Another Lake Powell Photo Flight

Perfect conditions, except for the cold.

It was a photo flight that brought me to Page and Lake Powell on my annual trip from Wickenburg, AZ to Seattle, WA for cherry drying season. Although I usually depart Arizona mid-May, when this flight came up, I figured I’d use it as a springboard to start my trip. Rather than return to Arizona, I’d use the ferry fees to take me two hours closer to Seattle. Not exactly a direct trip — it would likely add at least an hour to my total flight time — but it was something.

I flew to Lake Powell on Thursday afternoon so I could be ready for the photo flight at 6 AM on Friday. I planned to spend Friday night in Page as well, since our second flight of the day was scheduled for 6 PM.

So at 5:30 AM, I was at the airport, preflighting the helicopter. It was already light; the sun rose while I was working. There was a thin layer of clouds to the east and it filtered the sun, softening the light. Exactly the kind of light my best Lake Powell photo client likes.

Fifteen minutes later, I was waiting in my rented car in front of the terminal. I own a hangar at Page — it’s for sale if anyone’s interested — so I have a key card to get onto the ramp. I figured I’d pick up my passengers in front and drive them out to the helicopter. I’d then leave the car out there until we got back.

My passengers were on time. The photographer was a big guy, weighing in at 240 pounds. His wife was much smaller and thinner; only 130 pounds. We said our introductions and I drove them out to the helicopter. I pulled the front passenger door off and put it in the trunk of the car. Then I gave them a safety briefing and handed out the life jackets. (I always make my passengers wear life jackets over the lake.) They were Russian — it seems that most of my Lake Powell photo clients are — but they spoke English well. I don’t speak any Russian.

It took a while to warm up. The temperature had dropped down near freezing overnight. But by 6 AM sharp, I was pulling pitch and taking off. We climbed out and I turned uplake.

That’s when I realized that my photographer client had no idea where he wanted to go. He told me to take him to places that I knew were good. Places I’d taken other photographers. And Rainbow Bridge.

Gregory ButteSo we did a quick circuit around Gunsite Butte and Alstrom Point, then hustled up the lake toward Rainbow Bridge. Although the forecast had called for 50% overcast and morning winds up to 10 MPH, the sun was bright but filtered and the wind was not an issue at all. Flying toward the sun, the views weren’t very good, but glances to the left or right as I turned showed stunning morning views of the lake and red rock cliffs. My client didn’t provide much instruction, so I didn’t waste his time by being a tour guide. (Note that this photo and all the others that appear in this post were taken at other times; I can’t take pictures while I’m flying with clients aboard.)

I should mention here that it was bitter cold. With my heat not functioning — it had triggered a carbon monoxide warning the day before — and a door off, the cold morning air rushed in. I had a hat, scarf, and three layers of long sleeves on, but no gloves. My client was out in it. His wife sat behind me and didn’t say much, so I don’t know how she was taking the cold. But I figured that Russians were probably a lot more accustomed to cold weather than thin-blooded Arizonans.

Rainbow BridgeWe had a bit of a tense time when my client asked me to fly lower near Rainbow Bridge. The problem is, Rainbow Bridge is at the bottom of a narrow canyon. I know that I can get down to 5,000 feet MSL safely, so that’s what I always shoot for. He wanted me lower. That would bring me very close to a canyon wall. When I pointed that out, he backed down. I think he may have been spooked. But I gave him a good look at the bridge and was satisfied to hear his camera clicking right through his microphone.

Afterwards, he told me he wanted to go into a canyon he’d seen on the way up. But we were so close to Reflection Canyon, which everyone loves, that I asked if I could take him there. It added about 10 minutes total to the flight. He seemed happy with what we saw. The light was breaking through the clouds by then and although it was still soft, it was brining out the colors of the lake and the rocks.

We headed downlake and I reminded him several times that all he had to do was direct me and I’d fly wherever he wanted me to. We went into Wetherall Canyon on the south side of the lake, but he turned me around after only a few minutes. Then back to the main channel. Another side trip up Rock Creek’s main branch and back. Then another trip up Last Chance Bay and back. And some time over Padre Bay. There was a lot to see and the light was good, but he didn’t direct me to do much. I just flew, trying to enjoy myself, trying not to worry about my client not getting his money’s worth because he didn’t tell me where to fly.

