Fuel Strainer Lottery

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

It was a Wisconsin-warm, winter-to-spring day. Roads and runways were plowed, but the heavy rains from the previous night did little to diminish several feet of snow on lawns and fields. My friend (let’s call him Bill) and I drove to the airport at 12:30 p.m. to practice some aerobatics in a rentable Super Decathlon. I had recently completed my first 10-hour course of basic aerobatics, while his history included teaching aerobatics in Cubs, Stearmans, Decathlons and gliders, and he was typed in a handful of corporate jets.

On recent flights, I had goaded him into showing me rolling 360-degree turns, which he eventually limped through (no fault of his own in a Decathlon), and I intended to learn this because it was so damn hard to do. This would not be the day I’d learn anything like that.

The FBO guys had added 15 gallons of fuel before our 90-minute practice session. We back-taxied on the 2100-foot runway, did the runup and began the takeoff roll. Bill was in the rear seat. Shortly after we lifted off, I noticed that Bill was reducing power. I thought, “That’s a weird thing to do, but I guess it’s a surprise emergency test.”

I pushed the throttle forward, and realized it was already full forward. I wondered about the mixture, but wait … he doesn’t have a mixture control. All of this happened, of course, over a few seconds of time that seemed like a minute. When Bill suddenly recognized I wasn’t fiddling around, he said, “I got it!” I relaxed my grip on the stick. (I was OK with that. Remember his credentials.)

So here was the problem. We were nearing the end of the runway, which terminated at 90 degrees to the airport’s road. Across the road was a pure, white, inviting farmer’s field. But we both knew something about that field. Under the snow were trailers and harvesting equipment. If we knew exactly where they were buried, maybe we’d survive. Or maybe and rip up the belly, and end up inverted. The fuel line down there could then light us up.  And it’s hard enough to get out of that back seat at all, much less doing so upside down and on fire.  We didn’t discuss this as there wasn’t time, but we were thinking alike.

“Take the road?” I asked or wondered.

“Got to,” he said, already in a descending, steep bank.

So, the airplane did what they all do when they run out of lift. We impacted with a large, crunching noise, and then experienced Newton’s third law -- every action having an equally angry reaction. We were lofted maybe 15 feet in the air, and again, no time for discussion. My fear was again that damn fuel strainer on the bottom of the fuselage. Did we rip it up on impact? After we put it on the road, did we have functional landing gear, or did we splatter them?

Sirens of soft, non-flammable snow called to me. “Put it in the snow!” I demanded or suggested or begged. Luckily we had enough airspeed to allow some aileron control, and after clipping a road sign with the right wing, we swooshed to a stop, about six feet off the road. All was quiet, including us.

I secured the airplane – mags, fuel, mixture, master,  and finally said, “You OK?”

“Yeah. You?”


After another 10 seconds, I said, “I think I broke my back.”

“Yeah. Me too.”

We talked a bit more. Bill was in pain. Cars passed us, but nobody stopped. We had barely left the airport, so I told Bill I’d go back and get some help. I walked in and asked the new employee if the airport owner was available. He told me Leo had gone to lunch. I asked if the owner of the Decathlon was there, and was told Jim was at lunch with Leo.

Suddenly he jerked back and said, “Didn’t you just take off in the Decathlon?”

I explained what happened. He called the fire department. This was his second day on the job. A day earlier, an instructor and student cartwheeled a Cessna 150 on a landing. When Leo and Mark returned, their new employee explained what just happened, and quit on the spot.

By the time I walked back to the scene, there was a fire truck, several squads, and lots of cops and sheriffs and medical people, some of whom were talking to Bill. He was still in the airplane. As I trudged through the snow, a sheriff intercepted me, asking where the hell I thought I was going.

“Back to the airplane. I’m the pilot.”

“No you’re not. The pilot is in the airplane.”

I am! I‘m the guy who walked back to the airport to report this.”

“I can’t let you get any closer to the airplane.”

