Downloaded from https://www.everyspec.com MIL-STD-81OD 19 July 1983 (a) Measured data ~. The shock reaponae spectrum required for the test will be determined from reduction of the environmental data .toobtain either maximum abaolute acceleration spectra or equivalent static acceleration spectra. The spectra will be a composite of spectra for positive and negative directions, sometimes called maximax spectra. The analyses will be performed for Q . 10 at a sequence of natural frequencies at inte–rvalaof 1/6 octave or smaller to span at least 5 to 2,000 Hz.
When a auffioient number of representative shock spectraare available?,an appropriate statistical basis should be employed to determine the required test spectrum. Nonparametric statistical techniques are covered in reference e. Parametric atatiatics can be employed if the data can be sbowfito satisfactorily fit an assumed underlying probability distribution. (For example, in MIL-STO-1540 (USAF) the test levels are baaed upon a maximum predicted environment defined to be equal to or greater than the 95th percentile value at leaat 50 percent of the time.
When a normal or lognormal distribution can be justified,,,appendix 516.3A provides a method for estimating such a test level.) When insufficient data are available for statistical analysis, an increaae over the maximum of the available spectral data should be used to establiah the required test.spectrum to account for variability of tbe environment and uncertainty in any predictive methods employed. The degree of increase is based upon.engineering judgment anh ,shouldbe supported by a rationale.for that judgment. Effective transient durationa, TE, will alao be determined from the time histories of the environmental data analyzed.
The effective duration is defined to be the minimum length of time which containa all data magnitudes exceeding 1/3 of the peak magnitude associated with the shock event. Figure 516.3-3 illuatratea this process. The shock input time history used for the response spectrum analysis will be 2TE in duration starting at a time to include the most significant data prior to and following the effective duration TE. An average of environmental shock duration, TE, will be determined and included in the specification of the required teat shock. In general, each individual axia of three orthogonal axes can have a different shock test spectrum and average effective duration.
(b) Measured data nOt a vailable. If a data base ia not available, the applicable spectrum from figure 516.3-1 can be emplOyed as the.. teat spectrum for each axis, provided that the effective (above 1/3 its peak value) duration of the test shock time history falls between six and nine milliseconds. (This spectrum approximates that of the perfect terminal sawtooth pulse shown in figure 516.3-4.) Any test transient is suitable if it eQualS or exceeds this sDectrum requirement over the freauencv range of 5 to 2000 Hz (for example, a ~alf-sine pulse applied in both direc~ions-of each axis).
As stated in I-3.3b, this shock testing may be omitted if random vibration testing is required with input acceleration spectral density in excess of the applicable spectrum in figure 516.3-2. A leas desirable alternative, only permissible if the teater has no shock response spectrum analsWis caDabilitv. iS to emDlOv a terminal Desk sawtooth vulse that meets the tole;ancea Riven i;’fiuure 516.+4 with aDDli&ation in both”directions alon!a each axis. ‘The presen;e of superimposed ;~pple falling withitithe waveform- tolerance band should be minim;zed to avoid significant overtesting at-the frequency of such ripple. METHOD 5163 19 July 1983 516.3-3