Downloaded from https://www.everyspec.com AMCP 7•09.1 The cumulative hazard H is related to the that the same general plotting technique is Sf by the equations wed as for the K-S Bounds method. These instructions are also given on the Instructions Sf {x} = exp I- H(x)] (2-51) side of the Practical Probability Paper. H(x) I - In (Sx{)}] - - In (1 - Cdf {x}] Plot failure r at the two points (2-5b) FPH - 1 - exp (-- H) (2-G.60 So a plotting scale can be calculated for any FLO 1 1 - exp - HM.r.) (2.-6b) probability paper by using Eq. 2-5b. Special paper can be drawn for more convenient to convert the sample cumulative hazard H, hazard plotting and is mentioned in Ref. 23; to the Cdf.
Connect the points with hor- but It is not at all necessary. zontal and vertical lines; this Is the sample Cdf. Calculate and plot the K-S bounds as Even if no rigorous method is available for in par. 2-2.1.1 and Eq 2-2. The s-confldence estimating the uncertainty, it is desirable to bounds will not be exact at all. pet some idea about it. The procedure that follows provides grossly-approximate K-S Table 2-2 shows some failure data on field bounds (par. 2-2.1.1). It has the advantage windings of electric generators.
The hazard TABLE 2.2 CUMULATIVE4HAZARD CALCULATIONS FOR FIELD WINDINGS OF SOME ELECTRIC GENERATORS The table lists time.to-failure (months) for failed units and time.to-ond-of-test (monthu) for unfilled units. All times are listed in order of increasing time. Rank C vdaed event Reverse Harsed Cumulive" r time rank increment AH hoard Hr 1 31.7 F 16 0.0625 0.0=25 2 39.2 F 15 0.0687 0.1292 3 57.5 F 14 0.0714 0.201 4 65.0 13 5 C5.8 F 12 0,0533 0.284 6 70.0 F 11 0009 0.375 7 75.0 10 a 75.0 9 9 87.5 8 2 10 88.3 7 11 94.2 6 12 101.7 5 , 13 105.8 F 4 0.250 0.625 . 14 109.2 3 15 110.0 F 2 0.500 1.12 is 1 3 0 .0 1 dV F Indicates failure.
Other times are censorings (removed from test before failure, for a rason not connoected with thi state of the item). .1z