FAA Advice on Icing: Incomplete At Best

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

    • A
    • A
    • A

From the doesn't-everyone-know-this-#$%$ file comes this nugget this week: The FAA says icing may affect stall warning systems on older aircraft. Well, duh. Unfortunately, the reality is that everyone doesn't know this stuff, so the point is, sadly, well made and well taken. I'll get to what people don't know in a moment, but first some observations on icing accidents.

I don't think there are as many as there used to be. Back in the early to mid-1990s, we could reliably mow a lot of editorial hay out of icing accidents. A sweep of the NTSB data would reveal about 20 to 25 each year, half of which were fatal and most of which resulted from naked stupidity not necessarily related to the decision to launch into ice in the first place. Now, when sweeping accidents, a honest-to-goodness structural icing prang is harder to find. My guess is there are several reasons for this.

First, there's less flying activity and less of that flying is done in challenging winter weather. Second, there are more de-icing systems out there, especially TKS. And third, icing forecast and reporting has gotten better and although it's not as significant a direct help with ice as it is with thunderstorms, datalinked weather to the cockpit has proven a real benefit to weather awareness. So, that's my theory. I'm open to other ideas.

So, the FAA says the stall warning system might not work right, so watch it. Fine. On the other hand, if you're iced up and thinking the stall warning system has anything to do with your survival prospects, you may be too stupid to have tied your shoes on the way to the airport. Which is another way of saying there a bunch of things you need to be thinking about to survive and the stall warning is the least of them.

Now, the don't-know part. Some years ago, I was an instructor in a flying club and one of the club's pilots got into a spring icing event in upstate New York. Prudently, he diverted into the closest airport and flew an approach under high ceilings and good visibility, although with a lot of ice on the airframe. A little gentle probing revealed that he flew a normal approach, at the normal airspeed and with flaps. He was not aware that icing would alter the airfoil enough to likely raise the stall speed by some significant margin and that he should avoid use of flaps. But he learned the lesson the hard way. The airplane stalled above the runway and hit hard enough to damage gear components and bend the prop. It was an expensive repair.

I don't know why he didn't know to fly a faster approach. Or maybe he did know but in a state of mild panic, he just forgot. (He wasn't the first; the accident record reveals many examples of the same scenario.) The FAA's Special Airworthiness Information Bulletin (PDF) this week sort of addresses this.

But I'm not impressed with its advice, frankly, for a couple of reasons. One, it says that the POH should be consulted for "limitations and procedures" for flight in icing. For airplanes not approved for icing, that will be to simply say "flight in icing not approved" and if you get into it, good luck. Even the flight supplement for the TKS-equipped Cirrus SR22 doesn't go into great detail on this subject, although it does advise against the use of flaps in icing conditions, although this isn't much of a worry if the airplane is clean.

I find several things troubling about the shallowness of the FAA SAIB. It only mentions the flap issue in passing and refers to POH guidance which probably doesn't exist. Second, it completely ignores the tailplane icing and stall issue, something that has been well documented in flight tests if not in accident histories. (The flying club accident mentioned above could have been a tail stall.)

The SAIB recommends increasing airspeed by 25 percent in all phases of flight while in icing. Further, it argues treating a perceived buffet as an impending stall and immediately lowering the angle of attack. This would be precisely the wrong response to a tailplane stall. Yet the SAIB doesn't address this. It should.

Here's a far more complete and, in my opinion, well-reasoned approach to surviving with an iced up airframe. I think the best advice is to land a little faster, on the longest runway you can find and with no flaps. You will haveóor should haveódone no-flap landings in your training.

This is one place to put that training to work.

Comments (25)

I agree 100%. I too let out an audible "duh" when I saw the title of the warning.

Posted by: Mark Fraser | January 26, 2011 12:08 PM    Report this comment

Yeah when I first saw the paper's title I thought 'Wait, they have to mean something else.' Expecting something about how icing permanently affects the warning system or just about anything else. But then I read it.

