Cross-Country Fuel Management by
Running a Tank Dry



The following is a common method of fuel management (quoted from the Grumman-Gang list mailer):

Regardless of when I depart, I try to switch tanks on the hour and on the half hour. The rationale being that it is very unlikely that I  will have BOTH feeds fail, but If I have a problem with ONE of the feeds, I will learn about it while I still have a comfortable amount of fuel in the remaining tank.

There's an inherent gottcha in systematically switching tanks, whether to prevent a wing from becoming too heavy or, as expressed above, to have an out if the switch-to tank won't feed.  I became aware of this gottcha when flying cross country in Nevada and Utah. I had planned a comfortable reserve of 1 hr and switched tanks every half-hour as always.  I then found myself having to divert around some rather nasty thunderstorms which ate into my reserve by about 30 minutes.  Continuing to switch tanks, I proceeded to my destination uneventfully, but upon approaching the airport I realized that I couldn't reliably ascertain which was fuller (the gauges are useless, of course). 

To add to the pucker factor, I couldn't be absolutely certain that I had 5 gallons remaining. If I had timed the switching perfectly, and calculated the fuel burn perfectly, I had a maximum of 2.5 gallons in each tank. There might, however, be 2 gallons in one tank and .5 in the other--with no way of knowing which was which.

What I wouldn't have given to know that all my remaining fuel was in one tank--and which tank it was in!

I reasoned it to be near suicidal to fly a normal approach to land--to do so would be to risk unporting the fuel line while banking in the pattern. I flipped a mental coin, selected a tank, maintained altitude until overhead the airport, throttled back completely, and performed an "engine-out" landing on the longest runway.

My engine died on the ramp while taxiing to fuel. The FBO had been monitoring UNICOMM and sent the fuel truck. I had 3+ gallons in the unselected tank.

Now, I can hear the tongues clucking about the things I did wrong, but the fact is...I didn't do anything wrong.  The build-ups that caused the diversions were not forecasted, and there was no airport nearer than my destination. Once I realized my predicament,  I throttled back for best economy.  The morals of this story are: (a) situations can arise that will stretch any fuel reserve, and (b) know where your fuel is.

Since this experience, I've changed my fuel management technique.  I take off and fly for 1.5 hrs on one tank.  I then run the other tank dry (more on that in a moment).  I am therefore guaranteed to arrive at my destination with all my fuel in one tank.  Rear-seat passengers become ballast against the heavy wing (I've also got a wing- leveler). I plan for a 45-minute fuel reserve, even in VFR. 

Now, about running a tank dry.  In my '79 Tiger I've experimentally determined that the fuel pressure gauge begins to fluctuate about 15-20 seconds before it drops to zero. It then takes another ten seconds for the engine to stop. (BTW, it doesn't sputter--it just goes bang-dead quiet and keeps wind-milling at several hundred RPM.)

During cross-country trips, I turn on my electric pump about 30 minutes before the tank is due to run dry. If I have a passenger, I ask him to watch the fuel pressure gauge; if not, I keep my eye on it closely. I continue to scan for traffic; if I get a traffic call from ATC, I switch back to the fuller tank while I resolve the situation. I then switch back to the "dry" tank.

At the first sign of fuel-pressure fluctuation, I switch tanks. I've done this dozens of times without once letting the engine go quiet.  Since this incident, I've also added a fuel-flow monitor to my Tiger, which enables me to calculate to an accuracy of a couple of minutes when the tank will be empty.  I've also added a fuel-pressure warning LED that blinks when the pressure drops.  In truth, however, if I'm solo or with an experienced pilot, I don't pay much attention and just let the engine run the tank dry.  It's really a non-event.

Note: if you have first-hand knowledge or documentary proof of an engine failing to start on its own when switched from a dry to a wet tank, please with the details.

Joe Campbell
N4524P (Hayward, CA)  

P.S. For a similar article, read the esteemed John Deakin's Run That Fuel Tank Dry! article from his "Pelican Perch" column at AvWeb. John systematically dismisses every objection to this practice either as bad reasoning or deriving from a old wives tale.


Originally posted on the Grumman-Gang 31Dec1996