Menu Sign In Contact FAQ
Welcome to our forums

Icing (merged threads)

I'm still somewhat new to IR and I am scared of icing. Heck, I occasionally wake up in the night after dreaming that I'm in an evil icing cloud...

What was your worst icing event? How much ice did you get? How did your airplane react? How long did it take until you reached a serious level of icing?

I've been receiving two types of advice:

  • don't worry, a Cessna can handle a lot of ice and it's no big deal, people are greatly exaggerating
  • stay away from any cloud in freezing conditions unless you're FIKI

Icing is very "random" and this is partly why quite a lot of pilots say they have never seen much of it. There is a lot of bravado around... take it on board at your peril.

However when you get it you get it! And then it is definitely not random

My worst was collecting about 30mm of clear+rime, in about 5 minutes, at about 4000ft, in the south east UK. Bases were 1500ft, tops 4500ft (from memory). SFC OAT +3C. Temp at 4000ft (IMC) was -5C and that is the worst temperature to be in for supercooled water droplets in stratus cloud. It was a very smooth stratus, sunshine and ~100nm vis above, and OK conditions below.

I got about 5mm of ice on the climb up through it earlier, and got the big stuff when descending back into it.

The plane would not climb, and in fact started to stall, at about 100kt. At max power (and I have a TKS deiced prop, without which it would have been much worse) it would only do about 100kt. So that's a drop from 165kt to 100kt, purely due to aerodynamic degradation. No stall warning (switch frozen in the ice). A slow descent fixed it; gave me about 120kt and once I was in +1C the stuff started to melt off, but took about 30 mins to completely clean the elevator (which had the much more dangerous classical "horn" ice formation, about 3" thick). Control of the plane with the ice was "interesting" (very twitchy) but OK.

I was aware of the ice and had originally allowed the buildup to continue because I knew it was +3C on the surface and there were no applicable MSA issues, so a viable escape route existed. Had this been over the Alps I would be dead, but then I would not have been flying in IMC over the Alps in the first place... well not in -5C. At -25C, stratus, one is OK.

I took photos, of course, but won't post them here

The other day I read of a real near-death icing encounter of another pilot who used to write that icing is rarely an issue, etc. One has to take it seriously, and prolonged flight in potentially freezing conditions (IMC between 0C and about -15C, assuming non-convective cloud) needs deice equipment - or an instantly available escape route in the form of a descent.

Here you find an astonishing (especially astonishing that he published it, given the likelihood of an FAA bust on several counts) writeup of an icing encounter over mountains. And this

is a photo of the ice which was still attached after he landed.

I've had loads of ice encounters where I got say 5mm during a climb or a descent through a layer known to be a few thousand feet thick. For the TB20 that is OK; it barely shows up on the IAS. But that is with a TKS deiced prop. On some other types it would kill the wing performance. And yet other types can carry much more ice OK.

One uses strategy to limit the risks; e.g. you don't climb up into a warm frontal system (base say 800ft, tops say 25000ft, 0C level say 5000ft) for a FL100 cruise for 4 hours because there is no way you will come out as anything other than an icicle. If there are no MSA issues then a descent down to 4000ft will limit the damage and eventually remove the ice but you will just cause a bunch of ATCOs, and probably a bunch of airline pilots, to spend their next half a dozen coffee breaks slagging off "GA cowboys" So what's the point in doing that?

Consequently I don't do enroute flight in potential icing conditions e.g. enroute flight through frontal weather at "Eurocontrol IFR" altitudes. But the above encounter was in smooth stratus.

Icing forecasts do exist but are IMHO all but worthless. If you overlay them onto the MSLP ("surface analysis") chart it's obvious where the icing might be (fronts, basically!). But my biggest encounter by far would have never been forecast. There is no reliable way to forecast where supercooled water is going to be and in what concentration. If you stay in IMC below 0C for long enough, you are sure to collect ice eventually.

And yes this does limit hacking around in the winter, in cloud, in non-deiced planes, below controlled airspace as we all do so much of - 2400ft, 3400ft etc It's the price you pay for flying a cheap plane... If you have an IR then in general you can climb to VMC and this is one use for an IR for UK flying.

Shoreham EGKA, United Kingdom

A very relevant thread given the time of year. I have to say even with a deiced plane, I view the equipment as a way to safely pass through an icing layer rather than sit in it.

EGTK Oxford


My worst and most dangerous flying incident so far was icing related. During descent into Stockholm-Bromma in a Pa44 Seminole, we encoutered unforecast icing starting at FL120 and lasting till touchdown. The weather was horrible enough, heavy snow showers, strong wind 45 degrees off the runway, gusting to 50kt. Before takeoff we had obtained a personal weather briefing, at our time of arrival it should have been CAVOK after passage of a cold front. I never found out what went wrong there, he probably gave us the weather from a year ago...

The Seminole kept picking up ice all the way down, there was not a single gap in the clouds or cloudless layer to relax. At around FL70 the autopilot gave up, when I took over manually I knew why: The all-flying tailplane kept stalling all the time. I had to stay above 110-120kt to have elevator control. Eventually we were cleared on the ILS. We flew down that ILS clean (so to say) and I had to set flat-out power in order to keep the speed above 110kt. There was marginal control in all three axes so that our bank angle kept oscillating between 45 degrees both ways and localiser and glideslope were all over the place. Twice we were reminded by the controller how many miles we were off centreline... Communications were hardly readable as the ice on the antennas together with static discharges from the snowflakes and ice crystals made it sound like an HF radio. The propellers were visible against the grey murk as bright white disks as there was more ice rotating around the hubs than aluminium. Shuddering and vibrating heavily of course, but there was now way of reducing power without stalling. At this stage, the aircraft responded only to full deflections of the controls.

