Why beware of icing at anything colder than +2 C?

[RobertE]

I’m sure we’ve all heard what is usually described as the temp range at which airframe icing becomes possible but what I don’t understand is why that range includes a temperature that is warmer than freezing.  Is that to allow for possible measurement error?  Or, maybe, the lower pressure (and, thus, cooling effect on that air) of the air above the top surface of the wing? 

If the outside air (well, to be precise, my airframe) is +.5 C I’m not really at risk of icing am I?  

 

Thanks. 

[N201MKTurbo]

My experience is that the ice starts to melt when the OAT needle gets one hairs width above zero. 

Maybe they are accounting for the accuracy of the OAT gauge? 

[Danb]

One item to consider is the temp of the airframe after leaving a colder temperature to a warmer one. How long will it take our plane to be warm enough if we leave -5 to +2 our frame will still be below 0 for how long? Consider no sun above us we’re under a cloud deck. 

[Jerry 5TJ]

15 minutes ago, RobertE said:

 

If the outside air (well, to be precise, my airframe) is +.5 C I’m not really at risk of icing am I? 

Ignoring freezing rain?  Then at a static air temperature of +5C it is unlikely 

As DanB points out, if you cold soak  the wing and fuel to well below freezing point of water then descend into warm rain your wings can collect a fair amount of run-back ice.  

[jaylw314]

You can also run into supercooled liquid coming out of colder air from above or below.  It will freeze, and then melt eventually, but if it is heavy enough, it will accumulate faster than it freezes.

[RobertE]

Thanks for all the answers.  It sort of confirms the problem with suggesting icing is possible at +2C.  If I’ve been up at 20k feet and the fuel and airframe are, say, -15 C and I enter a cloud I can get iced up at +5C or pretty much any temp, I think.  Still makes me wonder the magic of all that advice “beware of icing from +2 to - 20 C”.

 

 

[GLJA]

Where did you get the +2 to -20C? 

Ground school taught me that icing is most prevalent between 5 and -15C

[RobertE]

Well, that’s a good question.  I’ve seen those numbers a couple of places but +5 C as the beginning of the danger area just magnifies what I see as the problem.  Is it genuinely reasonable to be unwilling to enter a cloud at 5 C, which is about 41 F?  Assuming, of course, no special circumstances like a cold soaked aircraft that is below freezing.

i guess I’m looking for honest rules to live by, stripped of any cushions for measurement error or other circumstances.  

[Jim Peace]

9 hours ago, RobertE said:

Well, that’s a good question.  I’ve seen those numbers a couple of places but +5 C as the beginning of the danger area just magnifies what I see as the problem.  Is it genuinely reasonable to be unwilling to enter a cloud at 5 C, which is about 41 F?  Assuming, of course, no special circumstances like a cold soaked aircraft that is below freezing.

i guess I’m looking for honest rules to live by, stripped of any cushions for measurement error or other circumstances.  

great post.......

[Bob - S50]

The Air Force used to teach us that icing was possible from 2000' below to 10,000' above the freezing level.  Assuming a standard lapse rate that would be from +4 to -20C.

I suspect the reason for temps >0C are due to reduced temps because of decreased local pressure.  My Aviation Weather book calls it aerodynamic cooling.

If icing was due to airfoil cold soaking then they would not have a temperature range because you can get thick frost on a wing while sitting on the ground in Florida.  That's because of the cold soaked fuel in the wings.  We were allowed up to 1/8" of frost on the bottom of the wing as long as the tops were clean.

[mooniac15u]

Wet highway bridges can freeze at about 38F (3C) due to evaporative cooling from wind.  Presumably the same evaporative cooling effect would be possible on airplane wings although the exact temperature might be slightly different.

