What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

To carb heat or to not carb heat.....

Latech15

Well Known Member
So, I was taught in an old Cessna, like most of us, and the only time I ever put on carb heat was when entering the pattern and pulling the throttle back. Now I have a very nice RV6A with a Dynon panel which is equipped with a carb temp gauge. The carb temp turns yellow at 40 degrees, and red at 35 degrees. I have been flying this plane for a couple of years and I usually put the carb heat on when it gets below the 35 degrees, regardless of the stage of flight. FWIW - I didn't not build the plane. The original builder said that he never used the carb heat, even in the pattern. As long as the carb temp gauge showed that is wasn't near 32/0 that he flew the pattern and landed with no carb heat. I have been following those instructions for 2.5 years and almost 300 hours.

On a flight yesterday, it was cold the entire trip. I left the carb heat on the whole time. Clear air, no moisture. I found that when I turned it off, adjustments had to be made to the throttle and the mixture to get it back to a EGT, CHT, Fuel Flow setting that I am used to. In speaking to a flying buddy, he recommended that I ask on here as to if I even needed carb heat at all in that situation. I am positive that if I didn't have the carb temp gauge, that I wouldn't have turned it on unless I was experiencing the signs of carb icing, low RPM and engine roughness. So What say you VAF, when do you pull the carb heat knob?
 
I oftentimes use constant carb heat in my 172 with an O-360 when it’s cold out. I find that it makes my EGT’s much more uniform. Cold 100LL does not like to vaporize very well and the warm air helps. If I were in a dusty environment would certainly exercise caution...

BAR
 
So, I was taught in an old Cessna, like most of us, and the only time I ever put on carb heat was when entering the pattern and pulling the throttle back. Now I have a very nice RV6A with a Dynon panel which is equipped with a carb temp gauge. The carb temp turns yellow at 40 degrees, and red at 35 degrees. I have been flying this plane for a couple of years and I usually put the carb heat on when it gets below the 35 degrees, regardless of the stage of flight. FWIW - I didn't not build the plane. The original builder said that he never used the carb heat, even in the pattern. As long as the carb temp gauge showed that is wasn't near 32/0 that he flew the pattern and landed with no carb heat. I have been following those instructions for 2.5 years and almost 300 hours.

On a flight yesterday, it was cold the entire trip. I left the carb heat on the whole time. Clear air, no moisture. I found that when I turned it off, adjustments had to be made to the throttle and the mixture to get it back to a EGT, CHT, Fuel Flow setting that I am used to. In speaking to a flying buddy, he recommended that I ask on here as to if I even needed carb heat at all in that situation. I am positive that if I didn't have the carb temp gauge, that I wouldn't have turned it on unless I was experiencing the signs of carb icing, low RPM and engine roughness. So What say you VAF, when do you pull the carb heat knob?

I think each carb/engine set up is slightly different and carb ice can form at different conditions for each. for me I notice carb ice when the OAT is between 15-25C and relative humidity is over 50%. I can feel the engine miss very slightly as the ice sheds. I normally do not use carb heat even in these conditions and just let it shed. at higher altitudes I never have carb ice.

one thing I did that may have helped is that I polished the carb venturi. I notice some flash on the carb venturi casting and I spent some time to polish it. if you like to polish metal, that is probably the best location to spend your time.
 
Last edited:
I operate mine the way my Lycoming operators manual says to. Check it on the ground before takeoff to make sure it is working. Mine shows a very slight RPM drop. Then in flight, only used if encountering icing conditions. I don’t use mine in flight very often, basically never because I don’t fly in icing conditions. I have pulled it on in flight to check it, and saw the same indications I see on the ground. I have a carbureted O-360-A1A with a fixed pitch prop.

On my first RV (RV6), I had a carb temp gauge, and I don’t think it was accurate. It indicated 35-40* all the time, even in the hot summer. Carb heat on only increased the indication 1-2*.
 
According to the book, carb heat can form at up to 70 degrees, when the relative humidity is at 50%. That encompasses a lot of the year in the south.
 