“Go ask the guy in the back seat. It’s the passenger seat. You can’t fly this airplane from the back seat unless a pilot or student or passenger is in the front seat. I’m the pilot. Ask him.”

Soon after Bill convinced the sheriff I wasn’t crazy, he accepted their offer of an ambulance ride to a hospital, and I was invited along. He was on a stretcher; I had a seat next to him. To pass the time, we sang “Stayin’ Alive” most of the way to the emergency room. Later, Bill was told he’d have to stay a night or two for observation. I was free to head back to the airport and retrieve Bill’s car and the six-pack he had buried in the snow for our post-flight briefing.

When I got back to the field, the Decathlon was there. Leo told me they started it up in the presence of the NTSB investigator, and it ran fine. He wanted to know exactly what I had done before, during and after the accident. I felt I was being accused of something. (I probably was.)

Later, they drained the fuel. Turns out, there were six gallons of water in that poor little airplane. At least a quarter of the fluid in the tanks was H2O. That prompted them to check the fuel in the underground tanks, which rewarded them with another 150 gallons of municipal water.

Did we forget to drain the sumps? Sorta. The owner had stated to his students and renters that he didn’t want a fuel sampler banging around inside the cockpit during aerobatic flight, so he suggested draining fuel right onto the ramp and just smelling it. (Today, that could cost you a $10,000 fine.) Following that advice was just stupid, and it would result in the NTSB’s final conclusion that I failed to do a complete preflight.

Leo’s insurance company came to visit me a few weeks after the accident. They wanted to agree on a settlement based on how much I earned per day, how many days of work I lost, and what kind of money I was looking for. They were shocked that I had not retained an attorney. They were elated when I said that the accident could have, and should have, been prevented by me. I eventually and somewhat reluctantly accepted $200 for lost wages, and signed a full release. Bill and I never discussed whether he received a settlement, but the NTSB report stated there was one minor injury, and one serious.

To this day, something puzzles me about that engine runup after the landing. The Decathlon has two fuel exits per tank, one fore and one aft. After that airplane sat tail-low for an hour or more, why wasn’t the water immediately sucked in from the low side? How could it have run on water? I’m open to suggestions.

I still meet pilots who never bother with a fuel sampler. Why? Because (tada) they have never found that mysterious half inch of water in the bottom of the device. Yet by not checking, they guarantee they never will. Sure, a situation such as mine might be as rare as winning a state lottery. But people buy tickets anyhow, just in case. Properly straining fuel is free, and every time you don’t get what you’re looking for, you win.

Comments (17)

Could the aft pickup point have been blocked by ice, and thus the water wouldn't get sucked in until the tail came up and it migrated to the forward pickup?

Posted by: Robert Gatlin-Martin | December 1, 2018 6:06 AM    Report this comment

Did they ever determine how 150 gallons of water got into the fuel tank? That's not condensation. That had to either be brought in OR there's a leak OR -- worse -- vandalism?

Posted by: Larry Stencel | December 1, 2018 9:05 AM    Report this comment

Several years ago a student asked how far would a C172 go with the fuel lever turned OFF. "I don't know but we'll soon find out" I said. So we preflighted the aircraft, then set the fuel selector to OFF, started the engine, taxied about .3 miles to the runup area , did a normal runup checking the ignition system, lined up for takeoff applied full power, started roll and just before Vr (53kias) the engine died. Total time from startup to engine out 13 minutes. Fuel flow on taxi 1.6gph, on take off 14.6gph, engine Lyc. O-360. Same procedure without the distance to runup area would allow the aircraft to become airborne with fuel selector OFF. I would think.

Posted by: Rafael Sierra | December 2, 2018 10:23 AM    Report this comment

So Raf, tell me, what would you have done if the student rotated and went airborne? Would you have allowed the student to rotate? How far down the runway would you have allowed the plane to accelerate and role? Just wondering.

