snow that continued to fall as we stepped out the hotel door in the middle of the night. We looked at each other with a sinking feeling as our expectation for good weather evaporated. We had driven halfway down the mountain before the snow turned to rain and fog. During the past two months, determining the weather while driving down the road had evolved into a “fighter-pilot science.” Noting the elevation on the mountain at which we could see Aviano AB gave us a better estimate of ceiling and visibility than the weatherman could. Today seemed to be the worst yet—since we didn’t see the base until we drove through the gate.
We went through the standard preflight planning and briefings hoping that conditions would improve. I was scheduled to fly with Buster, a squadron flight commander and a no-nonsense pilot who had total concentration on the job at hand. He was especially focused this morning since the latest weather brief said the weather in Kosovo was breaking up. We were briefed and ready to go an hour before sunrise, but the weather still had not lifted. Low ceilings forced the whole package at Aviano to sit on the ground and remain on standby. The weather was the same down the whole length of the Italian coast—everyone was on standby.
About 30 minutes after our scheduled takeoff time, the weather finally improved to the required 500-foot ceiling and one-mile visibility; we then got approval to launch. Taking off in poor weather puts additional demands on Hog drivers. Most fighters use their air-to-air radar to maintain positive separation between members of their formation when taking off and flying into the weather. Since we did not have air-to-air radar, we had to fly with our instruments and perform a procedural trail departure to ensure that we had safe separation between our aircraft. We took off 20 seconds apart, flew set airspeeds, maintained the same ground track, and climbed at a specific power setting. We kept the variables constant and relied on the differences between our departure times to keep us safely separated during the climb out. As Two, I waited 20 seconds after Buster started his roll before I released my brakes. I watched Buster disappear into the weather, and 20 seconds later so did I. If the instrument departure were executed properly, I would break out on top of the clouds with Buster slightly above and about a mile and a half in front of me—the distance he would cover in 20 seconds at climb speed. That was the theory but not what happened. I did not realize it at the time, but the moment Buster entered the weather would be the last I would see of him until we were both back on the ground at Aviano—almost two hours later.
Our two-ship was the third A-10 flight to launch that morning. We monitored the common VHF radio frequency and could hear the members of other flights describe the weather they were experiencing. Those descriptions became the best forecast of what lay ahead of us. The weather had become a real problem. Passing 6,000 feet, I noticed that ice had formed on the nose of the Maverick missiles, the rocket pods, and the leading edge of the wings. I told Buster, and he acknowledged having the same problem. We continued to climb in an attempt to find an altitude where the icing would stop, but it only seemed to get worse until we passed 17,000 feet. The A-10 flights ahead of us had also reported icing, and all aircraft were still in the weather when they leveled off at 25,000 feet, our final altitude for the track down south. Fortunately, a flight of two F-16s, which had taken off behind us and had climbed steeper and flown above us on departure, had just broken out at 30,000 feet and said that the weather above was clear. We decided to continue our climb to find clear weather and help the ice sublime. Unfortunately, our Hogs were loaded for combat with four Mk-82 bombs, two Maverick missiles, two rocket pods, an ALQ-131 jamming pod, and two AIM-9 missiles. That load made us extremely heavy, increased our drag, and precluded a quick climb.
We remained in the weather during our climb and flight down the Adriatic. After about 45 minutes, we finally reached 30,000 feet and could see sunlight above us. Leveling off approximately two miles in trail behind Buster, I was still in the weather and unable to see him. While doing an ops check to see how much fuel was remaining, I saw the master caution light begin to flash and looked to investigate. The right-generator caution light was illuminated, and the number-two (right) engine tachometer was wildly fluctuating between 30 and 90 percent. At this point, I was flying the aircraft about five degrees nose high to maintain level flight at 170 knots indicated airspeed (KIAS).
