Having flown many thousands of hours in the clouds of the Northeast, I used to believe that I had all the answers in dealing with structural ice. I admit to being very aggressive, flying into conditions that were likely to produce ice. My experience had convinced me that ice usually formed rather slowly and that changing altitude just a couple thousand feet would take care of any icing problems.
That attitude changed one day in early May of 2000 when I took off from Utica, New York (KUCA, no longer an active airport) for a flight to Nashua, New Hampshire (KASH) in a Piper Seneca, which did not have any ice protection. I was alone in the airplane with no baggage except a duffel bag and a laptop case. Even with full fuel, I was well below the maximum allowable gross weight and should have been able to climb very well.
The weather had us on the backside of a cold front, and there were a few cumulus clouds protruding through the low layer. There was no specific mention of icing in the forecast other than the usual phrase about icing in clouds and in precipitation above the freezing level. The freezing level was at the surface and the ceiling was 400 feet. I could see breaks in the clouds and I expected to climb through the low overcast and be in th
e clear very quickly after departure. It was clear all the way from Albany, about 80 miles east, to my destination, and I was expecting a 40-knot tailwind at altitude.
I departed IFR on Runway 27 into a 30 knot surface wind right down the runway and was cleared direct to the Utica VOR where I could intercept the airway. As I turned toward the southeast the cabin got dark and I realized I had found one of those embedded cumulus clouds and was climbing through it. I expected to break out of the side of it at any moment as I climbed through 4,000 feet. As I scanned my instruments, I couldn’t believe what I was seeing. The VSI needle, which had been comfortably indicating a 1,500 feet/minute climb, was dipping rapidly toward zero. It kept going right past the zero mark and quickly settled on at 1,200 feet/minute down. A quick look at the altimeter confirmed the bad news; I was no longer climbing but beginning to descend.
The airspeed indicator showed that my airspeed was also decaying. I had both engines operating normally at cruise-climb power with the props set accordingly. Since the airplane seemed to be handling normally, I suspected a pitot-static problem. Having anticipated flight in clouds above the freezing level, I already had the pitot heat on.
I opened the alternate static source thinking maybe the static port had iced over. The instrument indications remained about the same. I lowered the nose to maintain airspeed while simultaneously pushing a large handful of throttles and prop controls forward and pitching to maintain the blue line on the airspeed indicator. The blue arc shows best single-engine-climb speed and is also usually considered to be the best airspeed to achieve maximum climb, even with both engines operating. The increased power and best-climb airspeed had very little effect on the airplane performance and certainly didn’t solve my problem. The power output of the engines appeared to be normal, but I opened both alternate air sources as a precaution.
I finally had time to glance up at the windshield and observed a little ice on the lower portions. The next instrument scan showed that I was descending at 1,500 feet/minute through 3,500 feet. I was still at full power at best-climb airspeed in a relatively light airplane! A glance out the left side window revealed that there might be some ice on the leading edge, but I couldn’t be sure. Ice on a white leading edge inside a cloud is very difficult to detect.
This was getting interesting.
I immediately started a turn back in the general vicinity of the departure airport and advised Griffiss Approach that I was having a problem and needed to return. The strong wind favored Runway 27, but because of the 400-foot ceiling, the controller gave me a vector to intercept the localizer for an ILS approach to Runway 33. He instructed me to descend to 3,000 feet. At about that time I was descending out of about 3,300 feet and coming down now at about 1,100 fpm. I advised him that I would be unable to stop my descent at 3,000 feet. He said the minimum vectoring altitude for my area was 2,600 feet and he could not guarantee terrain clearance below that altitude.
Fortunately, I grew up just a few miles from the airport and had done extensive flying in that area, so I was familiar with the terrain. I knew there were no obstacles or high terrain between me and the airport. The lower altitude helped the airplane perform a little better and the ice had apparently stopped forming, though none of it seemed to be leaving yet. The descent rate was now about 300 fpm at full power. That nice tailwind that I was anticipating for my flight to Nashua was now a serious headwind as I struggled to get back to the airport.
