The problem is that the ice forecast is never precise – and will not be for a long time to come …
That’s a great article.
Yes – 1.5cm per minute is certainly feasible. I have seen that much. But the worst I saw was in smooth stratus, which is contrary to what most people would think.
Peter, the beauty of the article is to show not only that 1.5cm ice/minute is plausible, but also that it is the maximum theoretical limit of ice accretion rate when flying at 130 knots within a supercooled and supermoist cloud, which is sort of reassuring to allow for a quick 180, or to program a climb or descent through a layer.
Only thing that can cause more than 1.5cm ice/min accretion on wings is freezing rain or freezing drizzle, so it’s very important to be able to forecast and avoid them: how would you forecast them? I know freezing rain can only occur when there is a strong temp inversion, i.e. typically when there is a warm front, but what are all sufficient conditions? What about freezing drizzle? Do you know any commercial source of 3D radar images (i.e. also with height)?
Another interesting equation related to this thread is the increase in wing leading edge temperature due to aerodynamic heating:
Square of TAS/87
That is 1 degree increase at 87knots, but 5 degrees increase at 195 knots (very common TAS on a Cirrus Turbo cruise or a Cirrus NA in descent from high altitude).
I like this formula because explains why jets are never much concerned about icing: at 350 TAS the delta-t is 15 degrees, and why turboprops, like the ATR 72, should be concerned by icing at 250 TAS, with just 9 degrees delta-t (supercooled water is much more likely at -9 degrees than at -15 degrees)
That’s very true – however “icing” implies flying in IMC, and do you really want to be flying at 195kt TAS in IMC, without radar?
Very often, in IMC, one can either feel turbulence or one knows from the wx data that a total lack of convective activity is not assured, so one needs to slow down to Va, which removes narly all the advantage of aerodynamic heating.
Big jets have a very high Va and they don’t have a problem with it. Also the passengers cannot harrass the pilots, which is a hugely important factor 
If you fly an SR22 into a proper CB at 195kt TAS, it will probably take the wing(s) off and you will become another pilot forum discussion topic 
At 350kt TAS the aerodynamic heating more or less take scare of the icing issue – excepting locations where there is a pressure drop and the temperature thus falls again.
Assuming supercooled water existing between 0C and about -15C, that translates to a vertical band of about 8000ft which you have to climb through, or descend through. On the way up it will take you about 5 minutes (SR22T) which is plenty enough time to collect a huge amount of ice, in the worst conditions.
The worst icing is at around -5C. I have seen this many times.
Also, for unprotected props, the power is going to be degraded via prop icing. That’s why I suggest anybody wanting to do this gets himself a plane with a TKS prop – it costs much less than the usual avionics eye candy.
Just come across this
-22F is -30C i.e. very cold.
-4F is -20C.
Another interesting equation related to this thread is the increase in wing leading edge temperature due to aerodynamic heating:
Square of TAS/87
That is 1 degree increase at 87knots, but 5 degrees increase at 195 knots (very common TAS on a Cirrus Turbo cruise or a Cirrus NA in descent from high altitude).
Thats an interesting equation. Do you have any reference for that? Does that mean that, if you are flying in IMC in say, -15c at 195 knots, then your leading edges could collect ice (due to the aerodynamic heating) but the remainder of the wing could collect ice?
no, your leading hedges will be at -10 degrees, so if there is supercooled water at -15 in IMC, then you will collect ice on both the leading hedge and the rest of the plane which is at -15. The interesting case would be with an OAT of -5.
I was using the rule of thumb that beyond -15c, it is too cold for ice to form but yes, your example of the OAT being -5c equally poses the same scenario.
