Many of you, it seems never read Deakin's article, or have given much
thought to the procedure. My normal fuel management, when we'd tow
banners, was takeoff on the left tank (the "main" because it has
pickups in the front and back of the tank) and fly for 30 minutes.
Depending on the day and the size of the banner, that would leave :45
to 2:30 hours remaining (fuel burn ranged from 5.2 gph to 9.5 gph.
Switch then to the right tank, and run dry.

Worst case scenario, and I have contaminated fuel in the right tank, I
still have enough in the main to get home on. After running the right
tank dry, I'd know EXACTLY how much longer I could stay on station, and
still land with my required reserve. Normally, I could predict within
5 minutes when the tank would run out, and I'd climb to a safe altitude
in preparation (usually 600' AGL) the most altitude I ever lost while
the engine restarted was 100' on a silky smooth day when that tank ran
TOTALY dry at once, normally it would sputter and cough a few times
first.

In addition to now knowing how much endurence I had, this also meant
that all my fuel was in one tank, and as others have said, if I was to
land with :30 minutes of fuel at 5.5/hr, I'd much rather have all 2.8
gallons in one tank, rather than 1.4 in each! In practice, we never
landed with less than 45 minutes though; we'd want to have enough to
get to our field, attempt a drop, go around with a fouled banner, and
then proceed to the nearest paved runway, land, and fly back to base.

A friend of mine very nearly put one in short of the runway once
because someone else wasn't familiar with the fuel burn, and fuel
capacity. He routinely would get a safe 5 hour endurance out of his
plane, with reserve. Someone else flew it on his day off, and landed
with both tanks on "E" This means very little in a Super Cub, as the
last mark on the gauge reads "3pt attitude- 1/4 to E".

The next day, we didn't have fuel at our strip, and had to refuel at a
nearby airport (4 miles distant). He took off on the last selected
tank (figuring that it was the one landed on, it would be the fullest)
it ran dry on him 1 mile from the airport. Swiching tanks, he was able
to get a restart, and taxi in to refuel. 17.4 gallons was pumped into
the dry tank, and 17.2 into the other. The other pilot later claimed
he "knew you flew it for 5 hours, so I landed after 4" He thought he
had at LEAST a 1:30 hour reserve when he landed, in reality, it was
about 4 minutes.

They were both very lucky; had the other pilot run a tank dry (damn
near did anyway) he would have known his endurance estimate was way
off, and he'd have to land sooner than anticipated. Of course, the
prudent thing to do would have been to drain 5 gallons from another
airplane and pour it into my friend's plane, to be sure he could make
it, but we both assumed that our coworker wouldn't land with less than
5 minutes of fuel in the tanks.

Food for thought,
Mike