Saturday, September 15, 2018


There I Was...

...plunging blindly earthward through the clouds, GPS-direct to the safe haven of a runway mere miles away. Again, I scanned the gauges. Something was very wrong. Both fuel gauges were nearly pegged at empty. How was that possible?

An hour and twenty minutes earlier, I departed Burke Lakefront Airport in Cleveland, OH with 25 gallons of fuel on board. The simple arithmetic that always worked to stave off fuel exhaustion in the past had failed. Over the last several minutes, I had watched the needle of the selected fuel tank visibly swing toward empty. I had the mixture pulled abnormally lean in an effort to keep the engine making power for as long as possible. Somehow, the Lycoming ran smoothly and the head temperatures read unusually low despite the aggressive leaning.

Too late, I understood that the possibility of fuel exhaustion before making port was very real.

How did I let myself get here?

Club Outing

Date Aircraft Route of Flight Time (hrs) Total (hrs)
15 Sep 2018 N21481 SDC (Sodus, NY) - BKL (Cleveland, OH) - GVQ (Batavia, NY) - SDC 5.0 1860.2

2018 proved to be a difficult year to organize fly-out excursions for the Williamson Flying Club. Most were scuttled by poor weather, including a planned August flight to Burke Lakefront Airport in Cleveland, OH. We last visited as a club in 2016. Everyone had a great time, but we recognized that there was more to see and do on a return trip. For a 2018 flight, we decided to skip the Rock & Roll Hall of Fame (been there, done that) in favor of touring the USS Cod, a WWII submarine, and the Steamship William G Mather, a restored early 20th century Great Lakers freighter.

A September 15 rain date for Cleveland was also nearly cancelled because of marginal forecasts along the route. We target good VFR conditions for fly-out trips to make them accessible to all of our club pilots. Fortunately, we held off on a decision until the morning of departure and, to our delight, conditions were much better than forecast. Eight people in three aircraft departed the Williamson Sodus Airport around 8:00 am.

Ed joined us in his recently purchased Archer II, Four Four Papa. He had Rick along for the ride.

Tom, Alicia, Mike, and Mike's son made the flight in Archer Eight Five X-Ray. Tom flew the outbound leg and Mike flew the return.

Brenda decided to ride with me when Kristy and The Bear bailed-out at the last minute (it might be tempting to call this prescience, but I think it was more about the early departure time). Brenda was an old friend and a former direct report from the Kalamazoo days who was visiting Rochester for the weekend. Having flown to Groton, CT with me in 2014, she was no newbie to General Aviation flying.

Despite some low clouds south of the New York Thruway (above), the sky was clear most of the way to Cleveland.

Brenda and I had a wonderful opportunity to catch up after a few years out of touch. I had changed jobs twice since the last time we flew together and her company had undergone a significant change in ownership and culture. We had a lot to talk about.

Though the Warrior purred smoothly through the air, something was a little off. Cylinder head (CHT) and oil temperatures were cooler than usual and, at 6,000 feet, the airplane was not quite achieving the usual RPM, even at wide-open throttle. True airspeed hovered around 115 knots, about 5-7 knots lower than normal. Sure, I was still pulling away from Eight Five X-Ray, but Ed's Archer II was outpacing us.

Troubleshooting silently in the midst of conversing with Brenda, I discovered that pulling the mixture back more aggressively than usual achieved better power, though the CHTs still ran low. Although something was clearly different, the airplane ran so well that I was not alarmed.

Burke Lakefront was landing runway 6R that day and we were held high and vectored out over Lake Erie for sequencing before being cleared to make a short approach.

On left base for 6R, we flew directly at the Cleveland waterfront with the steamship William G Mather and football stadium prominently filling the windscreen.

As I reduced power in the descent, I noticed that the #1 cylinder was running much cooler than the others. It reinforced that something was different, but given that low CHTs are not harmful to the engine, there appeared to be no specific danger.

As directed, we parked in front of the control tower and waited with Ed and Rick for the others to arrive in Eight Five X-Ray.

When Tom lined up on final approach, I could see Mike's son in the back seat actively taking photographs with an iPad.