Last edited:
It's local pressure dependent

The throttle position has a great deal to do with it. The carb venturi lowers the static pressure at the throat by design so the fuel will vaporize easier. The molecules only feel the static conditions. Everything else is "relative". Water is special. The lower pressure at the venturi throat raises its freezing point. A closed (even partially) throttle has the same effect compounded with that of the carb venturi so moisture can freeze at a very high temperature. Never use partial carb heat. It can make conditions worse based on the aforementioned principles.

BTW, if you're kicking it old school with a carb and manual mixture control, always cruise WOT and adjust speed with Mixture and RPM. This minimizes your throttling losses. This was religion in Mooney world.

SB attached for reference. It may have been updated but the basis would be the same.

Fly safe.
 

Attachments

  • Use of Carburetor Heat Control.pdf
    31.7 KB · Views: 211
Last edited:
Just because its cold out doesn't mean carb ice is going to form. Google the chart that shows temperature and the probably of carb ice. Flying around with the knob pulled just because the guage shows under 35 degrees is most likely not only a waste of power, but also sucking in unfiltered air that bypasses the air filter. I only use carb heat upon symptoms of carb ice.
 
old wives tales die hard... ;)

My experience: Carbed O-360 with standard Vans FAB. Several hundred hours of flying in sometimes pretty hard weather, hot & cold, rain, mist, clouds, icing, etc.
Never had to use carb heat... which sure don't mean I will never have to.

Usual practice is to test it's function during the run-up, then leave it alone for the rest of the flight. It will be used if I ever see an unusual drop of the MP.

As for the dedicated carb T gage, it is an analog relic that will disappear during my next panel update... absolute useless indication.
 
What engine do you have? 0-320's and 0-360's with the up-draft carbuarter are not prone to carb ice. I fly a RV6-A with a 0-360-A1A and a Sam James cowl. I have a carb temp probe. I never used carb heat even in carb ice conditions. Colder air is better for your effincency unless you need the carb heat to get a good lean.

Many RV's have very poor carb heat set-ups. The fact that you had to re-adjust your lean when you turned yours off would suggest that yours works well, but have you measures the inlet temp with and without it? I had a 5 degree F tempature change until I re-worked everything and go it up to 30 degrees.
 
Point of clarity

Cold air being better for engine efficiency is misleading. There's a lot of variables at play.

It sounds contrary but warmer air and warmer fuel make for more efficient subsonic combustion, which our IC engines are designed for (talking deflagration. Detonation and diesels are a different topic). Colder air allows the engine to make more power as it can burn more fuel. Becoming more or less efficient depends on the specific design point of the subject powerplant. There's real engine guys on the forum who I'm sure will chime in.
 
Better put it on the plane

Someday somewhere that carb heat will be needed. If you dont have it available it will become a very high pucker factor experience.

Ive had my 0320 ice up between the hangar and the runway and Ive had airborne battles with carb ice. I know from the previous posts there is a lack of understanding of carb ice and how you deal with it. Flying around with the carb heat on is a bad idea.

Cm
 
Someday somewhere that carb heat will be needed. If you dont have it available it will become a very high pucker factor experience.

Ive had my 0320 ice up between the hangar and the runway and Ive had airborne battles with carb ice. I know from the previous posts there is a lack of understanding of carb ice and how you deal with it. Flying around with the carb heat on is a bad idea.

Cm

Yep, the two instances of carb ice I recall in the 21 years of flying my O-320 carbed RV-6 occurred during long taxi to the runway. The runup was a little rough and a few moments of heat cleared it up. Otherwise, carb heat has never been used inflight, but I only fly VFR.

The O-200 in the biplane is a first-class ice maker, however. :)
 
Last edited:
Get rid of the carb temperature gauge. Install a carburetor ice detector instead.

I have one in my non-RV, and it's worked great for more than twenty years. It has a probe that senses ice and a panel light that comes on when there is ice. No light, no ice. Sometimes partial heat will cure it but occasionally it needs full carb heat.

