It was a sunny January morning in Northern Baja, Mexico, when we landed at San Felipe. I taxied my Twin Comanche over to the fuel pumps as I had on nearly a hundred other occasions since I began flying as a volunteer pilot with Mexican Medical Ministries in 1992. Our team consisted of myself, my good friend and mission copilot Jon, and two women who were to serve as medical volunteers. Our plan was to refuel, clear Mexican customs, and fly another 350 miles south to Loreto, where we were to participate in a free medical clinic in support of a local church.
Jon supervised the refueling of the Twin Comanche while my passengers and I started on the paperwork. An hour later we were off again, heading south.
Shortly after departure, one of my passengers said she observed fuel flowing out of the left auxiliary fuel tank cap. I looked out and saw the steady mist as the low pressure above the wing sucked fuel from the tank. I discounted the significance of the leak, because on numerous other occasions the same thing had happened after the tank was filled to the brim. In the past, however, it was the left main cap that leaked, not the auxiliary.
I discussed the situation with Jon, and he said he’d noticed during the refueling that the cap was loose. He had attempted to tighten the cap by increasing its diameter with the built-in cap adjustment key. Satisfied with his answer, I assumed that the fuel leakage would stop within a minute or two, as it always had before with the main tank cap.
Thirty minutes after departure and level at 7,500 feet, I switched from the mains to the auxiliary fuel tanks and started my countdown timer, which I set at an hour and 45 minutes. I glanced over at the left auxiliary fuel cap, and was surprised to see that fuel was still flowing out in a steady mist. I looked down at the fuel gauge and chuckled at myself under my breath for doing so. I knew that the gauges never showed the actual amount of fuel remaining in the tank, and the fact that it still showed full meant that I had anywhere from half a tank to full fuel.
Due to the inaccuracy of the gauges, I always used my timer to monitor the fuel burn. Shortly thereafter the flow stopped, and I promised myself to inspect the cap after we landed in Loreto.
Next, we crossed the Baja Peninsula southwestward to the Pacific Side for a detour to Guerrero Negro (Black Warrior) and descended to 1,000 feet for some whale watching. California Gray whales congregate there to calf their babies and frolic in the warm waters before returning to their Alaskan feeding grounds. I reduced power to 18 inches of manifold pressure and slowly pulled back the props to 2,100 rpm, in order to fly as quietly as possible. While we circled I mentioned to Jon that I typically didn’t fly with the rpm settings that low. Just then the aircraft yawed slightly back and forth with the engines making a surging sound as the propellers went in and out of sync. I leveled the wings and pushed the throttles, props, and mixtures full forward, but the surging continued. A quick glance at the fuel gauges revealed both to be well above half full.
My first thought was that the much lower than normal rpm setting had somehow blown the nitrogen charge in the left prop, which I knew had a slow leak through the Schrader valve. Jon was also aware of the problem, and we both jumped to the conclusion that the nitrogen charge had leaked down to zero, causing the left prop to run away.
Fortunately, we were only five miles west of Guerrero Negro Airport, so I pulled both throttles back to reduce the yawing and set up for a right downwind to Runway 27. There was no common traffic frequency on the chart, so I selected frequency 122.8 and broadcast our position and intention to make right traffic to land on Runway 27. Guerrero Negro Airport is located in a sparsely populated area with very little air traffic. Nonetheless, I scoured the area for other aircraft and was quite surprised to see a Cessna 206 on final for the same runway. I extended my downwind for spacing. Flying further away from the airport was the last thing I wanted to do, and I made a mental note to have a word with the other pilot after we landed.
I turned right base and then final and made an uneventful touchdown. During the rollout on the runway, the left prop stopped, and it was then that I finally realized that I had run out of fuel. The left engine, I discovered, had not been generating any power at all. A wave of embarrassment engulfed me. The passengers’ compliments on a smooth landing did little to diminish my shame for having jumped to a wrong conclusion and ignored standard engine-out procedures, which would have included switching tanks.
Doing so would have restarted the left engine, supplying it with fuel from a nearly full main tank. A glance down at the left engine fuel gauge showed that it was now indicating empty; evidently the landing vibrations reset the mechanism.
We taxied on one engine to the tiedown area next to the Cessna that had cut us off. We exited the aircraft, and Jon looked into the left auxiliary fuel tank and confirmed what I already knew. It was bone dry.
I approached the pilot of the Cessna and asked him if he was on frequency. He said that he was—on 123.3. I asked how he came to use that frequency as he had missed my calls during my emergency landing, and he said, “It is painted on the runway.” Sure enough, it was, but I didn’t see it nor had anyone else on board.
As I look back on this experience, I realize that I had allowed myself to become complacent. I didn’t want to believe the fuel leak was a big enough problem to justify turning back to San Felipe. Although the leak persisted much longer than it did in the past, I kept on believing that nothing was wrong.
Complacency is defined as an attitude of invulnerability, or being self-satisfied and unaware of possible dangers. After 2,400 hours as pilot in command, with no significant in-flight problems, I was clearly guilty of all of the above. Blessed with the good fortune to not have experienced a significant in-flight emergency, you might think, “It can’t happen to me.” But it can.
E-mail the author at [email protected].