I captured a "behind the scenes" shot of Tom and Alicia's Burke Lakefront selfie.

Alicia, Ed, Tom, Me, Mike, Mike's son, and Rick. Photo by Brenda.

Finally, we had a quorum assembled on the Burke Lakefront ramp. It was time to explore the waterfront!

The Three-Armed Astronaut

The Burke Lakefront terminal building is home to the International Women's Air and Space Museum and features a number of exhibits spread throughout the interior.

One artifact that I had missed previously was this Coke machine that flew on the space shuttle Discovery in 1995 as part of an experiment (one evidently sponsored by Coca-Cola). I can say without hyperbole that, as I read the plaque, a voice in my head proclaimed "Tooth-rot in spaaaaace!" with the same inflection used by the "Pigs in Space" narrator.

We also discovered this strange three-armed astronaut at the museum. They sure do grow 'em differently in Cleveland!

Das Boot

Though Mike and I toured the USS Cod on our previous trip to Cleveland, we enjoyed our second exploration. It was a first time visit for everyone else.

First, we basked in the undeniable gravitas of a bronze plaque.

The USS Cod is a diesel-electric submarine first launched in March 1943. It is 312 feet long and only 27 feet wide. That is not a lot of space for a crew complement that eventually numbered 97.

It was a hot day and this big fan provided absolutely no cooling whatsoever. Useless.

To quote Risky Business, "Who's the U-boat commander?" For the record, quoting Risky Business is not usual behavior for me and, yes, the Cod is American and not German, but I could not resist the opportunity to make an eighties callback.

My friends finally decided to stop goofing around on shore and board the submarine.

If the USS Cod were about to attack Cleveland, I believe that the sight picture would look something like this.

Photo by Brenda

Whoever heard of a bunch of pilots hanging out on a submarine? How fast does this thing go, anyway? Like 22 knots maximum? Does that even qualify as moving?

Photo by Brenda

The same group of pilots with a big gun behind them needs to be taken a bit more seriously, however.

Mike literally went down the hatch, as did we all. The USS Cod is mostly preserved in its active duty configuration. That meant climbing ladders and negotiating tight spaces with a strong likelihood of a bumped noggin or bruised shin while going from compartment to compartment.

Evidently, the bunks in the forward and after torpedo rooms were popular because those spaces were quieter, cooler, and less prone to foot traffic than bunking midships. Pictured above is the forward torpedo room with enough space to bunk fifteen crewman. Looking around, I would have guessed something closer to five. Clearly, my concept of personal space is not applicable to submarines.

That looked like a door meant to withstand some pressure.

Gauges and valves were everywhere. Which spigot was hot and which was cold?

I am not claustrophobic by nature, but I think sleeping on that middle bunk would have made me anxious. Side sleepers need not apply.

Officers on board the Cod had more elbow room, an obvious benefit of moving up through the ranks.

This little office was so tight that I wondered if it came equipped with a crowbar to facilitate crew member extraction.

As a pilot, I am well aware that red lighting is used to preserve night vision, but I wondered about the need for red lighting in the control room. A nearby plaque supplied the answer: so that crew members could go topside at night with their night vision intact.

The USS Cod control room had enough gauges to put the complement of instruments on a typical Cherokee panel to shame.

Pitch is set with two brass wheels operated by a bow planesman and a stern planesman acting as a team. No gyroscopic attitude instruments here, simple bubble levels provided suitable attitude information.

Clearly, the entire 77 - 97 crew complement did not dine simultaneously. I noticed that the tables and benches were on rollers. Was that to open up the space for dancing?

Located in the engine room, it was not clear to me where this temperature reading was being taken, but any temperature gauge that pegs at 1200°F clearly needs to be taken seriously (though, to be fair, the Warrior's exhaust gas temperatures run hotter than that).

One of four sixteen cylinder, 1600 horsepower Cleveland-built General Motors diesel engines used to recharge the Cod's batteries. Because the diesel engines cannot be run while the boat is submerged, all energy for propulsion is provided by electricity, hence the name of the Cod's manufacturer: Electric Boat.