Griffiss cleared me for the approach and told me to contact Utica Tower. The approach controller started the usual spiel about maintaining altitude until established and then thought better of it. He simply told me to do the best that I could.
The vectors provided by Griffiss Approach were excellent, having me intercept the localizer just as the glide slope needle reached the center of the indicator and the marker beacon receiver showed outer marker passage. I turned inbound and still needed full power to barely maintain the glide slope. I was able to keep the glide slope indicator on scale, but I couldn’t do any better than to maintain the needle at two dots above the center reference. That is just barely within acceptable limits. I also had to deal with a substantial crosswind since I was flying the localizer for Runway 33 and the wind was 30 knots from 270 degrees.
Normal procedure calls for extending the landing gear at the outer marker, but I realized that the last thing I needed was more drag. I elected to leave the gear up for now. I broke out of the overcast at about 1,200 feet MSL which put me about 450 feet AGL. I realized for the first time since the ordeal began that I was going to make it to the runway. I accepted the fact that I might be landing gear-up, but I also realized that sliding to a stop on a concrete runway was much better than careening through a farmer’s field or a stand of trees.
As I broke out of the overcast I was still maintaining blue-line airspeed. I had thought about slowing the airplane a little as I got closer to the runway, but decided to maintain the higher airspeed for two reasons. First, the airplane was flying fairly well at that speed and I didn’t want to do anything to change that. Second, I wasn’t sure what the stalling speed was now with all the ice on the wings. Once I was in the clear and past the runway threshold, I was able to trade some airspeed for lift and begin a level flight down the runway. I grabbed the gear lever and selected the down position.
At first, nothing happened, probably because of the ice on the bottom of the airplane. After what seemed like an eternity but was probably only about five seconds, I heard a noise that is not typical of gear extension in a Seneca, followed by the sound of the gear extending. Finally, the last of the three green gear indicator lights blinked on. I backed the power off and the tires made contact with the runway about a thousand feet from the landing threshold with plenty of room to stop. What a relief to be on the ground, uninjured and in an undamaged airplane!
There was surprisingly little ice on the windshield and I was able to taxi to the parking area without assistance. I noticed that I needed substantially more power to taxi than I was accustomed to. I resisted the urge to see how much ice was on the leading edges, concentrating on taxiing to the ramp. I would have felt really foolish if I had just come through this harrowing flight unscathed only to clip the lights or a parked airplane.
When I shut down and climbed out of the airplane, I was amazed at what I saw. The ice had formed a flat plate about two inches high all along the leading edges of the wings. I later found out that this is an unusual formation called an “ice horn.” It not only severely disrupts the airflow but it greatly increases form drag. As big a problem as the leading edge ice was, the props were even more startling.
The blades were no longer shaped like airfoils but more closely resembled baseball bats. I was amazed that I’d been able to develop as much thrust as I had, and I immediately understood why it took so much power to taxi to the ramp. Virtually none of the ice had come off the props or the rest of the airplane.
Having been involved in aviation education for many years, I wanted desperately to get pictures for use in training. I almost always carried a digital camera with me, but not on that day. I ran into the FBO looking for someone with a camera, but it was not to be. One of the line service people walked out to the airplane with me to take a look. The first drop of water was dripping from a prop blade. The temperature had increased slightly and the ice was melting. I had just lived through the icing adventure of a lifetime and was to have no pictures of the evidence.
I learned a valuable lesson on that day; that my self-perceived skill in dealing with structural icing for the past 30 years had been laced with a large dose of luck. Of the many times that I had ventured into conditions that had a high probability of producing ice, I had never encountered anything like this.
What really got my attention was that nothing indicated ahead of time that this flight might encounter serious structural icing. I learned that airframe ice can be unpredictable and unforgiving. I have become much more conservative in my fly/no-fly decisions when there is any possibility of encountering structural ice.
-Gene Benson