Posted by: Tom Cooke | December 5, 2018 7:11 AM    Report this comment

Not sure if how good a sump of your plane's tank would really be if you did it immediately after the FBO pump churns up the water and gas together, It's false security to sump the tanks immediately after filling them; IMHO.

You won't catch Jet A nor suspended water buy looking at the few ounces that occupies the sump.

Posted by: Mark Fraser | December 5, 2018 7:17 AM    Report this comment

"I still meet pilots who never bother with a fuel sampler. Why? Because (tada) they have never found that mysterious half inch of water in the bottom of the device."

Lucky them. On several occasions I have found water in the fuel samples I have taken over the years. Sometimes a scarily significant amount (it was a leaky fuel cap).

That's a bit longer than it took the Piper Archer I tested this on to quit. If I recall, it was around 5-8 minutes (I don't remember the exact amount of time). Definitely long enough to get from the run-up pad and airborne, but not long enough to get to a safe enough altitude for a turn back. I tested this just by switching the fuel lever to OFF and waiting for the engine to quit at the tie-down, but it would have been interesting to see how a run-up and taxi would have changed things. And I presume the 172 got further because perhaps there's more fuel line distance from the fuel shutoff to the engine than in the Archer.

Posted by: Gary Baluha | December 5, 2018 9:12 AM    Report this comment

I've had water show up in the fuel samples several times; purging it can be problematic because you don't ever really know if you got it all out. Leaky vented caps in rain/storms and overnight condensation from aircraft parked outside with near-empty tanks is usually the culprit. I check it every single time unless it's a hot turn-around.

But 150 gals in the underground tank is ridiculous.

Posted by: A Richie | December 5, 2018 3:28 PM    Report this comment

What the... my comments don't seem to be saving. Hotel internet problem? Testing one two...

So one more try:

Regarding the 150 gallons of water, we eventually assumed it was due to the rains, and some massive leaks in the underground tanks. They were soon removed and replaced with above-ground tanks. No further incidents such as this occurred.

Posted by: Jeff Parnau | December 5, 2018 8:01 PM    Report this comment

Several years ago, I flew my son to a paradrop area in our 1974 C172. It was located on a grass strip and the taxiway where I parked was on a slight incline, so I turned the fuel lever to off to prevent any cross feeding.

On departure, we were in a bit of a hurry (ain't it always the case), so I skipped the pre-flight, fired up, taxied to the runway and took off.

Halfway down the runway and at an altitude of around 50 feet, the engine died. Thankfully, this was in my pre Alzheimer days and I, without thinking, quickly reached down and turned the selector to BOTH. The engine surged back to full power and we continued our climb.

I don't think we lost any altitude. I instinctively lowered the nose and we stayed pretty much level with the tree tops. I don't think I was even breathing hard. But I sure was on the flight home when I thought about how close we had come to ending up in the trees.

Time from engine start including a 30 sec taxi and maybe 15-20 sec at full power--maybe a minute. That's with a 150 HP C172. YMMV.

Posted by: Peter Millard | December 6, 2018 10:39 AM    Report this comment

Tom Cooke, I expected the engine to stop before the take off roll. The runway is 8500 ft. My intention wasn't to determine approximate time from startup to shutdown. This was in a 150hp C172. I'll try the same in a 180hp C172.

Posted by: Rafael Sierra | December 7, 2018 9:37 AM    Report this comment

I never skip the sumping, maybe because I have found water in my fuel tanks. After refueling my 210 at a decent-sized airport in Illinois I drained three full testers of water. Something about the appearance of that first tube made me suspicious. If you're in a hurry it wouldn't be too hard to mistake 100% water for 100% fuel. Beyond just the color, which isn't that obvious in a small fuel tester, the surface tension is different, and contaminants act differently. Tiny bubbles of water in fuel fall to the bottom, but tiny bubbles of fuel float up in water.

Posted by: David Finamore | December 7, 2018 10:33 AM    Report this comment

I just re-read the first comment submitted by Robert Gatlin-Martin. I initially thought something similar happened, that lifting the tail allowed the water to hit the the front fuel port.

