i = incident angle: the angle between the source-target line and a line normal to the skin surface L = heat-loss factor p = density, lb/ft Cp = specific heat, Btu/lb-F t = skin thickness, feet Similar relationships were established for the B-47 tests. In addition to the theoretical calculations above, thermal effects on certain critical panels were determined by experimental furnace testing. The limiting thermal response for the B-36 was a 400 F rise in the 0.020-inch magnesium hat panels of the elevator. For the B-47, the critical thermal response was a 370 F rise in the 0.020-inch aluminum skin of the ailerons. The characteristic: of the blast wave in free air include a sharp rise to its peak posi- tive pressure (the shock front), followed by a relatively slow decrease through the initial ambient value to a minimum of approximately a third of the peak positive value and a slow return to initial ambient pressure. The difference between the peak-positivetransient and initial-ambient values is the overpressure. For the B-36 in Castle, this was expressed empirically as: 1/3 ‘ AP = 31.3 V— [18 (<3) -0.88 1/2 R wis (6.3) Where: AP = peak overpressure, psi W =yield, ibs TNT equivalent R = Slant range, ft (mm)’” Pb 4b p = air density, slugs/ft® a = speed of sound, ft/sec h = altitude of the measurement b = burst altitude Equation 6.3 was used only for overpressures less than 2 psi. Both equations 6.1 and 6.3 were derived from limited test data from previous operations. The second important property of the blast wave is the material, or gust, velocity — the air movement behind the shock front. The equation used to predict material velocity was: u=1.89a, AP —— “2B, ( ap \~1/2 {|7+6— Ph ) Where: u = material velocity, ft/sec %;, = speed of sound at measurementaltitude, ft/sec AP = peak overpressure, pai Py = initial ambient pressure at measurement altitude, psi 14 (6.4)

Select target paragraph3