Posted by: Corey Phelps | January 26, 2011 10:01 PM    Report this comment

I *still* teach at an Upstate NY flying club! and I concur on your findings. First there are fewer flying in general and those that launch into winter icing fun usually have at least basic TKS now. This does not stop people from scaring themselves and "learning the hard way." A Mooney crashed up here two days before Christmas (doctor commuting to work; leaving a wife and four daughters) and one of our members came in at the very same time with his TKS Mooney. Darwin is still at work culling the incautious. It is true that though this FAA info seems obvious to us it is necessary warning (and legal disclaimer) to the new cadre of "digital pilots" with those truly amazing white plastic steads that fly you anywhere ("direct please") with the push of a button.

Posted by: David St. George | January 27, 2011 6:04 AM    Report this comment

You hit it right on the mark, Paul. I suppose this can serve as a call to flight instructors to ensure our students really understand this issue, beyond the basics.

Posted by: Meredith Tcherniavsky | January 27, 2011 7:30 AM    Report this comment

A young couple died at Christmas at Colorado Springs, on the second attempt in minimum weather that undoubtedly (due to the temp, moisture, etc.) included ice on their older Mooney. He was a highly trained B-1 pilot from Ellsworth AFB, but no GA airplane without FIKI certification is a predictable airplane in actual icing--or maybe it is predictable, that it will be in deep trouble, no matter how sophisticated its panel may be.

I think the lesson that the FAA and instructors have to push really hard is that icing is a killer, and it's only a little less deadly if the airplane is a light single certified for FIKI.


Posted by: Cary Alburn | January 27, 2011 9:34 AM    Report this comment

only a little less deadly if the airplane is a light single certified for FIKI.<<

How ya figure?

Posted by: Paul Bertorelli | January 27, 2011 9:46 AM    Report this comment

I saw the title of the email from the FAA and didn't actually read it. I've encountered light ice on several occasions when the freezing level was lower than forecast, and the stall warning wasn't on my list of "much pay close attention to" items. What I noticed every time was how badly performance was degraded by even a trace of ice, airspeed dropping significantly for a given power setting due to the drag. (In all cases, I had warm air and low terrain below, so it wasn't life threatening.) It was hard to believe the plane would still fly with the "inch of ice" I keep hearing about people carrying on their 206 and still able to climb . . .

Posted by: David Chuljian | January 27, 2011 11:00 AM    Report this comment

I'm not sure that simply teaching "icing is a killer" is the way to combat icing accidents. While I would certainly not recommend any flight in known icing with a non-FIKI airplane, most pilots who fly IFR long enough will inadvertently encounter airframe icing. In the vast majority of situations, it is possible to stop the accident chain provided the pilot acts quickly, decisively and correctly - which can include landing short of your destination, making your approach count because you may be unable to go missed, carrying extra airspeed and not using flaps, gradual s-turns on final to see around a frosted over windshield, etc. We also need to teach proper decision making technique to allow students to interpret the weather and have an escape route (two or three are better) in case of an inadvertent encounter with ice. Ironically, one of the dicier encounters I had with ice was in VFR conditions and no precip was in the forecast.

Posted by: Josh Johnson | January 29, 2011 9:19 PM    Report this comment

Light singles and light twins are not weather warriors, and a FIKI system on either of these is there to help you get out of ice ASAP, not press on in bad conditions. I don't care what the FAA says on ice bridging, I have seen it and more than a couple times. I have also lost a vaccum pump in the ice more than once. Then you are stuck in a bad situation wondering if you will loose the other pump the next time you cycle the boots. Many pilots don't realize how much cycling the boots wears a vaccum pump. Light airplanes with FIKI are marginal in moderate ice conditions, and need to be treated that way to keep from killing their pilots. I agree that the light FIKI singles are only a little less deadly than their non FIKI equivilants. If pilots are not smart enough to leave themselves and use an out that does not rely on boots or tks they are asking for trouble.

Posted by: Barrett Roessler | January 30, 2011 1:38 PM    Report this comment

I invented tailplane icing! Well, not exactly, but working as an ASI for the NTSB in the late '80s and '90s had the duty of investigating two YS-11 events, one fatal and one with no injuries, both at West Lafayette, IN. To keep the story short, tailplane icing was discovered in both instances. As a result the FAA changed icing certification requirements to include tailplane icing wihch it had not previously considered!

