Analysis of Impact Sensitivity Experiments

Variations in procedure revolve principally around the number of drops, required to make a Anlid determination of the sensitivity, and the sequence in which the successive heights are varied. The sequence of height selection depends on the subsequent analysis of the results and the level of probability at which the test is being conducted. The two most common sequences are the up-and-down and the run-down methods. 

The up-and-down method, also called the Bruceton or staircase method, is a widely used statistical procedure for the determination of the height at which 50% of the samples detonate or react and its standard deviation [17,18]. The advantages and disadvantages of the method are described elsewhere [18-20] basically the test gives a reliable 50% value but not a reliable standard deviation. The number of tests required at each height also has been investigated [21]. 

Median 50% height of sample Median 50% height of standard 

The standard normally employed is a specially prepared RDX having an assigned FI of 80. 

The run-down method was developed at Frankford Arsenal, Philadelphia, and produces a more complete frequency and probability of reaction curve [22]. Although a larger number of tests is used than in the up-and-down procedure, the method makes possible a better evaluation of the distribution of the population. Starting at a level expected to be between 0% and 100% reactions levels, 20-25 tests are made at each of several levels above and below the starting level until the 0% and 100% levels are reached. The increments between the levels normally are equal to or less than the expected standard deviation. A cumulative probability curve is then plotted from the results of the entire test which is considered to be the frequency distribution of the population. 

---