Horseshoe BendHe asked me to take him to Horseshoe Bend and I headed out that way. I purposely swung past the Glen Canyon Dam, putting it on his side of the aircraft so he could take photos. I heard his camera snap. Then I climbed — Horseshoe Bend is best seen from at least 6,000 feet — and headed downriver. A few minutes later, I pointed out Horseshoe Bend and began making a climbing circle to the left around it. I was about 270° into it when he told me we were done.

I dumped the collective and started a steep descent, banking east toward the airport. I landed on the taxiway as a National Park Service patrol plane took off.

Here’s a look at our route on the official Park Service map:

Photo Flight

My client jumped out while I was cooling down the engine. He snapped a few pictures of the aircraft as his wife got out and joined him. Then I shut down, got the blades stopped, and climbed out. We’d flown 1.3 hours.

I brought them back to the front of the terminal. My client told me that he was going to cancel the evening flight — it was just too cold. I was shocked. He’d paid me for 5.3 hours of flight time but had only been airborne for 1.3 of those hours. A second flight would cost more, but it would also help spread the cost of the ferry time and help him get his money’s worth.

But who was I to argue? The wind was supposed to kick up later in the day and I wasn’t eager to be out over the lake with the wind howling around the canyons.

So at 8 AM, I was done for the day.

Two hours later, I was heading toward Bryce Canyon. But that’s another story.

Wickenburg to Seattle by Helicopter: Day 1

Wickenburg to Page, AZ.

Regular readers of this blog who don’t follow me on Twitter might have been wondering where I’ve been. Did I fall off the face of the earth? Or finally, after six years, get tired of blogging?

Neither. I was making my annual helicopter repositioning flight to Washington State.

This year, I got an early start, piggy-backing my long cross-country flight at the end of a photo flight at Lake Powell. A photographer was willing to pay for the 4-hour round trip ferry time for me to get the helicopter from the Phoenix area to Page, AZ. At the end of that flight, I continued north to Salt Lake City instead of heading home. This put me several hours closer to my destination. At Salt Lake City, I picked up Jason, a low-time CFII interested in building R44 time for much less than the cost of renting. With Jason at the controls, we continued to Seattle.

The trip can easily be summarized by the number of days it took to complete. I’m putting it all down here, in four parts, while it’s still fresh in my mind. The photos aren’t terribly good due to glare through the bubble, but I hope the illustrate some of the terrain and weather we encountered.

In this first part, I’ll cover my trip from Wickenburg to Page, AZ. You can click here to see my approximate route on SkyVector.com.

I’ve made the trip from Wickenburg to Page (or Page to Wickenburg) countless times. It’s the kind of trip that I don’t even need to consult a chart to complete. I know the landmarks by heart.

But on Thursday, the weather promised to be a factor. Although it was sunny down in Wickenburg as I preflighted around 2:30 PM, the clouds were building to the north. I could see them thickening over the Weaver Mountains 15 miles away. And all the forecasts for all the points north of the Weavers called for high winds gusting into the 30s. It would be a bumpy ride.

So bumpy, in fact, that my friend Don and his wife decided not to join me on my trip to Page. Don’s got a helicopter very much like mine and he’d planned to fly up there with me, spend two nights, and let me show him around Lake Powell between my photo flights. But with the forecast so nasty, he bowed out. I didn’t blame him. No one wants to spend 2 hours getting thrown around the sky in a relatively tiny bubble of metal, Fiberglas, and Plexiglas.

And if wind wasn’t enough of a deterrent, the forecast also called for isolated showers and thundershowers north of I-40. So I knew I’d be dodging weather, too.

But I had a contract and my client had paid me to fly up there. I had a pilot waiting for me in Salt Lake City for a Saturday departure. The weather would have to be impossible to fly through to prevent me from making the flight.

So at 3:30 PM, I took off from Wickenburg (E25) into a 15 mph wind from the west and turned out to the north.