The way out!

The electric motors used to propel the Cod were controlled by this panel in the Maneuvering Room. Here, electricians controlled the speed of those motors and selected their power source, switching between diesel-powered generators when the Cod was on the surface and batteries while submerged.

Ed and Mike in the aft torpedo room.

As I learned the hard way, it is very challenging to climb a ladder through an access with a diameter less than the length of your femur. There was just not enough space to lift a foot from one rung to the next. I hope that the average WWII submariner was shorter than I am.

Mike and son setting up to defend the Cod. So long as that thing is not pointed anywhere near aircraft parking, I'm fine with it.

To quote an old Far Side cartoon, "Never, never do this."

I was fascinated by the rifling of the gun barrel. This photo was taken from the breech of the gun to avoid "pulling a Mike".

After a great tour of The Cod, we departed around noon for lunch.

Trendy, Hipster Tacos

The nearest restaurant was right on the waterfront: Nuevo - Modern Mexican and Tequila Bar.

Our food was excellent. Most astronauts in Cleveland agree, three thumbs up! Nuevo was as good a $100 hamburger destination as I have ever sampled. It may have been trendy, hipster Mexican, but it was genuinely good trendy, hipster Mexican. Prices were reasonable, particularly for an upscale eatery set in a tourist destination. As for the tequila, no one partook.

Suspicious Behavior

Over lunch, I raised the issue of the Warrior's aberrant engine behavior with the group. I did not feel that the airplane needed to be grounded based on its performance flying to Cleveland, but I suggested that Brenda ride back to Sodus with Ed just in case. I also asked Ed to delay his departure until after I was aloft in case I discovered anything worrisome during run-up or initial climb out.

Tom and Alicia seemed to enjoy the Cleveland waterfront.

Cleveland is a pretty good looking town! At least, from where we stood.

We skipped the Rock & Roll Hall of Fame because I'd/we'd been there and done that before.

Even learning that the incomparable John Oates was playing an outdoor concert there, we still took a pass. Sorry, John.

Wally World...Again

Finally, we were about to explore something new to everyone, the Steamship William G Mather!

This was the closest I had ever been to a Laker and it was impressive.

When we reached the entrance, we discovered that the freighter was closed for the day because of a fire that occurred the day before. We had all been Griswolded!

Trusting the Clock

Despite the disappointment of the freighter's closure, everyone had a great trip to Cleveland.

For me, the journey became more complicated when it was time to fly home. As decided over lunch, I flew solo in the Warrior and Brenda rode with Ed.

Pilots are taught that a clock is a better gauge of fuel quantity than the notoriously unreliable gauges built into our aircraft. Leaned for cruise at 75% power, the Warrior will reliably burn about 8.5 gallons per hour. For planning purposes, I usually round this up to 10 GPH both to simplify the arithmetic and to conservatively pad the calculations. This approach has served me well for years; it is not just a rule of thumb, it is a mindset.

On preflight, I discovered that the Warrior's tanks collectively held 25 gallons (2.5 hours of fuel remaining per the 10 GPH assumption described above). Given that the flight to Cleveland was 2.1 hours, I should have had closer to 32 gallons remaining in the tank based on the usual fuel burn. Instead, I consumed close to 12 GPH on the way to Burke Lakefront.


Because I always plan to land with an hour of fuel in the tanks, I decided to amend my clearance partway home to include a fuel stop, most likely at Genesee County Airport in Batavia. With 25 gallons on board, I should have had plenty of fuel to reach Batavia with an hour reserve. On run-up, the engine ran smoothly and behaved well. I launched and the Warrior climbed easily to 7,000 feet.

Fairport Harbor, OH

My cleared route kept me slightly offshore along Lake Erie and out of the clouds. In cruise, I noted that the engine was running cooler than usual again. Aggressive leaning achieved proper cruise power, but the engine still ran cool. This time, cylinders 3 and 4, usually the two hottest, were indicating 315 and 295°F, respectively.