My buddy Greg Koontz flies a Decathlon in airshows. He's also one of a few American Champion U.S. dealers. He called his folks at the factory. They felt that the large volume of water had not settled, and that the large percentage of water mixed with fuel was enough to prevent combustion.

Unless someone wants to test these theories by experimenting with mixing water and fuel, I'm not sure I'll ever know exactly what happened, but I still lean toward the raising of the tail as a contributing factor.

Posted by: Jeff Parnau | December 7, 2018 12:12 PM    Report this comment

Years ago my company was assessing different aircraft to fulfill a feeder freight role. One aircraft considered was the Cessna 402, so the company entered into a lease agreement to evaluate a 402 in daily operations. The aircraft was delivered and the next day it was my job to go fly the plane, check out what we have to work with.

Preflight didn't go well. The first fuel sample from the left sump filled my fuel tester with...mud. The second and third samples were mud also. The fourth sample was not mud and I thought, "Finally, fuel."

Nope. Water. Got to fuel after 2 more samples of water.

The right sump was the same story. End of preflight, the airplane was turned over to maintenance to have the tanks and fuel system drained and flushed.

Obviously the sumps hadn't been checked in some time, and the scary part was that the airplane had been in scheduled part 135 passenger ops right up to the day before it was delivered to my company.

Posted by: Bob Bostick | December 7, 2018 3:13 PM    Report this comment

"If you're in a hurry it wouldn't be too hard to mistake 100% water for 100% fuel"

Seems like after a dozen or so years, those fuel samplers do take on a blueish hue.

For that reason, I take 2 samples from each point. If both are "clean", I move on.

If I get water, I sample until clear and require 3 clear samples afterwards.

Now, the logistics..how many of you have taken samples and the airport hasn't provided a proper means of disposal of said samples?

Posted by: Robert Ore | December 7, 2018 6:34 PM    Report this comment

Good point. If my sample is "good " I just pour it back into the tanks. We've got one of those half-liter jars with replaceable filter paper in the pour-back outlet.

Posted by: YARS (Tom Yarsley) | December 8, 2018 10:30 AM    Report this comment

Note to those who may mix 100LL with Auto Fuel:

When 100LL is mixed with ethanol auto fuel the water separates from the ethanol. We had a rotax powered aircraft that was regularly using 10 percent ethanol from the local fuel station. When the aircraft did a cross country flight the pilot filled up at the FBO and topped-off the tank with 100LL. No problem the flight home but, the next morning the 25 gallon tank with 10 gallons in it had a coffee can of water.

After that instance we experimented. It only takes about 20 percent 100LL to release the water from the ethanol. Try it, use a glass jar of auto fuel add a little 100LL and check out the next day.... scary!

Posted by: Klaus Marx | December 8, 2018 12:02 PM    Report this comment

Last spring at Sun 'n Fun, I flew in with several other friends in their aircraft to spend the week together. Upon landing, most of us had the fuel truck top off our tanks to prevent condensation in the tanks. During the week there was a heavy rain storm one night. Upon preparing to depart, most of us experienced water in the fuel, which we blamed on the rain storm. However, the planes had all different fuel caps, so it was a surprise that we all had at least some water. None of us thought that there could have been water in the fuel truck, and since none of us had tested the fuel back then, there was no way to tell. But, after reading this article, it does look like a good possibility that the truck was a source of at least some of the water. From now on, I will wait 10-15 minuets after refueling and then test for water.

I am also with YARS in that I use a larger sample jar and drain at least a full cup of fuel from each tank and the belly drains. That gives the fuel its more distinctive blue color, and makes it easy to spot any water interface. If the sample is clean, I pour it back into the tanks through a filter to trap any debris.

Posted by: John McNamee | December 8, 2018 6:29 PM    Report this comment

Add your comments

Log In

You must be logged in to comment

Forgot password?


Enter your information below to begin your FREE registration