YS-11 addicent 3/15/89... http://www.ntsb.gov/ntsb/brief.asp?ev_id=20001213X27867&key=1

YS-11 incident 4/3/90... http://www.ntsb.gov/ntsb/brief.asp?ev_id=20001212X22877&key=1

Although I was not directly involved with the investigation, there was a fatal BAE-3101 accident at Pasco, WA on 12/16/89, with a similar probable cause. This was due to the recent investigations of the YS-11 events... http://www.ntsb.gov/ntsb/brief.asp?ev_id=20001213X29944&key=1

As a further result, the FAA was looking at a previous Convair 240/340 accident at Boston, MA. It is difficult to know how many accidents have been the result of tailplane icing. I agree the SAIB leaves a lot to be desired considering the amount of time spent and the little new information contained in it.

Posted by: Steve Wilson | January 31, 2011 8:46 AM    Report this comment

Josh Johnson's comment reminds me of an Angel Flight mission I flew a few weeks ago in a 172SP. The mission assistant was IFR and MEI rated and had a commercial license, but had only 250 hours as PIC. She was also working on becoming a CFII.

Conditions were hovering around freezing at altitude, and possibly on the ground. I flew the mission leg above the clouds IFR, and then had to descend through the clouds to land at the destination. Pilot reports indicated icing in the clouds. During the IFR approach, I descended, watching for visible ice, but did not see any, although water drops appeared on the airframe and the outside temp was zero. We broke out at about 100 feet above DA with the runway in sight. I intended to land at 90 knots and no flaps - when my assistant suggested I slow down and extend the flaps.

I said, "No. We have sufficient runway length and I have extensive experience with high speed landings in this aircraft." She somewhat freaked out, but we had an opportunity to discuss what was not taught in school. This knowledge comes with self-education, such as these forums and articles like this.

Posted by: Alan Hoffberg | January 31, 2011 9:09 AM    Report this comment

Icing and fog are the two things that scare me the most when flying because both can appear in an instant and get really, really bad in an amazingly short period of time.

I don't believe there is a single engine airplane today that is truly capable of FIKI. I do believe, however, that TKS systems give pilots a false sense of security. The only safe way to deal with icing is to immediately get out of it. You don't have the luxury of staying around and waiting to see how bad it can get because it can be bad enough in a very short amount of time that further flight is impossible. When you see ice move immediately. This is one of the occasions that you don't ask ATC, you tell them. Tell them you must manuever to escape icing conditions and ask them which way they would like you to go. There may also be some pireps from other aircraft as to temperatures/icing at other altitudes. (Be careful with PIREPS, however. Those from aircraft similar to your own and near your area are likely to be more accurate.) Make a turn off the airway (or what ever course guidance you are following) and begin a climb or descent depending upon OATs which all pilots should monitor when there is the tiniest chance of icing.

Luckily, I have only seen severe icing twice in a 36-year career, but that was enough. Both times I was on an approach and made it to the runway under climb power.

The only ice I'm comfortable is that clinking around in my glass of Scotch!

Posted by: Linda Pendleton | January 31, 2011 10:49 AM    Report this comment

On a Wx Recc Misson over the South China Sea in 1973 I was the Co-pilot in a USAF KC-135 ie. B707 with certification in known icing, heated leading edges, heated inlet nacelles from bleed air. With 4 engines at normal cruise power and anti-ice activated approx. 2 Hrs. prior at FL280 we encountered Ice so intense 3 of our 4 engines flamed-out and we stalled at 300Kts Indicated and recovered at 4K feet. All anti-ice systems were later found to be functioning normally. So don't think it can't Happen To You! After floundering around at low level for 5-8min. we began shedding what appeared to be approx.4in. thick Ice and got 2 engines re-started and climbed up to 14000ft. and did a 180. Then declared a No-Strike Day. Pretty sure tailplane ice caused the stall!! One of those Never forget Moments!! Respect Ice!