As I passed Erie, I saw a few isolated rain showers farther inland. ForeFlight depicted Eight Five X-Ray and Four Four Papa in trail behind me.

Chautauqua Lake

Significant build-ups towered over the Canadian shore, pillars of white that reflected off the remarkably calm surface of Lake Erie.

At the eastern end of Lake Erie, I asked Buffalo for a diversion to Genesee County airport. This was quickly granted and I adjusted my trajectory a few degrees left to proceed directly to Batavia.

As I approached the wall of clouds ahead, it seemed that the fuel gauges were indicating much less fuel than I would have expected after an hour of flying. Still, I had departed with a nominal 2.5 hours of fuel on board, so reaching Genesee County should not have been a problem.

Nonetheless, as the fuel quantity decreased, I actually observed the needles moving in real time. Though the engine was leaned aggressively, the Warrior appeared to be guzzling fuel like never before. While notoriously unreliable, the gauges for the left and right wing tanks both exhibited the same behavior despite being completely independent of one another. Though I did not understand what was happening with the fuel quantity, the indications were undeniably real.

Farther inland now, I entered the clouds and stayed high, unwilling to give up any altitude prematurely. As I bounced through the clouds toward Genesee County, the fuel levels dropped precipitously. Finally, eight miles from Batavia, I dropped out of the bottom of the clouds, cancelled IFR with Buffalo, and pointed the Warrior's nose directly for the numbers at the end of runway 10.

A Cessna announced five miles west of Genesee County with intentions of entering the pattern. I spotted it off my left wingtip.

"Genesee County Traffic," I broadcast. "Cherokee Two One Four Eight One is five miles southwest, landing one zero. I have a mechanical issue and am planning to land straight in."

The Cessna's pilot indicated that he did not have me in sight, but circled to allow me to land first. I glanced at the fuel gauges and saw that they appeared to be bottomed out, raising doubts that I would reach the airport.

How did I get here? Why did I let things get this far?

For a moment, I considered the reality of becoming another fuel exhaustion statistic, but then focused on flying the airplane. I had stayed high -- as high as possible while still remaining VFR -- and, eventually, had a sight picture clearly indicating that I could glide to the airport. Relief washed over me. The engine could have sputtered and quit right there, I still would have made it.

It never quit. I rolled the Warrior onto the runway at idle with surprising grace, then added power to taxi to a tie down.

Technical Support by Phone

Unprecedented fuel consumption paired with low CHTs pointed to an excessively rich mixture rather than a leak. Somehow, I managed to burn close to 25 gallons of fuel in 1.3 hours, more than double the usual rate of fuel consumption. But why? Though it was a Saturday afternoon, I called Ray.

"Uh oh," he said by way of greeting. With the unusual timing of my call, he knew that something was wrong before answering.

I explained what had happened and answered all of Ray's questions. He suspected that the primer was leaking and drawing excess fuel into the engine.

I suggested that I could leave the airplane at Genesee County for Boshart's crew to examine. Maybe Ed could be persuaded to retrieve me in his Archer. Ray did not think that was necessary. Though he allowed that abandoning the Warrior in Batavia was a fine solution, he suggested that the airplane was running fine other than an excessively rich mixture.

"Fill it up, lean it out, and fly it back here. I'll look at it on Monday."

After thinking about it for a bit, I decided to follow his suggestion.

Welcome Home

The Bear's favorite canoeing destination just off the south end of Irondequoit Bay

A straight line flight from Genesee County to Sodus takes about 30 minutes. Despite the unusual position of the red lever, the airplane flew wonderfully going home. I just used a lot of fuel getting there.

After rolling to a stop on runway 10 at the Williamson Sodus Airport, I added throttle for taxi to the hangar. The engine actually sputtered, running as though started with a full rich mixture at a high elevation airport like Leadville, CO. Pulling the mixture way back soothed the stumbling powerplant. Had it idled that way in Cleveland, I would have never departed.

Ed and Brenda, Tom and Alicia, and Rick had all waited to make sure that I made it home. They heard my request to Buffalo Approach for the diversion to Genesee County, but did not know anything beyond that. I was gratified that they cared enough to wait.