Wickenburg RanchI climbed steadily at about 200-300 feet per minute, gaining altitude slowly to clear the 5,000 foot Weaver Mountains ahead of me. Below me, I could see the scars the near-bankrupt developers had left on the desert where Routes 93 and 89 split off. Greed had scraped the desert clean, built a golf course, and then let the grass wither and die. Where there was once pristine rolling hills studded with cacti and small desert trees, there was now flattened dirt, void of vegetation, shaped by bulldozers and men. A dust bowl on windy days covering hundreds of acres of Sonoran desert.

I continued to climb, looking out at the Weaver Mountains ahead of me. The clouds were low over the mountain tops and I could clearly see patches of rain falling. The wind was moving the weather along at a remarkable pace. I picked a spot to cross the mountains, preferring the place where Route 89 climbs into Yarnell over the more direct crossing at the ghost town of Stanton and the valley beyond it. I knew from experience that the wind would be setting up some wicked turbulence in that valley as it gusted over Rich Hill and Antelope Peak. I braced for the turbulence I expected as I topped the hill at 5,000 feet MSL and was surprised when I wasn’t blasted.

That’s not to say there wasn’t turbulence. There was. But it was the annoying kind that bounces you around every once in a while just for the hell of it. The kind that makes flying unpleasant but not intolerable. The kind pilots just deal with.

Ahead of me was Peeples Valley, which was remarkably green. Our winter and spring rains had fallen as snow up there and it wasn’t until the warm weather began arriving that the grass could start sucking it up. The result was a carpet of new green grass that made good eating for the open range cattle and horses up there. All it needed was a little sun to give the illusion of irrigated pasture. But the sun was spotty, coming through breaks in low-hanging cumulous clouds.

Peeple's Valley to KirklandThe weather up ahead gave me a good idea of what I’d be facing for much of the trip: a never-ending series of isolated rain and snow showers. They appeared as low clouds with hanging tendrils of wispy precipitation. But unlike the gray rains hanging below summer rainclouds, these were white, making me wonder whether I was looking at rain or snow. With outside air temperature around 4°C (40°F), it could have been either. Or something worse; damaging hail or icy sleet.

I’d been taught at the Grand Canyon that if you can see through it, you can fly through it. But I didn’t think that rule applied to late spring storms at high elevations. I wasn’t going to fly through anything I didn’t have to.

Knowing which way the storms were moving made it easy to skirt around their back sides. Up near Kirkland, this put me several more miles west of my intended course. Before taking off, I’d punched the waypoint to our property at Howard Mesa into my GPS; I always fly over anytime I’m close, just to make sure everything is okay. Having flown the route dozens of times, I should be flying much closer to Granite Mountain. But that mountain was completely socked in by one of the storms, so I passed to the west of it, adjusted my course line with the push of two buttons, and continued northeast.

Near the Drake VORNear the Drake VOR, I detoured more to the east to avoid a rapidly approaching shower. Raindrops fell on my cockpit bubble and the 110 wind of my airspeed whisked them away. The sky was clearer ahead of me, although the tops of Bill Williams Mountain was still shrouded in clouds. The Prescott (KPRC) ATIS reported mountain obscuration and snow showers to the north, east, and west. I couldn’t see the San Francisco Peaks, which were likely getting more snow to extend the skiing season at the Snow Bowl.

Bill Williams MountainThen I was back out in the sun — a good thing, since the outside temperature had dropped to just over freezing and my cabin heat wasn’t able to keep up with the cold. I climbed up the Mongollon Rim just west of Bill Williams Mountain, trading high desert scrub for ponderosa pines. The mountain had recently been dusted with fresh snow. That didn’t surprise me; only three hours before, the airport at Williams (KCMR) had been reporting 1/4 mile visibility.

By now, the turbulence had become a minor nuisance that didn’t bother me much. I was listening to a genius mix on my iPod, hearing songs I didn’t even know I owned and trying to enjoy the flight. I was almost an hour into it and had more than an hour to go.

I reached Howard Mesa and flew over our place. Everything looked fine. I was surprised to see the wind sock hanging almost limp. Surely there was more wind than that.

I punched the next waypoint into my GPS; a point on the far east end of Grand Canyon’s restricted airspace. I wasn’t allowed to overfly the Grand Canyon below 10,500 feet. Since my helicopter starts rattling like a jalopy on a dirt road over 9,500 feet, that was not an option. Besides, one look out in that direction told me that no one would be flying anywhere near the Grand Canyon that day. All I could see to the north was a blanket of low clouds. I couldn’t even see Red Butte, a distinct rock formation that can normally be seen from 50 miles away. The Grand Canyon (KGCN) ATIS confirmed that things were iffy. The recording reported “rapidly changing conditions” and instructed pilots to call the tower for current conditions. You don’t hear that too often on an ATIS recording.