The light will come on at low voltage, too, FYI. It's a purely analog system.

The only source I know of is Aircraft Spruce.

Since my RV-3B engine has a carburetor, I bought one of these for that plane.

Dave
 
My Experience

Got carb ice at 500' AGL while flying around Catalina Island California at relatively low power. Sunny 60 degree day. If you aren't familiar with the location, pull it up and estimate your pucker factor. Engine - o-320 Lycoming, updraft carb. Clearly near the water the humidity was up there.... [Wasn't in an RV]

Also one other time (same plane), again at a lower power setting.

For the advice given on not bothering with it until you've determined you have ice... keep in mind the amount of carb heat is proportional to the engine power. If you don't have much power (due to carb ice), you won't have much heat in the muff, and the carb heat is going to be pretty slow to work. Not a good thing if your carb heat is marginal as several have noted above.
 
Latent heat of vaporization of the fuel literally sucks heat out of the air and lowers the temperature in the venturi a bunch. This is one of the reasons you can get carb ice at relatively high temps.

The carb on a Lycoming mounts right to the hot oil sump and on a Continental, it mounts to a cold air plenum. That's why the Conties will ice up so easily. I've never seen carb ice with a Lyc at high throttle. Doesn't mean it can't happen if the air is moist enough though. As others have mentioned, it's a lot more likely at low power settings - eg. taxi or descent.

A poster said to never use partial carb heat. I agree, if we are using carb heat for ice removal. We can, however, use partial carb heat to aid fuel vaporization while trying to get lean of peak in non-icing conditions. That and just a little tiny bit of throttle reduction to add some turbulence to the fuel air mixture can make the difference. With a touch of carb heat, you can often lean much further without running rough.


Ed Holyoke
 
Carb heat: not for cruise......

Yep, the two instances of carb ice I recall in the 21 years of flying my O-320 carbed RV-6 occurred during long taxi to the runway. The runup was a little rough and a few moments of heat cleared it up. Otherwise, carb heat has never been used inflight, but I only fly VFR.

The O-200 in the biplane is a first-class ice maker, however. :)

Someday somewhere that carb heat will be needed. If you dont have it available it will become a very high pucker factor experience.

Flying around with the carb heat on is a bad idea.

Cm

For decades, my preflight check includes carb heat, and left on for a count of at least 5 (and my hand left on it to make sure I put it back in) and see if there is a rise in RPM. You leave it on as it may take a few seconds or longer for the ice that is there to be heated away. Pulling it on and shoving it right back in is not going to tell you anything except, perhaps, that it is working.

Ice may form at any time. As fuel goes from a liquid to an "aerosolized gas", that takes energy, which is heat from the environment and that evaporative loss of heat is the reason ice can form: decrease in temperature as the air is mixed with the fuel. As long as there is moisture in the air, ice can form. I've seen studies that show ambient temperature sometimes is not a factor. Don't ask me where; that was years ago. Evaporative cooling: works in my house coolers as well.

That said, I do not use it in flight unless I suspect ice: usually from a perceived drop in RMP. In all the hours I have in both O-320 (carburetor) SuzieQ and the C-90 Cub, I have had maybe one or two episodes of carburetor ice in the Cub and cannot remember what the circumstances were. Most of my perceived RPM drops seem to be attributed to "auto-rough".:p

As it is a habit born of decades of flying, I will pull it before my turn from downwind to base on both airplanes. Consistency. I would rather pull carb heat for no reason than to have the engine go quiet on base or short final :eek::eek:

Warm air is less dense :rolleyes: and cruise with heat on makes your engine less efficient: there is less air going into the cylinders. What in the world did we do before carb temperature probes? The Cub has......well, a carburetor. Throttle, mixture, carb heat. And has been flying since 1946.........

IMHO; your mileage may vary......... What is your usual routine? Stick with that..........
 
I use about 1/3 carb heat at altitude during cruise. I find it evens out my temps and allows me to lean past peak with my O-360 RV6.
 