The Fly in the Ointment

The problem was not a leaky primer, it was far more obscure than that. A thin strip of felt from the flapper valve in the carburetor heat air box had detached and lodged itself in the throat of the carburetor. Its presence in the venturi appears to have directly affected the metering of fuel. Once Ray removed the strip of felt and replaced the seal in the air box, Warrior 481's engine went back to running normally again. It was an easy fix.

With the airplane back in shape, what about her pilot?


This episode represents a string of failures on my part. While openly dissecting them is not pleasant, I think it is necessary and worthwhile.

My first opportunity to manage the issue was right after Brenda and I leveled off for cruise flight in Rochester's airspace that morning. I noted the low CHTs and convinced myself that, because the airplane was running so well, there was nothing seriously wrong with it. None of the indications were inherently problematic, but they signified that something was unusual. I could have turned back at that moment, but chose not to.

The next best opportunity to manage the issue was during preflight in Cleveland when I noticed evidence of high fuel burn. As pilots, we are constantly reminded to trust the clock with respect to fuel planning. This is really good advice, except that it was a poor strategy in this case because I had data suggesting an unpredictable burn rate. Granted, it is difficult to overcome a mindset established over 18 years of flying, but aviators need to react to changing circumstances.

In my subjective reality, the rate of fuel burn seemed to increase drastically after I was approved to divert to Genesee County. Whether that is true or merely my perception, an earlier diversion would have pulled me back from the ragged edge in a more satisfactory way. Was I too fixated on Genesee County after making the decision to go there while still on the ground in Cleveland? I cringe when I contemplate the number of airports I passed only to just barely reach Genesee County.

With the fuel burn approaching 18 GPH on the way to Genesee County, all of my endurance calculations were significantly flawed. I was slow to realize that the fuel burn was higher on the way home than it had been going to Cleveland. Unfortunately, I do not have any fuel flow equipment installed that would have indicated that quantity directly. As a result, I continued to put my trust in the clock for too long.

I think that I can lump all of these issues under a general failure to recognize and meaningfully react to changing conditions in a timely manner. Usually, I think that I manage changing conditions quite well, particularly with respect to variable weather and en route decision making. I did not do well in this case. The difference is that I expect weather to be dynamic, but view fuel consumption as a constant. It took a lot of data to convince me otherwise.

Did I do anything right that day? Maybe one thing. Once I was suspicious of an issue, even a poorly defined one, I asked Ed to fly Brenda back in his airplane as a precaution. I think that was a good choice; do no harm.

As for making it to Genesee County Airport with the engine still running, I may have improved the situation with aggressive leaning, but the positive outcome was really just dumb luck. Considering the farmland around Batavia, I had many good options for landing out, but am delighted that I did not need to use any of them.

For me, this was a relatively inexpensive lesson in the dangers of complacency. Airplanes talk to pilots in many ways. Sometimes those messages are unambiguous, but at other times, the meaning may be unclear. In those latter cases, it is essential to stop and listen.



    My own experience with the same sort of issue. Even with a LOT more data in the air it is amazing how difficult it is to believe reality. Laurence Gonzales (who every pilot should read) calls this "bending the map," since hikers who are lost will look at things that are different from what they expect and bend the map to fit what they see, rather than admitting they are not where they *think* they are.

    1. "Bending the map" ... I like that. It seems like a less clinical-sounding way to express "confirmation bias". And thanks for the tip, his work looks very relevant.

      I remember your post, Colin. I remember cringing while I read it and wondering what I would have done in the same situation. It seems that I did not remember it well enough.

  2. Great lessons all around on this trip. I can't say I'd have done anything different - it's basically instinctive to trust the clock when you're unsure of something with the fuel gauges. As much as we all know a precautionary landing is almost always the right choice, it's still much harder to make that call in the air; I think most of us have been in that situation at some point.

    Glad all turned out well - and thanks for another useful nugget of aviation wisdom that'll get buried in my brain, hopefully to be recalled if I ever need it!

    1. Thanks for the comments, Steve. Mental inertia can be a tough thing to break.