[squeaky.stow]

One of the reasons for this guidance is the venturi effect. As air accelerates over a surface like an airfoil, the pressure and the temperature both drop. That is how it is possible to get carb ice in temperatures above freezing. For the same reason, the engines on the big Boeings require engine anti-ice on any time you have visible moisture and temperatures below 10C. Below 3C you have to do a fan blade ice shedding runup before takeoff. I suspect it would be pretty rare to pick up ice on the leading edge of a Mooney wing at 2C but it all depends on the amount of mosture and the pressure/ temperature drop. The propellor might be a different story.  The attached photo is a 787 fan blade after about a 10 minute taxi in on a foggy day at +2C. You can see the ice on the leading edge of the blade. What you can’t see is the ice on the back side of the blade where the pressure drop is the highest. It was a pretty impressive accumulation for 10 minutes at idle thrust. 

[Scott Dennstaedt, PhD]

On 3/7/2019 at 4:51 PM, RobertE said:

I’m sure we’ve all heard what is usually described as the temp range at which airframe icing becomes possible but what I don’t understand is why that range includes a temperature that is warmer than freezing.  Is that to allow for possible measurement error?  Or, maybe, the lower pressure (and, thus, cooling effect on that air) of the air above the top surface of the wing? 

If the outside air (well, to be precise, my airframe) is +.5 C I’m not really at risk of icing am I?  

This is a question that I get asked a lot and I'll be discussing this a bit more in my webinar next week.  For the sake of argument, what is to follow assumes the aircraft is in visible liquid moisture.  First, when the static air temperature (SAT) is warmer than 0°C, you shouldn't expect to accrete ice.  Once the SAT reaches about -7°C, almost all aircraft will begin to accrete some ice.  The important temperature for determining ice accretion, however, isn't the SAT, but the total air temperature (TAT) or what some refer to as the ram air temperature (RAT).  The TAT is the temperature just above the skin of the aircraft.  By the way, some aircraft show the TAT and then infer the SAT from that (honestly that's a more sensible approach).  But just be careful...when you offer a PIREP, please use the SAT, not the TAT.  

The TAT is different on different parts of the aircraft.  Excluding the prop for now, the immediate leading edge of the wing or horizontal/vertical stabilizer is typically the warmest.  That's due to an effect called kinetic heating.  Kinetic heating is primarily a result of both friction and adiabatic compression.  As the wing moves through the air, the air just in front of the leading edge will "pile up" and compress causing a rise in temperature due to the laws of thermodynamics. 

While not exactly linear with airspeed, you can use 1°C rise for every 50 knots of airspeed as a quick estimate.  So an aircraft traveling at 150 knots will see a TAT of 3°C warmer than the SAT.  However, as you move away from the immediate leading edge, the kinetic heating drops off quite rapidly.  If you've ever flown in visible moisture with an SAT of about -2°C or -3°C, you may notice horns building on the top and bottom of the leading edges with liquid dancing around in between.  That's because the temperature at the immediate leading edges are too warm (above 0°C) for freezing to occur.  But just above and below the immediate leading edge, it's cold enough to accrete ice.   

The blades of the prop are moving at a much higher speed and typically don't accrete ice out at the ends.  Again, this is due to kinetic heating. Prop ice tends to collect at the hub and then progress outward...in colder temps, it can move out toward the ends which will decrease propeller efficiency.   Losing thrust is very bad...it really limits your options very quickly.  If there's anything you can add to your aircraft for ice protection, it's prop de-ice.   

Not to add complexity, but those surfaces with a high radii of curvature will collect ice more efficiently than a low radii of curvature.  So, thin wings will collect ice more efficiently than fat wings.   It also depends on the size of the drops.  Except for the immediate leading edge, small drops tend to just flow over the wing and don't penetrate the boundary layer above the wing.  Larger drops have more momentum and will typically penetrate the boundary layer further back behind the leading edge.  In other words, there are other factors besides air temperature that come into play.  

One last point.  It's really hard to define temperature.  SAT is the undisturbed air around the airplane (when it's measured by an aircraft, it's usually referred to as the OAT).  The SAT is typically measured by an immersion thermometer on many GA aircraft.  On the ground in the shade, the thermometer might be very accurate, but when in the air, it may also suffer the same kinetic heating as the airplane's leading edges.  When that immersion thermometer gets wet or accretes ice, evaporative cooling can quickly drop the temperature of the probe by several degrees almost instantly.  Have you ever flown into a cumulus cloud (at temps above 0°C, of course) and saw the OAT drop by several degrees?  That's a wet immersion thermometer.  This is a long way of saying that if you are measuring a temperature of +2°C with your immersion thermometer and then fly into visible moisture, you should expect the OAT to drop a few degrees creating a risk of icing.  

