I have some insight on carb icing and the use of carburetor heat, which may be of some use to the RV community. A long time ago, I purchased a basket case Luscombe (still have the airplane, 40 years later!), and assembled it while I was a college student. In the box of all the parts was an old airbox, with no provision for a filter. The combination of the A65 and unfiltered air was very conducive to carburetor icing. I had one engine failure and dead stick landing on an airport due to carb ice, noted by water dripping out of the carb 15 minutes after changing my shorts!
I replaced the carb air box with one requiring a filter and the situation changed significantly. There were fewer times when I picked up carb ice, although it would still occur and I had to be ready for it. The circumstantial evidence is that the turbulence created aft of the filter, or reduction in speed of the air flow reduced the formation of carb ice.
On that airplane and others with the small Continental engines, I would routinely do a normal run-up, check for rpm drop with carb heat, then check the engine at idle with full carb heat applied. If there was any roughness, or if the rpm was lower than it should be, carb ice during taxi out was suspected, and I would run with carb heat applied at a higher power setting until the check was satisfactory at idle. It is very possible to get carb ice when taxiing on a moist morning, or on grass wet with dew.
Moist conditions can exist in layers, for example if you are flying at the level where clouds will begin to form, but before they actually do. I was flying one day along the eastern shore of Lake Erie in the Luscombe and encountered icing conditions where the wind was blowing onshore in a fairly narrow band.
Now about the Lycomings and ice. Some of my early test flying experience was on an FAA contract to study carb icing. We used a Piper Apache and flew IFR in a published holding pattern near the airport at Springfield, OH. While the hold was within gliding distance, we never had to use that feature. We ran one engine with full carb heat and let the instrumented Lycoming O-320 ice up. We found that if we allowed the unprotected engine to ice below 15 inches MP, no amount of heat would recover it, and we would land single engine. A good takeaway here is that in worst case conditions, a Lycoming can lose power and stop due to carb ice.
Many of the earlier European designs (Jode Musketeer, Fournier RF-5, as examples), did not have carb heat. Instead, their filtered air came from inside the cowl space, after being heated by the cylinders. In spite of conditions strongly conducive to carb icing, it was a successful approach to the problem, with few events. Of course, there was a performance penalty to pay for the loss of rpm.
What I think makes sense for us RVers is to have carb heat available, be continuosly aware of outside conditions, and have a monitoring system available. During construction of my RV-8, I did some reading about carb ice detectors. I found that while they work, the literature says they are less reliable than a temp sensor mounted in the carburetor itself. Temperature can be both monitored, and set to alarm if conditions are conducive to carb ice. So I monitor carb temp on my GRT EIS, and use that as both an alarm, and as a check that my carb heat system is working (weakly, as most of us admit!).
I also think, going back to my experience in the Luscombe and the European designs, that the carb air box design Van recommends creates significant turbulence in the airflow. That, and the warmth from the engine and exhaust in a fairly tight cowl keep the carburetor fairly warm. These may be the reasons we see few cases of carb ice in our RVs.