My course would take me east of that area, but the weather was also moving east. It soon became apparent that I was in a race against the storm. I was halfway to my waypoint when I realized that I wasn’t going to be able to go that way. Although there was a gap there between two storms, I knew from experience that gaps can disappear quickly, swallowing up whatever naive pilot slipped inside. The temperature had dropped down to 0°C and the clouds were only 300 feet above me as I zipped across the high desert, 500 feet off the ground. I’d have to detour to the east.

I aimed for the leading edge of the storm, hoping I could reach it and go around it. But the leading edge was racing eastward to cut me off. My course kept drifting eastward until I was heading due east. That would put me, eventually, over the Navajo and Hopi Reservations, far from any major road or town. I didn’t want to go that way.

I came down off the Coconino Plateau just southwest of Cameron. At least that’s where I figured I was. The low-hanging clouds had blocked all of my normal landmarks from view. To the north, where I needed to go, was a solid sheet of gray rainfall, blocking out whatever lay beyond it. As I descended from the plateau, still heading east, I began thinking of making a precautionary landing and waiting out the storm.

Then I saw a break in the storm with bright sunlight beyond it. It was still raining there, but I could clearly see my way through and what I saw looked pretty good. I banked to the north and entered the rainstorm. Soon, I was being pelted by rain. Visibility was still tolerable; I could see well enough to fly. Thankfully, there were no downdrafts to contend with. Just turbulence, rocking me around, punishing me for interfering with nature’s gift of rain. I held on and rode it out.

And that’s when the carbon monoxide detector light went on. On departure, I thought the heat had smelled a little more like engine exhaust than usual, but had put it out of my mind. The warning light brought it right to the front of my mind again. I opened my door vent and the main vent and pushed the heater control to the off position. I took stock of the way I felt: any symptoms of carbon monoxide poisoning? No, I felt fine. But soon I’d feel very cold.

Echo CliffsI passed through the rain and emerged on the other side with a clean cockpit bubble and likely very clean rotor blades. Ahead of me, now due north again, the sky was brighter and I could clearly see the Echo Cliffs. Before entering the storm, I’d punched Tuba City (T03) into my GPS to keep my bearings; now I punched in Page (KPGA) and was pleased to see that I was already right on course. I aimed for The Gap, a small town at the gap in the cliffs, right on Route 89, adjusted power to maximize speed, and sped forward, 500 feet off the high desert floor.

The rest of the trip isn’t very interesting. I did get some different views to the west, where the Grand Canyon was still covered by a thick blanket of clouds. It would be snowing there, especially over the North Rim. My usual path was at least 20 miles to the west, much closer to the Canyon. It included overflying the Little Colorado River Gorge and mile after mile of nearly deserted flatlands on the far west edge of the Navajo Indian Reservation. I was still over the reservation on this path, but the land below me was sculpted by wind and water into mildly interesting patterns. This was the western edge of the Painted Desert, which is not quite as picturesque as most people think.

The GapI crossed Highway 89 at The Gap and flew through the gap in the Echo Cliffs. I was now about 45 miles from Page, flying among three sets of high tension power lines that stretched from the Navajo Generating Station on Lake Powell to points south. There was a dirt road here that made a short cut to page — if you didn’t mind driving more than 40 miles on a dirt road. Navajo homesteads were scattered about. The sky was a mixture of clouds and patches of deep blue. I warmed in the sunny spots and cooled in the shadows.

It was after 5 PM and I was starting to worry about reaching Page in time to pick up my rental car at 5:30. I began making calls to the FBO from 25 miles out. No answer. A while later, another plane called in from the northwest. Other than that, silence.

I was fifteen miles out when Lake Powell came into view. It looked gray and angry under mostly cloudy skies.

I landed on the taxiway parallel to runway 33 and went right to parking. A line guy from American Aviation arrived with a golf cart to pick me up. I shut down and jumped in. It was 5:30 PM; the 200 NM flight had taken almost exactly 2 hours. I was just in time to get my rental car.

