Page:CAB Accident Report, TWA Flight 891.pdf/57

This page needs to be proofread.

-57-

-57-

(b!) As mentioned in pars. ll.h.h and ll.h.18, the pOSSibility that electrical discharges might ignite flammable gasoline vapors issuing from the vent out- lets has already been studied in the United States, mth peeitive results.

The tests mentioned in par. 11.11.15 were made in the tunnel, as follows:

On only one of the original outlets, placed on the trailing edge of an @011, from whose four outlet pipes (See par. 114.0) issued vapors containing a mixture that was within the limits of flammability in the case of tanks Nos. 7 and 6, and not within those limits in the case of tanks 3-1.1;


at a pressure corresponding to an altitude of 1,700 ft.; at an air flow speed of 170 knots I.A.S.;

for an outgomg speed of the vapors, for each indiv1dual outlet pipes, corresponding to climbing speeds of 900, 600 and zero feet per minute.

They disclosed that in the presence of electrical discharges, said vapors become ignited only if the plane is climbing, and that the flames Will not spread to the interior of the tanks.

The tests mentioned in par. 11.14.18, conducted on the premises of a SpBClallzed institution, showed, specifically, that the weather conditions eXisting at the time of the crash were most appropriate for creating, on the vent outlets, electrical discharges fully capable of igniting flamable gasoline vapors.

(c') With regard to the possibility that, once the gasoline vapors had ignited at the vent outlets, the fire may have spread to tank No. 7 through the pipes, it is observed that the tunnel tests (see preceding paint ht ), during which such spreading did not take place:

[1] Did not reproduce the real vent outlet system of tank No. 7 (existence of two outlet pipes and, therefore, two vent outlets;

[2} Did not bring about the true conditions in which the plane must hare found itself at the time of the accident

In particular, the tests did not take into account the effects generated by the turbulence, by sudden variations in flight trim, etc.; such conditions, in