tables may be used for FR levels expressed either in terms of percent per 1,000 cycles, operations, or in terms of duty cycle and stress level.

d. Where a FR level is required for periods other than 1,000 hours, an appropriate conversion factor may be applied.

Example: 1%/10,000 hours are equivalent to 0.1%/1,000 hours, or when specifying %/10,000 hours, the unit-hour requirement is to be multiplied by 10.

e. Tables are provided to show the relationship between true failure rates and selected confidence levels.

4.1.1 FR level determination. Determination of FR levels shall be based upon data from all FR tests. Data shall be accumulated from:

a. The qualification FR sample. The specification designates the number of sample units to be inspected, number of permissible failures, duration of FR test, and other criteria that may apply.

b. Inspection lots which have been submitted for FR conformance inspection during any qualification period or interval. Data accumulated shall meet the specification requirement referenced in 4.1.1a.

c. Samples subjected to extended FR tests. The data shall be added at the specified time of measurement and not at the end of the FR test.

4.1.2 Qualification approval for higher FR levels. Qualification approval granted on one of the lower FR levels shall include approval for all of the higher established FR levels. For example: Qualification approval for level "R" shall include approval for levels "P", "M", and "L", (if designated in the ER specification).

4.1.3 Supplying to higher ER levels. Parts qualified and marked (color coded or part numbered) to lower failure rate levels are substitutable for higher failure rate level parts with acquiring activity approval. A manufacturer may supply to all higher FR levels than that to which they are qualified, and may elect to use the sample size associated with the FR level to be supplied on orders or contracts. Election by the manufacturer to apply this option does not negate the requirement to maintain qualification in accordance with procedure III (see 5.3).

4.1.4 FR marking. All parts shall be marked with the FR level to which they are qualified, except when the contract or purchase order specifies higher FR marking under the substitution criteria of 4.1.3.