Autorotation is Not a Low Rotor RPM Emergency Procedure

Especially when you’re two miles out at sea.

Picture this: An R22 helicopter without floats operating two miles off the coast of Miami, FL. On board is the CFI-rated pilot with 600 790 hours of total flight time and the private pilot rated “passenger” with 115 hours total flight time. They’re operating at about 40 knots 100 feet above the waves on an aerial photo mission, photographing boats. The wind in Miami, 13 miles away, is from 120 at 13 knots and it’s 26°C with a dew point of 21°C, resulting in a balmy 74% humidity.

The pilot had just completed a 180° turn to the south when the low rotor RPM horn sounds.

The pilot adjusts the throttle to compensate — in other words, we should assume that he adds throttle. The horn stops blaring, but 3 seconds later, it does it again.

So what does the pilot do? Despite the fact that the helicopter does not have floats, he enters an autorotation. The helicopter crash-lands in the ocean, the occupants escape, and the helicopter sinks. The pilots are rescued 10 minutes later by a privately owned boat. The helicopter is left unrecovered (so far) in 150-250 feet of seawater.

What We Don’t Know

There are a few things we don’t know that could explain the reason for the low rotor RPM horn:

  • How much did the pilots and their equipment weight? An R22 Beta (not Beta II) is a very small helicopter. Although they had burned off 45 minutes of fuel, there is a possibility that they were still heavy for the flight conditions.
  • Which direction did they turn? A turn that would have put them into a tailwind situation — especially at low speed — could rob them of airspeed. If airspeed dropped below ETL, the helicopter would have to work harder to stay in the air.
  • What speed were they operating at? Without the benefit of forward airspeed and effective translational lift, the helicopter would have to work harder to stay in the air. If the speed was close to zero, the aircraft might have gotten into a settling with power situation. The natural (but incorrect) reaction of increasing the collective to arrest the rate of descent could have triggered a low rotor RPM warning if available power was exceeded.
  • Were the engine and its components functioning properly? If the engine or magnetos were not performing to specifications, the resulting reduction of engine power could cause a low rotor RPM horn. We have to assume the engine was still running because the NTSB report didn’t mention an engine failure.

But regardless of the reason for the low rotor RPM horn, it’s the pilot’s decision to perform an autorotation to into the ocean that needs to be questioned.

The Robinson Low Rotor Horn

In a Robinson helicopter, the rotor RPM green arc is 101% to 104%. (Please don’t ask why; I don’t know. Yes, it is weird.) The low rotor RPM warning system is designed to alert the pilot at 97% RPM. (See it in action for yourself here.) This is a very early warning. The idea is that if rotor RPM is deteriorating, once it gets past a certain point, it could could become unrecoverable very quickly. The earlier the pilot is warned, the better off he is.

At the Robinson factory safety course — and, one might assume, at many flight schools that train in Robinsons — pilots are taught that a Robinson can generally fly at an RPM of 80% plus 1% per 1000 feet of density altitude. Given the temperature, dew point, altitude, and altimeter setting (30.01), the density altitude was 1,612 feet. That means that the helicopter should have been capable of flight when operating at only 82% RPM.

I need to stress here that this is a general rule of thumb. Do not attempt to fly around at low rotor RPM to test this. While it’s true that my flight instructor at the Robinson safety course had me fly for a few minutes in the Long Beach, CA area at 90% RPM with the horn blaring just to prove that flight was possible, RPM is not something we play with in non-training situations. The formula is simple: RPM = life.

Low Rotor RPM Emergency Procedures

The Robinson R22 Pilot’s Operating Handbook is quite specific on what to do in the event of a low rotor RPM warning. On page 3-10, in the red-tabbed “Emergency Procedures” section, it states:

A horn and an illuminated caution light indicates that rotor RPM may be below safe limits. To restore RPM, immediately roll throttle on, lower collective and, in forward flight, apply aft cyclic.

The NTSB report indicates that the pilot initially “adjusted the throttle to compensate for the [low rotor RPM warning] condition” and was immediately rewarded with recovery. But that was followed by the horn sounding again only 3 seconds later.