For many years I flew a 182 with an O-470 which, as others have noted, seems more carb ice prone than many other aircraft. I got carb ice a few ‘typical’ times (taxiing in 50 F pouring rain, flying low over the bay at 59F, high humidity) and at least once atypical (IMC, 100% humidity but 40F, full power climb) time. A few thoughts:
1. The recommendation to pull carb heat for landing is for landing approaches made at idle power, because under such conditions carb ice can actually shutdown the engine without the pilot knowing.
2. As others noted, most of the cooling is due to the evaporation of fuel.
3. In some engines (like the O470) the mixture distribution is awful. Partial carb heat can increase fuel vaporization, which may help improve the distribution. In the 470 this effect resulted in more power from the engine, even though the most optimum cylinder produced less, most notable at altitude/full throttle.
 
I’d need more data on that

Regarding;
“2. As others noted, most of the cooling is due to the evaporation of fuel.”

I’d need more data but with a lack of such, this seems doubtful. Low local static pressure levels seem to be the prime driver. Example: Axial Compressors ice at very high ambient temps when inlet guide vanes are less than full open (no fuel evaporation effects). Likewise, carb heat is rarely needed at power levels greater than 80% i.e. throttle valves open/low static pressure loss. Similarly, closed throttles (resulting low local static P added to that of Venturi) where very little fuel flow (very low latent heat losses) are most prone to icing for given conditions.

If someone has related data, it would be appreciated. As of now, I would need some convincing.

Thanks.
 
The FAA says:

Regarding;
“2. As others noted, most of the cooling is due to the evaporation of fuel.”

I’d need more data but with a lack of such, this seems doubtful. Low local static pressure levels seem to be the prime driver. Example: Axial Compressors ice at very high ambient temps when inlet guide vanes are less than full open (no fuel evaporation effects). Likewise, carb heat is rarely needed at power levels greater than 80% i.e. throttle valves open/low static pressure loss. Similarly, closed throttles (resulting low local static P added to that of Venturi) where very little fuel flow (very low latent heat losses) are most prone to icing for given conditions.

If someone has related data, it would be appreciated. As of now, I would need some convincing.

Thanks.

From AC 8083-32:
Ice can form in the induction system while an aircraft
is flying in clouds, fog, rain, sleet, snow, or even clear air that
has high moisture content (high humidity). Induction system
icing is generally classified in three types:
• Impact ice
• Fuel evaporation ice
• Throttle ice

and:
Fuel evaporation ice or refrigeration ice is formed because of the decrease in air temperature resulting from the evaporation of fuel after it is introduced into the airstream. As the fuel evaporates, the temperature is lowered in the area where the evaporation takes place. Any moisture in the incoming air can form ice in this area. It frequently occurs in those systems in which fuel is injected into the air upstream from the carburetor throttle, as in the case of float-type carburetors. It occurs less frequently in systems in which the fuel is injected into the air downstream from the carburetor. Refrigeration ice can be formed at carburetor air temperatures as high as 100 °F over a wide range of atmospheric humidity conditions, even at relative humidity well below 100 percent. Generally, fuel evaporation ice tends to accumulate on the fuel distribution nozzle in the carburetor. This type of ice can lower manifold pressure, interfere with fuel flow, and affect mixture distribution.

Throttle ice is formed on the rear side of the throttle, usually when the throttle is in a partially “closed” position. The rush of air across and around the throttle valve causes a low pressure on the rear side; this sets up a pressure differential across the throttle, which has a cooling effect on the fuel/air charge. Moisture freezes in this low pressure area and collects as ice on the low pressure side. Throttle ice tends to accumulate in a restricted passage. The occurrence of a small amount of ice may cause a relatively large reduction in airflow and manifold pressure. A large accumulation of ice may jam the throttles and cause them to become inoperable. Throttle ice seldom occurs at temperatures above 38 °F.