When we approached the minimum, nothing was visible, but there was no way we could go around. The aircraft kept descending on it's own on full power and just as the controller called "I understand you are going around?" the guy next to me (who was a self-flying businessman and who had taken me on the trip as safety pilot - but had passed controls to me when we started picking up ice) said: "I see the runway!" Due to the crosswind we were yawed 20 degrees to the right and he actually saw the lights through his left side window. The windscreen was totally frozen up and from my right hand seat nothing could be seen. I pointed the nose down throgh the hole (maybe 100ft above the tarmac) lowered the gear with the left hand, radioded "Negative, we land!" with the right hand, pulled the yoke fully back and in the last moment lowered full flap with the "handbrake lever" at the same time to cushion the fully-stalled and totally controlless pancake-landing. All the time looking through the side window only. I didn't have a spare hand to cut the power, but it made no difference anyway. We touched down in the last 200m of runway with a ground roll of less then 50m. Must have been my shortest landing ever.

During the last 3NM of this approach - which seem endless till today - only one thought kept me/us alive. I kept telling myself: "This is not the way I am going to die. I am not going to be a statistic today. Not today! Not today!"

Till this day, the only thing I am really scared of in flying is icing. As in instrument instructor, I am often faced with the decision wether to go flying in marginal conditions or not. If icing is forecast or anticipated and ground temperatures are such, that the ice will not melt away during descent, I will not fly. No exception. Never again. No problem at all if the freezing level is above 3000ft AGL and the obstacle situation permits descent below that height (even if below radar vectoring minima). Under such circumastances I may even encourage students to enter clouds with possible icing so that they can "experiment" with it. But if it gets too much - down we go.

Happy landings max

EDDS - Stuttgart

Blimey. That was serious.

Darley Moor, Gamston (UK)

Max, it may be impossible to say but what difference do you think a deiced wing/tailplane, prop and windshield would have made? The conditions sound terrible and I don't think being out there at all would have been pleasant.

EGTK Oxford


Max, it may be impossible to say but what difference do you think a deiced wing/tailplane, prop and windshield would have made? The conditions sound terrible and I don't think being out there at all would have been pleasant.

If I could have chosen, I would have taken hot props. As long as you have sufficient power you have options!

EDDS - Stuttgart

'What next' - it seems your bag of experience lasted before your bag of luck ran out. Well done for living to tell the tale, an instructive tale we can certainly all benefit from.

Does anyone have a view of the icing differences between metal and composite-structured aeroplanes?

Specific heat would seem to be a player here - the amount of heat required to change the temperature of one kilogram of a substance by one degree. Specific heat can be measured in kJ/kg Kelvin (in this context, kg Kelvin meaning that for the material the air-frame is made from).

My gut feeling is that metal aircraft may be more resistant to icing, as I suspect they are more absorbtive of heat energy on the ground and therefore may need more 'chilling energy' to drop the flying surface temperatures to 'icing temperature' compared to those of composite aircraft. That said, composite a/c are effectively surfboard technology, i.e. mainly plastic type materials that exhibit a lower thermal mass and conductivity - so surfaces may actually 'chill' faster.

What say you?


If I could have chosen, I would have taken hot props. As long as you have sufficient power you have options!

Only up to a point though

At least with a SEP with a TKS prop the overspray from the prop keeps the front window clear. When I had 30mm on the wings etc the window was fine. Plenty of ice forming on it but it was sliding straight off, in very small pieces.

I suppose every twin with hot props would have a heated window panel though?

My gut feeling is that metal aircraft may be more resistant to icing

I don't think the material makes any difference because what is happening is that the airframe is at the static air temperature (which is say -5C) and as soon the supercooled droplets splatter against it, they go solid.

(Actually the airframe, or each separate part of it, will be at the TAT - Total Air Temperature - applicable to the local airflow. TAT is the SAT corrected for aerodynamic heating, but in the piston GA context the heating is usually only 1-2C. In faster planes, the aerodynamic heating makes a huge difference, and at ~350kt it disappears altogether...usually).

What matters a lot is the aerofoil profile, etc, and allegedly this is a lot less tolerant of ice buildup on certain composite models.

There is some amazing research going on into surface finishes which prevent ice sticking. I reckon some years from now we will have a paint which just does that.

Shoreham EGKA, United Kingdom

The Cirrus is terribly vulnerable to icing. That is because with no ice, it owes its excellent cruise performance largely to its very clean aerodynamics. Take this away (2 millimetes of ice is sufficient) and it gets ugly real quick. This by the way also holds true for a TKS equipped models with the system running. The ice on the unprotected surfaces is sufficient to ruin the aerdynamics. I have no idea about the FIKI models as I haven't flown one. The worst is the somewhat underpowered SR20, which does not have TKS at all. My worst icing experiences have been with the SR20 that I operated previously to the SR22.

Mainz (EDFZ) & Egelsbach (EDFE), Germany
421 Posts
Sign in to add your message

Back to Top