It had to be scary for the pilot. After all, he’s only 100 feet above the water and he’s supposed to react by lowering the collective. But the emergency procedure and repetitive training doesn’t tell us to enter an autorotation, which would be a full-down reduction of the collective. The reduction of the collective, coordinated with the rolling on of the throttle, should be slight — perhaps an inch or so. This reduces drag on the blades while the increased throttle provides power to increase their RPM.

What Was the RPM?

One of the things we don’t know is what the RPM was when the pilot decided to enter autorotation. If it had deteriorated to the point where autorotation and cyclic flare were the only tools to recover RPM, his decision was probably a good one. Better to hit the water relatively softly than from 100 feet up, falling like a brick.

If RPM had deteriorated to that point that quickly, however, it’s important to recover the aircraft to learn why. Other than a complete engine failure — which was not mentioned in the report — it’s hard to imagine what would cause RPM to drop enough to warrant such a drastic recovery action.

Who Was Flying?

There may be more to this than what meets the eye.

The helicopter was operated by Helicopter Academy, a flight school with locations across the U.S. The school’s Web site clearly advertises it as a low-cost training company:

$250 PER HOUR R22 HELICOPTER TRAINING TIME BARGAIN and we are the ONLY company in the world that can guarantee you a job.  We operate a fleet of helicopters and like other schools our insurance requires 300 hours helicopter time and an instructor’s rating to fly for us. We train you to work for us and offer a job to all graduates, including transfer student and instructors who can’t get jobs elsewhere.

Helicopter Academy’s other business is BoatPix, which uses helicopters to photograph boats and then sells the photos to the boat owners and others. It’s widely known that BoatPix pilots pay BoatPix (or Helicopter Academy) for the time they fly aerial photo missions. The company’s Web site alludes to this:

…you pay for the first 100 hours at $250/hr, the second 100 hours at $200/hr and the third 100 hours at $150/hr….It’s  $25,000 for the first 100 hours where you’ll do mostly training, $20,000 for hours 100  through 200 where we’ll introduce you to our photo contract which will subsidize your flying and $15,000 for hours 200 through 300 where you’ll do almost exclusively photo and will learn this skill that is valuable to our photo contract and making you a valuable pilot to us.

I added the emphasis in the above quote. It begs the question: who was actually flying this aircraft? The NTSB report suggests that it was the 600791-hour CFI. But was that really the case? Was the 115-hour private pilot paying $200/hour to be “introduced” to the photo contract — as a pilot — while the 600791-hour CFI took the photos?

High Risk Operations

In March 1999, Robinson Helicopter issued Safety Notice SN-34. The latest version of this Safety Notice is dated April 2009. Titled “Aerial Survey and Photo Flights – Very High Risk,” it starts out saying:

There is a misconception that aerial survey and photo flights can be flown safely by low time pilots. Not true. There have been numerous fatal accidents during aerial survey and photo flights, including several involving Robinson helicopters.

It goes on to list some of the possible dangers of low time pilots conducting aerial photo flights. It also makes some recommendations for minimum requirements for aerial photo/survey pilots, including a minimum of 500 hours pilot-in-command. BoatPix is one of the operations that has chosen to ignore this recommendation.

My question to helicopter pilot wannabes out there: Are you that desperate to become a pilot that you’re willing to trade your safety for flight time?

Pilot Experience and Decision-Making

What it all comes down to is whether the pilot made the correct decision for the situation he found himself in. I’m not convinced that entering autorotation over the ocean on hearing a low rotor RPM warning horn is the correct decision.

True, both pilots walked (or perhaps I should say, swam) away. But if the rotor RPM could have been brought back into the green while in flight — something a well-trained or experienced pilot could have accomplished if there wasn’t a mechanical problem — the watery autorotation and the resulting loss of the aircraft could have been avoided.

Hopefully, the Probable Cause report for this accident will shed some light on what really happened. Until then, it certainly gives pilots some food for thought.

November 1, 2011 Update: The Probable Cause report doesn’t add much to what’s reported here other than to clarify airspeed and PIC experience. The official Probable Cause is “A loss of main rotor rpm for undetermined reasons.”

Update, March 17, 2012: Just found another accident report with someone else using autorotation as a cure for low rotor RPM. He crashed, too.