Impact ice is formed either from water present in the atmosphere as snow, sleet, or from liquid water which impinges on surfaces that are at temperatures below 32 °F. Because of inertia effects, impact ice collects on or near a surface that changes the direction of the airflow. This type of ice may build up on the carburetor elbow, as well as the carburetor screen and metering elements. The most dangerous impact ice is that which collects on the carburetor screen and causes a very rapid reduction of airflow and power. In general, danger from impact ice normally exists only when ice forms on the leading edges of the aircraft structure. Under some conditions, ice may enter the carburetor in a comparatively dry state and will not adhere to the inlet screen or walls or affect engine airflow or manifold pressure. This ice may enter the carburetor and gradually build up internally in the carburetor air metering passages and affect carburetor metering characteristics.

Ed Holyoke
 
I’m well aware of the different mechanisms of formation/types of induction ice. That was never the question. I’ve done plenty of heat xfer and associated cycle analysis in my professional life. I questioned if the latent heat of fuel is the primary driver versus the effects of low local static P. I don’t see anything in the previous cut/paste that begins to answer that. My previous examples of when induction heat is recommended and others poster’s examples of induction ice encounters backs up that hypothesis. If anyone has real data regarding the relative contributions of each associated contributor, I would like to know. So far, there is no contrary evidence to my previous statements.
 
Regarding;
“2. As others noted, most of the cooling is due to the evaporation of fuel.”

I’d need more data but with a lack of such, this seems doubtful. Low local static pressure levels seem to be the prime driver. Example: Axial Compressors ice at very high ambient temps when inlet guide vanes are less than full open (no fuel evaporation effects). Likewise, carb heat is rarely needed at power levels greater than 80% i.e. throttle valves open/low static pressure loss. Similarly, closed throttles (resulting low local static P added to that of Venturi) where very little fuel flow (very low latent heat losses) are most prone to icing for given conditions.

If someone has related data, it would be appreciated. As of now, I would need some convincing.

Thanks.

Evaporation of liquids is a very powerful cooling mechanism - that’s how most air conditioners work. As you point out, carb ice is a problem with the throttle closed - when air flow thru the venturi, and hence adiabatic cooling, is minimal. What happens is most of those fuel droplets hit the closed throttle plate, and evaporate right then and there, not spread out thru the induction system. Also, you might note that many (most?) fuel injection systems use a venturi to sense the air flow, but almost none of these systems have a ‘venturi heat’ system.
 
Evaporation of liquids is a very powerful cooling mechanism - that’s how most air conditioners work. As you point out, carb ice is a problem with the throttle closed - when air flow thru the venturi, and hence adiabatic cooling, is minimal. What happens is most of those fuel droplets hit the closed throttle plate, and evaporate right then and there, not spread out thru the induction system. Also, you might note that many (most?) fuel injection systems use a venturi to sense the air flow, but almost none of these systems have a ‘venturi heat’ system.

I am aware, Sir. Sorry if I'm being dense but I'm going to be stubborn here. Bare with me.

For a given RPM, the IC engine is a constant volume machine. (Probably) No one will argue that. The volumetric flow through the venturi (thus ~ fluid speed) does not changep with throttle position alone. I’m aware that usually RPM will not stay that constant Air density on the other hand... Your position assumes that the static P effects are limited to the venturi. That is far from true. At some point of closure, the throttle losses become the predominant pressure loss; very high velocities occur around the butterfly valve. Static P drops proportionately to the square of this velocity.

So I'll ask again. Does anyone have data that supports the previous statement that fuel evaporative effects are the primary driver? I'll state again, the engine OEMs operation recommendations and the experiences posted here by others (at least anecdotally) backs that up as does axial compressors icing, i.e.closing inlet guide vanes increases this risk still without no latent heat effects.

Sorry for the back and forth. Respectfully submitted. I’m really trying to understand the entire proportional relationships.
 
Last edited:
I am aware, Sir. Sorry if I'm being dense but I'm going to be stubborn here. Bare with me.

For a given RPM, the IC engine is a constant volume machine. (Probably) No one will argue that. The volumetric flow through the venturi (thus ~ fluid speed) does not changep with throttle position alone. I’m aware that usually RPM will not stay that constant Air density on the other hand... Your position assumes that the static P effects are limited to the venturi. That is far from true. At some point of closure, the throttle losses become the predominant pressure loss; very high velocities occur around the butterfly valve. Static P drops proportionately to the square of this velocity.

