eee: zontal tail loads made by two different and completely independent methods, Frior to IVY, the cesirn limit upload on the horizontal tail of the 636 aircraft was ruolished as 33,.00 lb. Further analysis and re- calenlation by Sorsolidated Vultee Aircraft Corporation (CVAC) after IVY produced a revised desiom limit load of 63,000 lb. This higher fig- ure was later confirmed by static tests conducted by CVAC under contract to WADC. The tests were not -:ompleted until aftet Shot 9; therefore, the 36 aircrart was positioned on the lower allowable tail load for this shot, as well as the two IVY snots. In this report, however, all coi varisons of mecsured loads with desiim limit load are made with res- pect to the hisher fisure of 63,000 lt. On this busis the peak measured loud «as only 45 per cent of desi.n limit loud; however, if calculated on the same tusis used to position the aircraft, the peak measured load would have approacned more closely the desivn limit. In 41] exrosures tne aircraft were positioned with the tail toward the e«plosion; therefore, symmetrical ioadins: on the wing and horizontal tail was ex;ected. TMA pao ath ~~ wIEY ai -3 hye’, rendinr meacurements on the rirkt and left wing at station 390 were ecudvailent for Mike and Kine Shots cut differed greatly in Shot 9. Althoush no direct evidence hus veen found that would invalidate either of the ~eacurerents, Sor reasons given oelow it is velieved the lower value, which wis thet meacured on the left wing during Shot 9, is incorrect, und aujmmetrical loadin, 13 not indicated, To Jetermine whether or not mecsured values were in approximate ayreement relutive to each other, the maximum positive bending moments meaourped at eacn instrumented station for each shot, except for the sta-— tion farthest outooard, were plotted as a function of the distance from the aircraft center line. The resuiting curves drawn Zor each shot are shosn in Fig, 4.1. The r«lationship of tendi:- moment versus span cal culated for the condition of uniform load snows it is ceasonable to exreet a plot of peak values to ;roduce a curve of the reneral shape shown, 1.4., tigher loads at the inbourd stations, decreasing with increasing span. .. dynamic analysis would be required ta determine the exact curve at any given time. The aveve method of cumrurison provides a good check on the validity of the test, duta, The peu: cending moments from station 106° have not veen plotted because the influence of the higher vibration modes caused the peak value to ve reached at a much earlier time than for the instrumented stations inboard of «tation 1062, The response curves of the suspect measurements made at station 2990 are practically identical. except that the left wing meacurement consistently equals one- half the right wing measurement. This low reading can easily be explain- ed as an instrumentation failure; however, «cxeludiniy data reduction er- rors, it is almost impossiole for an instrumentation failure to cause a hivh ceading. The peak value of these Shoat 4 measurements have both been rlotted in Fig, 6.1. The curves presis..cd were iIrewn c th? basis of the composite data, excludins the point: in question, and show that the nhicsher value is in the revion predicted ty the curve, whereas the lower value falls consicerably celow the curve. If the Shot 9 curve vl