I immediately radioed Buster to let him know I had a problem and pulled the throttle back on number two in an attempt to recover the engine. When I moved the throttle, the engine immediately spooled back and flamed out. The right engine nacelle now generated drag rather than thrust. That and the combination of high altitude, low airspeed, and a dirty configuration caused the aircraft to yaw right and begin a descent. I immediately pushed the nose five to 10 degrees nose low, attempting to gain airspeed. As the aircraft yawed to the right, a chopped tone came over the headset indicating a stalled condition, which was immediately followed by a loud pop and buzzing sound as the number-one (left) engine compressor stalled. Things were getting serious in a hurry. I tried to stay calm and inform Buster what was happening.
While making the radio call, I saw the main attitude direction indicator (ADI) freeze in the centered position, both steering bars came into view, and all my caution-warning lights illuminated—and the bottom dropped out of my stomach. Those indications told me that I had lost all alternating current (AC) electrical power and that the aircraft had just reverted to direct current (DC) battery power. Not wanting to believe what I saw, I looked at the number- one engine as it rolled back below 30 percent, approximately 12 percent below what was needed for the generators to provide the much-needed AC electrical power. With virtually no thrust, the aircraft began to descend rapidly into the weather, and I soon lost what little sunlight I had been able to see through the clouds. I knew that the only way to clear a compressor stall was to completely shut down the engine, but the idea was still not one I wanted to entertain.
I then checked the DC-powered standby ADI, which showed the aircraft 15 degrees nose high and 20 degrees of right bank—the same attitude it indicated when I had leveled off. My confidence in this old piece of equipment was shot. With the main attitude indicator frozen, the standby attitude indicator unreliable, and no visual references due to the weather, I was forced to use the “needle and ball” of the turn-and-slip indicator to keep the aircraft in coordinated flight. My voice jumped about 10 octaves as I tried to tell Buster that I had just experienced a “double-engine flameout.”
This situation calls for a “boldface emergency procedure.” I had long ago been required to commit to memory the steps I now needed to take—but I had never dreamed of actually using them. My initial A-10 training instructor had even made jokes about this emergency, swearing it could never happen in the A-10 because the engines were too reliable. Some guys in the squadron even joked in our monthly emergency-procedures training that if it ever happened they would just jump out using the ejection procedure. With that in mind, I had to decide and act in a hurry.
The five boldface steps that I had long before committed to memory:
1. THROTTLES – OFF
2. APU [auxiliary power unit] – START
3. FLIGHT CONTROLS – MAN [manual] REVERSION
4. LEFT ENGINE – MOTOR
5. LEFT ENGINE – START
Thinking that I still had a little time before needing to make an ejection decision, I started to rapidly repeat these steps in my mind. We had decided to fly without antiexposure suits because they would have made our eight-hour missions miserable. However, now the thought of ejecting at high altitude, in the weather, over the Adriatic, and without an antiexposure suit was not very appealing either. I knew that I had to attempt the restart—I did not know that the procedure, which had not yet been successfully used, was intended for use in good weather and at lower altitudes.
Passing 29,000 feet, I executed the first step, pulling the throttles back, forcing them over the hump, and into the cutoff position. I knew that I was now committed, and it was not a heart-warming feeling. As soon as the throttles were in the off position, the cockpit rapidly depressurized, and frost began forming on the inside of the canopy. Unfortunately, I was still far above the normal operating envelope for starting APU. I remembered that the aircraft flight manual (Dash-1) guaranteed that the APU would start only at or below 15,000 feet, but it might start as high as 20,000 feet. I waited for the aircraft to descend at least 9,000 more feet.
With both engines shut down, there was no hydraulic pressure to power the normal operation of the flight controls. The control stick locked, so I had no ability to roll or turn the aircraft. I bypassed the next boldface step and selected flight controls manual reversion.
The manual system is designed to give the A-10 a limited flight-control capability to improve its combat survivability in the event the aircraft is shot up and loses hydraulic pressure. It uses a cable-and-pulley system to move small electrical trim tabs which act as flight controls. The amount of control that these tabs can provide is a function of airspeed. Since the control surfaces are only a few inches wide, greater airspeed allows the tabs to
