On one hot July day when there were plenty of billowing summer clouds around, Bob Baker and I went out in these two fighters. The concept of the test was to have one plane fly through the storm cloud at 18,000 feet and 185 mph. A second plane would stay outside of the test area and conditions and fly at the same altitude and airspeed. Both pilots would take records simultaneously. This would produce ambient air and actual rough air records to compare. This flight would simulate the conditions that the B-17 and P-5l group encountered on their return from Germany. Because I was in the lead, I found a suitable cloud and flew into it. I experienced instant and considerable roughness. The accelerometer on the instrument panel hit the limits of plus 12 and minus 4. But these loads were of short duration and similar to the loads I had experienced on the preliminary tests.
The flight continued through very rough air for some time. Suddenly, I noticed a hole appear in the cowling in front of me. It was about the diameter of a piston. Strangely, it didn’t seem significant, nor were there any effects for the moment. Then the oil pressure started to decrease, and the RPMs started to increase. There was a sharp jolt as the engine threw a propeller blade. A wave of yellow flame swept back over the top of the canopy. I had come out of the clouds at that point, and Baker was alongside me at about one-mile distance. Up to the time of the fire, I had thought I could land in a field.
When the fire broke out, Baker radioed me to bail out. I jettisoned the canopy, took off my helmet and seat belt, and started a roll to the right. I thought it would be easier to bail out if the plane was inverted and if I just fell out of it. I rolled the plane over and tried to stay out of a dive by holding the plane level. This caused the flames to go under the plane and away from me. And as the plane continued into the roll, I started to slide up the side of the cockpit while still holding onto the stick. As a result, I never got completely inverted before I started to slide out of the plane. I was hit in the face by the 180-mph air velocity and had to let go of the stick. I grabbed for my knees and went out between the vertical and horizontal tail surfaces. It was fortuitous that the plane was neither level nor at a ninety-degree angle, because I wouldn’t have passed between the tail surfaces as easily as I did. I can still see the vertical numbers on the tail as it went by.
The catastrophic failure that Cavallo encountered was not a gust load on the wings as initially assumed, but the aircraft’s engine and the engine mounts that were failing.
The P-51 test flight was just one of many incidents where Cavallo faced violent buffeting inside an aircraft’s cockpit. While at Langley, he recognized the need for skull protection because of all the battering he encountered and designed and made a helmet to safeguard himself during flight tests. The helmet is probably the first to incorporate skull protection, optic shielding, voice communications, and an oxygen system. Although it was not adopted for issue, it may have influenced other flight helmets developed for jet pilots by the military services after World War II. The helmet was painted white and had Cavallo’s name stenciled on the front brow. He included five gold stars in the design to represent each borough of his hometown, New York City. According to Cavallo, these stars surprised the military ground crew personnel he encountered when landing at numerous Army Air Force bases during the war.
The events surrounding Cavallo’s wartime service with the NACA were critical to the U.S. Army Air Forces and Navy's successful operations during World War II. The flight tests conducted by this small group of test pilots helped improve American combat aircraft design's performance and capabilities. The testing was critically important for the war effort and helped save the lives of those serving in combat.