Posted by: Buz Allen | January 31, 2011 12:00 PM    Report this comment

On a Wx Recc Misson over the South China Sea in 1973 I was the Co-pilot in a USAF KC-135 ie. B707 with certification in known icing, heated leading edges, heated inlet nacelles from bleed air. With 4 engines at normal cruise power and anti-ice activated approx. 2 Hrs. prior at FL280 we encountered Ice so intense 3 of our 4 engines flamed-out and we stalled at 300Kts Indicated and recovered at 4K feet. All anti-ice systems were later found to be functioning normally. So don't think it can't Happen To You! After floundering around at low level for 5-8min. we began shedding what appeared to be approx.4in. thick Ice and got 2 engines re-started and climbed up to 14000ft. and did a 180. Then declared a No-Strike Day. Pretty sure tailplane ice caused the stall!! One of those Never forget Moments!! Respect Ice!

Posted by: Buz Allen | January 31, 2011 12:00 PM    Report this comment

I'm not sure that simply teaching "icing is a killer" is the way to combat icing accidents.<<

And this is the crux of the discussion. And it also gives a bit of a lie to the idea that known-ice equipped singles are somehow not up to the task. The TKS performance history strongly suggests otherwise.

The problem with the idea that all icing is a killer is that real world experience proves this not to be the case. And we don't help by repeating this mantra against everyday ops in which people deal with and survive icing.

The reality is that most icing events in light aircraft--ice-protected or not--are benign. You climb up through an icy layer, get on top, you descend into warmer air, you divert or...whatever.

Some small percentage of icing events will be unmanageable. No one knows what that percentage is, so when you decide to fly in clouds with the potential for icing, you assume a degree of risk. But everything in aviation has risk, yet you still do it. Icing is no different. The key to survival is having good outs available and to know when to say no. If you do, you will reduce but never eliminate the risk.

Posted by: Paul Bertorelli | February 1, 2011 5:01 AM    Report this comment

As for TKS not being effective and being little better than no protection at all, I've heard this before. It mostly comes from people who don't know much about TKS. We've done a number of owner surveys on the effectiveness of this system and this opinion is simply not supported by the people who have TKS.

We're doing another survey now. Here's a typical user comment:

"Only once in eight years of operation have I been uncomfortable with the system's ability to prevent ice accumulation. Always have a plan B and C. In this case, a 2000' altitude change was enough. Plan C, diversion to the nearest airport, was next."

You hear a lot of that sort of thing from TKS users. It's a very effective system. But it's not bulletproof. Nothing in aviation is. But it strikes me as uninformed to say an airplane with TKS is little better than one without it.

Posted by: Paul Bertorelli | February 1, 2011 5:06 AM    Report this comment

Paul - I totally agree with what you're saying. There is no true all weather airplane - not even a 747, but the more tools we have to deal with the weather the better. Can a light single with TKS handle everything a heavy jet can - heck no! But with careful preflight planning, I'd say it increases your ability to safely complete a winter mission in a given day from possibly 70% without it to around 90-95% with it.

Posted by: Josh Johnson | February 1, 2011 8:50 AM    Report this comment

In a 44 year career of mostly single-pilot freight only anxiety producing icing event happened early on in a lightly loaded, forward CG 560 Commander. Got buffet @ 90K on approach with very light ice on wing(?!)- years later discovered it was undoubtedly tail stall. Part of company investigation into smoking hole Caravan accident; light ice, forward CG, full flaps, sudden power application = tail stall(?!) Finally, like Paul, with plan A, plan B, and fall back on every flight.

Posted by: Stan DeLong | February 1, 2011 10:46 AM    Report this comment

Paul- Glad to see tailplane icing getting noticed.It can be most disconcerting to say the least. When you get into heavy ice,how you got there isn't as important as how you get out so you need another card to play.(The objective is to ALWAYS have another card to play). (1)Tailplane ice can make your bird unstable in pitch which can lead to ice behind the boots and belly ice. Try cracking your flaps just a little to increase downwash.Reduces tailplane stall. (2)Even with prop boots working tou can get blade ice. Try rapid movement of prop controls to throw it off.The noise it makes as the ice hits the nose will upset pax but you'll fly better so do it.(3)Test your bird at altitude if you can to see if you can use some flap on landing.(4)On landing,fly it on flat and hot. Tires 1" above blacktop before you reduce power. Works for me.