BTW, airplanes don't feel wind chill.  Evaporative cooling of moisture (or sublimation), yes, wind chill no.   And please, never use the standard lapse rate to estimate the freezing level.  When making weather decisions, if you catch yourself using the standard lapse rate, slap yourself in the face!  Now, for the tables in your POH, sure the "departure from standard" applies to understand performance.  But don't use the standard lapse rate for anything else.  

Hopefully that helps clear things up.

[David Lloyd]

I need to read that 2 or 3 more times.  Great answer!

[steingar]

You know nothing Robert E. I I live in the icing capital of North America. If it’s anywhere between 20 below to 20 above you can just about bet there be ice in those clouds.  Were I not put off the IR by my dire financial straights I’d probably be put off by the ice. Can’t think of a local pilot friend who hasn’t gotten into dutch because of it.

[RobertE]

Your are The Man, Scott.  Thanks for that comprehensive answer.

Are you the guy teaching the Pilot Workshop Skew T diagram reading class I began yesterday?

[Scott Dennstaedt, PhD]

1 minute ago, RobertE said:

Your are The Man, Scott.  Thanks for that comprehensive answer.

Are you the guy teaching the Pilot Workshop Skew T diagram reading class I began yesterday?

You are welcome.  Yes, that's me. I license Pilotworkshops to sell my Skew-T program.  

[David_H]

1 hour ago, scottd said:

Excluding the prop for now, the immediate leading edge of the wing or horizontal/vertical stabilizer is typically the warmest.  That's due to an effect called kinetic heating.  Kinetic heating is primarily a result of adiabatic compression (not friction).  As the wing moves through the air, the air just in front of the leading edge will "pile up" and compress causing a rise in temperature due to the laws of thermodynamics. 

I don't claim to know much about Weather other than knowing when to stay on the ground and being thankful that those such as yourself can give good weather predictions. However, Heat and Mass Transfer, Aerodynamics, and Thermodynamics...

The leading edge of an airfoil (depending on the angle of attack) is where the stagnation point is located. This is a point in which the airflow velocity is effectively zero and the pressure is at it's highest. Hence, no (or very little) convection cooling is present in this area. Hence, Conduction heat transfer (much less effective than convection) is dominant. Compression will not be present at the stagnation point as long as the airfoil is subsonic.

How did you get to the hypothesis of Kinetic Heating being the reason for the skin temperature on the leading edge of an Airfoil? Kinetic heating is a function of friction alone which has very little to do with the leading edge of an airfoil that's not in a stall condition.

 

[Danb]

9 hours ago, RobertE said:

Your are The Man, Scott.  Thanks for that comprehensive answer.

Are you the guy teaching the Pilot Workshop Skew T diagram reading class I began yesterday?

I took Scott’s skew t course and it’s worth every cent , I need to review it often do to its complexity for some of us more challenged folks

[carusoam]

Nothing like being fully trained in the science of thermodynamics... and finally finding a real world place to apply it...

It is pretty generous of ScottD to supply such great insight on MS!

There are so many things effecting the actual temperature of the wing’s surface while we are flying...

Temperature we sense...

• Some are sensor shape and size and location... and that is what we are measuring with...
• some include air within a boundary layer...
• some include sources of warm air from engines...

Temperature that physically gets adjusted...

• Any time air is in a low pressure area, it is expanding and cooling...
• Any time air is in a high pressure area, it is compressing and heating...
• So the top of the wing, low pressure, it is going to be cooler...
• Leading edge, under compression, it is going to experience some heating
• There is going to be some friction as well, but compared to these other sources of thermodynamic heating and cooling... the friction is less strong... 