So I'll ask again. Does anyone have data that supports the previous statement that fuel evaporative effects are the primary driver? I'll state again, the engine OEMs operation recommendations and the experiences posted here by others (at least anecdotally) backs that up as does axial compressors icing, i.e.closing inlet guide vanes increases this risk still without no latent heat effects.

Sorry for the back and forth. Respectfully submitted. I’m really trying to understand the entire proportional relationships.

As you point out, a closed throttle causes lowered pressure and causes the moisture in the air to freeze out. This is why throttle icing tends to occur at mostly closed throttle and on the back side of the throttle plate where the pressure is the lowest. Fuel vaporization icing occurs at the discharge nozzle and will be more likely to occur at higher throttle settings because the pressure is low enough in the venturi to draw the fuel from the float chamber and cause it to vaporize. lowering the temperature enough to freeze the moisture in the air. With the throttle closed, there isn't enough delta P to draw fuel through the main jet reliably and so an idle circuit with an outlet right by the edge of the closed throttle is necessary. The idle port could still ice at the same time the throttle is icing for a different reason.

So the question that I see you raising is which effect predominates. It depends on throttle position. I don't think it matters much to the pilot which effect is causing the power loss or making the throttle immovable. Yes, full carb heat is what you need for either cause.

Ed Holyoke
 
Last edited:
isentropic vs adiabatic expansion?

Regarding;
“2. As others noted, most of the cooling is due to the evaporation of fuel.”

I’d need more data but with a lack of such, this seems doubtful. Low local static pressure levels seem to be the prime driver. Example: Axial Compressors ice at very high ambient temps when inlet guide vanes are less than full open (no fuel evaporation effects). Likewise, carb heat is rarely needed at power levels greater than 80% i.e. throttle valves open/low static pressure loss. Similarly, closed throttles (resulting low local static P added to that of Venturi) where very little fuel flow (very low latent heat losses) are most prone to icing for given conditions.

If someone has related data, it would be appreciated. As of now, I would need some convincing.

Thanks.


Expanding a gas through the blading of the inlet guide vanes of an axial compressor, which was one of your points, could possibly produce much more cooling effect than a throttle process..
I suggest that that process will be adiabatic and produce a lower temperature that a throttling process will. I am less certain about the difference compared to the carburetor venturi though.

I don't have the numbers for the relative effect of expansion vs latent heat of fuel, but I suppose some one could grind it out using the different ratios.

Learned to fly in Michigan in the winter and early spring in my then C65 Champ nearly seven decades ago. It was stick/throttle/carb heat most days. Carb icing was expected and pretty much ho-hum. Later, early Bonanzas with PS-5's, a dry carburetor with the fuel introduced down stream, no icing. So, I would have to guess that the temperature drop from fuel evaporation is the more important factor. But again, no numbers.

Something related to fuel cooling effect: In Libia in WW2, the standard beer cooler was an empty drum, usually delidded with prima cord, filled with beer bottles and avgas, then bubbled with an air hose. Presto, cold beer due to the latent heat of gasoline. And that is how to win a war.
 
Someday somewhere that carb heat will be needed. If you dont have it available it will become a very high pucker factor experience.

Ive had my 0320 ice up between the hangar and the runway and Ive had airborne battles with carb ice. I know from the previous posts there is a lack of understanding of carb ice and how you deal with it. Flying around with the carb heat on is a bad idea.

Cm

Yup. This past July taking off out of Salida, CO on a chilly Rocky Mountain morning just before sunrise with virga in the sky around Poncha Pass. Barely made it over the low hills around the airport. Around 10.8 gph showing at WOT. Later concluded it was likely carb icing. Was not using carb heat on takeoff. I have a Carb T probe and find I get just a tad more than 10F rise on carb heat...which Im not happy with. Would be interested in designs to get more temp rise.
 
Back
Top