The Rules and Fuel

The most obvious requirement for continued powered flight-fuel in the tanks-seems to confound pilots more than P-factor. It's impossible to know for sure how many pilots run out of fuel each year, because many power-loss events do not result in an accident if there is no serious personal injury or substantial damage during the forced landing. But it is safe to say that fuel exhaustion is a leading, if not the leading, cause of power loss in general aviation flying.

I was reminded of pilots and fuel by the responses we received concerning Peter Garrison's "Aftermath" analysis of a fuel exhaustion accident that appeared in our February issue. In case you missed that issue, the accident involved an ATP-certified pilot with 25,000 hours and type ratings in several airline jets. The pilot was on an IFR flight in a Piper Arrow. En route he changed his destination airport and crashed after running out of fuel while flying what he told controllers was a "practice ILS" at the new destination.

Peter pointed out that the FAR fuel requirements are based on planning, not the actual amount in the tanks when you land. That interpretation is unambiguous under VFR conditions where FAR 91.151 says that no pilot can begin a flight unless consideration of wind and weather will allow the flight to reach the planned destination with enough fuel to cruise for 30 minutes at a normal power setting, or 45 minutes at night. A literal reading of FAR 91.151 indicates that running out of gas is not against the rules so long as your preflight plan was complete and accurate based on the winds aloft forecast. If the wind or weather forecasts are wrong, there is apparently no legal obligation under that rule to land and buy more fuel. Of course, FAR 91.13 appears to forbid running out of fuel, because that rule bans all "careless or reckless operation."

FAR 91.167, the rule that governs fuel requirements in IFR conditions, seems to require a pilot to land short of the planned destination if fuel runs low. This rule says "no person may operate a civil aircraft in IFR conditions unless it carries enough fuel" to complete the flight and land with enough fuel for 45 minutes of normal cruise. If an alternate airport is required, which is determined by the forecast weather at the destination, then the 45 minutes of fuel must remain upon reaching the alternate.

Aha, several readers have pointed out, the pilot of the Arrow that Peter wrote about was on an IFR flight and therefore was legally required to land as soon as the fuel onboard dropped below the 45-minute reserve at the destination. That is one possible interpretation of the rule, but when does a pilot on an IFR flight know that he lacks the required reserve? If, for example, the strong headwind you are now flying into continues to the destination you would not have the necessary reserve. But what if the wind is forecast to lighten as the trip progresses? You would in effect be "creating" reserve fuel as you flew into the lighter winds and thus be "operating" with the legal reserve.

Here's another, even bigger, ambiguity in the rules. FAR 91.151 and 91.167 specifically refer to flight in "VFR and IFR conditions," respectively, both in the label and the body text of the rules. In FAR 1.1, which contains the FAA's official definitions of terms used in the rules, "IFR conditions" is defined as "weather below the minimum for visual flight rules." So it appears that a pilot flying IFR, with a clearance, but not above 18,000 feet and in VFR weather conditions, is flying under the fuel requirements of FAR 91.151, which applies to flight in VFR conditions and only requires fuel reserve for the flight plan trip. But if a pilot flying IFR flies into the clouds, then FAR 91.167 seems to apply and he needs a different fuel reserve. Got that?

I don't have answers to these conflicts in what the rules actually say, compared to what we have all been told they actually mean, and probably what the FAA intends. And the fact is, the rules don't really matter. Pilots flying under the personal and business flying rules of FAR 91 are not going to have a fuel reserve problem with the FAA unless they actually run out. Flights under the airline or charter rules with all of the routine monitoring that goes on by the FAA are another matter, but those guys are doing a good job of managing their fuel reserves.

The reason I'm dwelling on the lack of clarity in the rules is to point out that there is no legal answer to the fuel exhaustion accident problem. The intent of the rules is clear-to require pilots to build in a reserve of fuel for the unexpected-but the hairsplitting the rules invite is, I think, counterproductive.

When it comes to fuel planning we need to throw the rule book out the window and apply common sense. Every airplane and situation is different, but pilots flying for personal reasons must determine a conservative reserve and not go below that amount of gas in the tanks for any reason. Dispatchers, with their shared legal responsibilities, do this for airline captains. When you fly on your own, you are also alone in enforcing your fuel minimum.

In piston airplanes I believe that you must think of fuel reserves as a number of gallons, not amount of cruise time. Even with big six-cylinder engines you could hold, or cruise at low power, for an hour on as little as five or six gallons of fuel. An hour's worth of fuel is legal and would appear to be conservative as a reserve. But five or six gallons is not conservative because it will be spread out over at least two tanks, and it is not possible to reliably measure such a small amount of fuel in many airplanes. Even a fuel totalizer system that measures fuel burned with great precision cannot know with certainty exactly how much fuel was onboard at departure. The five- or six-gallon reserve may not have even been there at takeoff if the fuel level was slightly below the filler port or the airplane wasn't sitting level when it was fueled.

Another problem is the widespread belief among piston airplane pilots that fuel gauges are unreliable. I would agree that many gauging systems lack great precision, but I disagree on the reliability issue. Every fuel gauging system that I know of is designed so that its most likely failure mode, or gross inaccuracy, is to show less fuel than actually is in the tank. For example, if the wire breaks off or corrodes, the gauge will show empty. If every pilot believed a gauge that showed empty and landed immediately, he might make some unnecessary precautionary landings, but the fuel exhaustion accident rate would go down.

I am sure that some fuel exhaustion accidents are, in part, caused by pilots loading partial fuel to stay within the maximum takeoff weight limit. This is absolutely the norm in jets with their excellent fuel gauging systems, but is a bit of a guess in many piston airplanes. In most piston airplanes your only precise fuel measure is a visual check of absolutely full tanks. After the top off you can have a pretty accurate notion of how much fuel you burned on a given flight and thus how much remains so you can add the necessary amount for the next leg with reasonable confidence. But if you partially fuel several times in a row, the opportunity for error-both in human calculation and in gauging-creeps in. Staying within takeoff weight limits is a worthy goal, but is not nearly as important as having enough fuel, so if you are going to err, do it on the fuel side, not the weight side.

My final bit of advice on fuel planning is to always put the fuel stop in the middle of the flight. If it is even close that fuel will be an issue, land and refuel at an airport as near the center of the flight plan route as possible. Human nature is such that our optimism grows as the destination nears, and it is almost impossible to make yourself land for fuel in the last hour of a planned nonstop trip.