Then there is the laws of physics related to mass and momentum of water droplets being approached by a wing in flight...

• Small enough they move with the air and go around the wing...
• too large they crash into the wing and spread back while freezing....

Then there is the oddities of pure water...

• Freezing requires a source of nucleation to start the crystallization...
• nucleators can be dust, dirt, ions, or airplanes...
• it is very possible for water in the atmosphere to be nucleation free... thus allowing it to be in the phase of water, but be super cooled waiting for the nucleation to arrive...

Then there is the dynamics of vertical air currents.... where nothing is static... it is always changing...

 

So what we learned...

• 32°F is where things freeze in a static dirty condition...in the shade...
• but, not too dirty, because that would cause something called freezing point depression... lowering the freezing point of the water solution...
• If we flew in those conditions icing would be much easier to predict...

Thanks, Scott!  For making Thermo 101 applicable to real life, while improving the safety of flying a GA plane... 

I am sure I missed a few details... my thermo classes were taught by a prof in his 40th year of teaching, and that was 30+years ago...

Thanks to Dr. Teddy Gela... Prof of thermodynamics in the 80s...

There is so much to know...

Best regards,

-a-

[N231BN]

I don't claim to know much about Weather other than knowing when to stay on the ground and being thankful that those such as yourself can give good weather predictions. However, Heat and Mass Transfer, Aerodynamics, and Thermodynamics...
The leading edge of an airfoil (depending on the angle of attack) is where the stagnation point is located. This is a point in which the airflow velocity is effectively zero and the pressure is at it's highest. Hence, no (or very little) convection cooling is present in this area. Hence, Conduction heat transfer (much less effective than convection) is dominant. Compression will not be present at the stagnation point as long as the airfoil is subsonic.
How did you get to the hypothesis of Kinetic Heating being the reason for the skin temperature on the leading edge of an Airfoil? Kinetic heating is a function of friction alone which has very little to do with the leading edge of an airfoil that's not in a stall condition.
 

I can't tell you the proper term for the heating that takes place, I'm just a mechanic. What I can tell you is there is significant heating of the leading edge of a fast(subsonic) jet. You can easily get a 10 degree rise which basically puts the aircraft out of the icing range when the atmosphere is right for it. This is one of the reasons the old Lear jets where known for their icing capability, high airspeed. The other being a lot of thrust as well as tons of hot bleed air for anti-icing.

[Bob - S50]

1 hour ago, N231BN said:

On 3/8/2019 at 7:46 PM, David_H said:

I don't claim to know much about Weather other than knowing when to stay on the ground and being thankful that those such as yourself can give good weather predictions. However, Heat and Mass Transfer, Aerodynamics, and Thermodynamics...
The leading edge of an airfoil (depending on the angle of attack) is where the stagnation point is located. This is a point in which the airflow velocity is effectively zero and the pressure is at it's highest. Hence, no (or very little) convection cooling is present in this area. Hence, Conduction heat transfer (much less effective than convection) is dominant. Compression will not be present at the stagnation point as long as the airfoil is subsonic.
How did you get to the hypothesis of Kinetic Heating being the reason for the skin temperature on the leading edge of an Airfoil? Kinetic heating is a function of friction alone which has very little to do with the leading edge of an airfoil that's not in a stall condition.
 

I can't tell you the proper term for the heating that takes place, I'm just a mechanic. What I can tell you is there is significant heating of the leading edge of a fast(subsonic) jet. You can easily get a 10 degree rise which basically puts the aircraft out of the icing range when the atmosphere is right for it. This is one of the reasons the old Lear jets where known for their icing capability, high airspeed. The other being a lot of thrust as well as tons of hot bleed air for anti-icing.

In the Air Force we called it the speed of heat.  I believe you are referring to compression heating.  If I remember right, the speed of heat was about 450 KIAS.  At that speed one of two things happened.  If the air was cold enough that compression heating didn't bring the temperature above freezing, then it didn't have enough moisture in it to make icing a concern.  Or, if it had enough moisture in it to make icing a concern, compression heating would bring the temperature above freezing.