Posted by: Ron Robinson | February 2, 2011 1:05 PM    Report this comment

guys do you think that a heated stall warning vane would help?(something simple like a pitot tube heater)

Posted by: Peter Horvat | February 2, 2011 2:06 PM    Report this comment

Paul, you and I will have to agree to disagree. A light single with FIKI certification is akin to a 4WD pickup with chains and a winch. You don't use the chains and winch to get into the rough stuff--you use the chains and winch to get out of it. You don't use the deicing equipment to get into ice, but to get out of it--that is, if you want to live a long life. I don't doubt the effectiveness of TKS or other de-icing/ice prevention equipment, but as others here have said, even extraordinarily well equipped "big iron" can be brought down, because ice accretion is not all that predictable.

I prefer the attitude of the TKS owner you quoted whose backup plan is to exit the ice as soon as possible, using the TKS to minimize the problem in the process of exiting. To continue on would be fool-hardy, IMOH. There is no way of knowing how long the ice will last, or how quickly it will build. There is only so much TKS fluid aboard, and when it runs out, there goes the FIKI ability. Hot props can fail; there goes the FIKI ability. Heated windshields can fail; there goes the FIKI abiility. And induction icing can occur, causing significant power loss or complete engine failure, and all the FIKI systems available won't restore power.

I do not believe that any pilot flying a light single with FIKI should use it except to get out of unforecast icing.


Posted by: Cary Alburn | February 2, 2011 7:47 PM    Report this comment

Cary...it's a question of emphasis, really.

No one one commenting on this forum--including me--has suggested certified TKS is anything other that a reliable, safe means to find an ice-free altitude. And neither are any of the 50 or so TKS owners we have heard from.

Logically and factually, that's not the same as a FIKI airplane being little better than one with no ice protection. TKS is a significant safety enhancer and the accident record--or lack thereof--shows that owners seem to understand whatever limitations it has.

I can tell you this: Many owners with TKS will comfortably dispatch into forecast or reported icing that they wouldn't tackle in unprotected airplanes. So would I.

But that's not the same as departing into icing with no reliable plan B. Warmer air at the surface, known layers, good VFR on top or an inversion, and so on. It requires a higher level of both fine point judgment and risk tolerance.

The fine point is knowing that you probably wouldn't cross the Sierra with strong Pacific inflow and cold temperatures, whether TKS-equipped or not. But you likely would fly through much of winter conditions you see in the northeast or midwest.

What I reject is this: That all icing is the same and that the small risk of it becoming unmanageable neutralizes the benefits of TKS. If we teach people to never consider flight in icing, we are giving them an artificially conservative risk model that doesn't match the real world.

Posted by: Paul Bertorelli | February 3, 2011 6:30 AM    Report this comment

I recognize that some people want that conservative risk model and that's fine. But those who pick a less conservative approach are not wild-eyed loons, merely people who see things differently.

Posted by: Paul Bertorelli | February 3, 2011 6:30 AM    Report this comment


I guess, like Cary, you and I, too, will have to agree to disagree on this on subject. When I experienced icing I was flying freight in the Midwest every night. I suspect my exposure was a bit greater than the occassional TKS-equipped trip that most owners would make. I suspect that many of those successful trips through icing in a FIKI-equipped airplane never encountered more that trace or slight icing. I thought icing wasn't worth all the bruhaha, either, until I saw moderate for the first time. I then realiezed that I hadn't seen anything previously. Since once you encounter ice there is no way to know how bad it will get and how quickly it will get that way, I will continue to teach the dangers of ice. I want to continue to come back from all my missions and I want my students to return from theirs. So far we're battting 100%.

Posted by: Linda Pendleton | February 3, 2011 1:43 PM    Report this comment

Some people may call it obsession but I call it passion. Some people do it to make money but I believe most of us just want to express ourselves. I also love writing a lot of stuff. If I could write everything I do in my blogposts, I'll do it. Continue your passion because I know that you love what you're doing. curt nike mid-sole

Posted by: mbdmbd twetty820 | February 14, 2011 7:47 AM    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