So kinetic heating + centrifugal effect seems to give some icing protection to the prop when only a few degrees below freezing.
I think you are right and prop icing is a smaller problem than airframe icing, but the extent to which this is true is probably dependent on various things.
A slow prop (geared engine) is going to ice up much more easily because the “TAS” is lower and the aerodynamic heating is nonlinear with the TAS.
Also I think if your prop ices up then it is game over a lot sooner!
Accordingly, I am always careful to mention, when I talk about icing on a TB20, that I have a TKS prop. That stuff is so effective that no matter how much ice I have had (and I have had only the one “experimental” ice encounter that collected a lot of ice) the front window was completely free of ice, so I think it is working pretty well on the prop.
So kinetic heating + centrifugal effect seems to give some icing protection to the prop when only a few degrees below freezing
While kinetic heating is probably worth 2o C at typical GA retractable piston speeds, also recall there are plenty of low pressure spots where venturi effect lowers the temperature, in particular the prop. IIRC ice tends to form earlier on the prop than on the airframe, there may be an old NASA film to that effect? Certainly it was not unusual for an MEP to sling ice at the nose, before there was much of an accumulation on the airframe.
I tested my TKS today, on a VFR flight … After one minute the fluid was all over the windshield
Remember, the manufacturer recommends to run it every 30 days and for several minutes, until all panels are clearly covered with fluid.
It’s not only the horrible TBM crash, read the files of the COMAIR crash in Buffalo and of AmericanEagle, Roselawn near Chicago. And those were FIKI AIRLINERS. No chance for recovery.
Presuming that you mean the Colgan crash at Buffalo (there was a Comair icing related accident near Detroit, 15 years ago), I find it ironic that you would cite it. A crew who didn’t understand icing were somewhat scared of it, which may well have contributed to the stall of an aircraft that was perfectly capable of dealing with the icing conditions it found itself in. It seems like a great example of a why “overdoing ice” by treating it as instant death is not, on balance, in the interests of safety.
After one minute the fluid was all over the windshield
Even the prop-only TKS system, without the spray bars which one gets with the full €40k system, covers the front window with a fine mist which takes a long time to come off.
One can land with it but normally I turn the TKS pump off the moment the OAT does positive, to give the stuff time to disperse before I land.
The quickest way to disperse it is in cloud (the droplets wash it off faster) but obviously the cloud needs to be above 0C 
Also if flying in VMC at say +1C, if you enter cloud the OAT is sure to drop to -1C or -2C and you start collecting ice again…
the manufacturer recommends to run it every 30 days and for several minutes
A lot of Socata TB owners, with the full TKS system, or even with the prop-only TKS system, never exercise it (many fly with no fluid in the tank) and then they post on the TB owners’ group saying their TKS pump has seized up and why does it cost $3000 (or whatever) 
In fact there is a real issue with the prop-only system, which has a mysterious little pump in the side of the 2 litre fluid container, whose manufacturer is not evident, and a lot of people would like to know who makes it!
Peter, FYI, Alexis’ TKS system is the non-FIKI system which neither has a spray bar for the windshield. So likewise, all the fluid on the windshield came from the prop slingers.
For the fluid to show up on the winscreen it usually takes less than half a minute after switching the system on. However, it takes something like 3-5 minutes for all the wing panels to be completely soaked.
Right ;-) I think it’s almost as good as the FIKI spray bars!
@bookworm
Yes, of course I mean Colgan Air, and I also mean American Eagle. You may find it ironic that I cite those, but my intention was to show that icing can, in no time, make a professional airplane unflyable. I had no intention talking about the qualification of the crew, but my intuitio tells mej that they were probaly at least as qualified as most of us here (okay ME if that sounds better to you). I could cite other crashes too: Read the one of a Cessna 310, some 20 years ago in Germany. And there’s many more.
But I think this leads to nowhere. I am convinced that beginning with moderate icing ice is a deadly threat to small airplanes, and since I’m not the type to take chances I avoid it.
BTW the “full” TKS system (like on the Cirrus) makes a real mess on the entire aircraft. Therefore, I usually do the 30-day run at the beginning of a cross country flight, right after reaching the TOC. This leaves enough time for at least most of stuff to fly off. However, it is so sticky that some of it will stick to the airframe until after landing. The best remedy is really to find a nice rainshower before landing ;-) …useful not only for the insects etc. but also for the TKS stuff.
Why is this important? Well, the people on the COPA forum never really agreed on whether the TKS fluid is (or can potentially be) corrosive. Nor do I know for sure. Fact is that if you didn’t find that rainshower at the end of the flight, you will have some of the stuff on the ailerons and the elevator, which are made of metal even on the Cirrus. So, when that happens, I usually take a rag and clean it off before closing the hangar doors.
