Gremline Flight Safety Report: Aircraft Carburettor Icing

the online flight safety digest

Text Box: carburettor icing

the gremline digest — carburettor icing in aircraft

Understanding Carburettor Icing

A high proportion of accidents to GA aircraft in the United Kingdom are either caused, or contributed to, by carburettor icing. Perhaps a better understanding of carburetion, carburettors and how and why carburettor icing forms would help pilots to avoid that sinking feeling as the engine splutters and the power fades away.


Most pilots have probably looked at the classic graph illustrating the relationship between dewpoint, temperature and relative humidity that shows areas of different rates of carburettor icing likely to be encountered at various power settings. This illustration is included later in this article, but the graph alone does not seem to get the message across. A deeper understanding of the mechanics of carburetion and of ice formation within the carburettor may help.

About Carburettors
Carburettors are fitted to petrol (gasoline) fuelled piston engines to control the supply of fuel to the cylinders. This article will not consider fuel injected or diesel engines. The purpose of a carburettor is to supply the correct quantity of fuel to mix and vaporise with the air being induced to the cylinders so as to form a combustible mixture of the correct proportions. For complete combustion, the chemically correct proportion by weight is about three of oxygen to one of fuel. Oxygen represents about one-fifth of the weight of air, so the correct volumetric proportion of air to fuel is about fifteen to one.
In the cruise, engines are usually run on a weaker mixture than is chemically correct. This ensures that the greatest practicable proportion of the fuel is converted to useful work, thus improving economy. Conversely, at high power settings it is necessary to use a mixture that is richer (i.e. contains more fuel) than is chemically correct to minimise overheating of the valves, cylinder head and pistons and to reduce the risk of detonation.
Although the ratio of weight of air to fuel required remains constant at all heights and temperatures, the necessary volume of air increases as height is gained because the air density (weight per cc) decreases as the atmospheric pressure falls. Thus one should lean the mixture as altitude is increased. The carburettor is designed to supply the engine with a mixture of the required strength and, in complex carburettors fitted to high power aero engines, to do this under all conditions of altitude and power.

There are three main types of carburettors fitted to petrol fuelled piston engines. These are:
(a) Float type carburettors.
(b) Injection carburettors.
(c) Fuel injector pump systems.
We are only interested in the float type carburettor for the purposes of this article.

In a float carburettor the fuel is supplied through one (or more) jets opening into a restriction in the main air duct through the carburettor. This restriction is also known as the choke or the venturi. When the engine is not running, the air pressure in the carburettor choke is atmospheric and the fuel level in the float chamber is held just below the opening of the jets. When the engine is running and air is being drawn through the carburettor, the pressure in the choke falls to an extent depending upon the speed of the airflow through the choke. This fall in pressure due to venturi action sucks fuel from the jets in proportion to the speed of the airflow through the choke. Fuel being drawn from the float chamber causes the fuel level in the float chamber to drop, and the float drops raising the needle valve and admitting more fuel into the float chamber.


Gremline Illustration: A simple updraft float carburettor, a type often fitted to petrol-fuelled piston engines in GA aircraft.

More powerful piston engines, as fitted to ex-military aircraft for example, may have an aneroid controlled needle within the carburettor jets so as to maintain a constant weight of fuel to weight of air at all altitudes. These engines may also have an automatic mixture control to bring in additional jets to enrich the mixture at high power settings by an interconnection between either the throttle lever or the boost regulator.
When the throttle is opened rapidly, a temporary weakening of the mixture occurs until the balance of pressures and fuel flow is re-established. To prevent a ‘weak cut’ in these circumstances, some carburettor systems have an accelerator pump that forces a small additional quantity of fuel into the air duct as the throttle is opened. For aircraft without an accelerator pump, it is up to the pilot not to induce a ‘weak cut’ by too rapid opening of the throttle.
Most of the petrol engines in use by GA aircraft in the UK will have a simpler type of float carburettor fitted with a manual mixture control that must be set by the pilot to provide the mixture required for each operating condition.

Engine Icing
There are two distinct types of icing that will affect aircraft engine performance, impact icing and carburettor icing. Each occurs in a totally different manner and may occur singly or both at the same time. They can occur with outside air temperatures (OAT) over a large range, typically between –15C and +25C.
Impact icing occurs at the same time and in the same manner as airframe icing, by the freezing of super-cooled water droplets as they strike the airframe. Impact icing can affect engine performance by gradually blocking the engine air intakes. This upsets the mixture reaching the cylinders by reducing the airflow through the intake, causing a drop in boost and finally stopping the engine. There is then a considerable risk of an engine fire due to neat fuel entering the induction system.
The OAT required for the formation of impact ice is sufficiently low for the air reaching the carburettor to be further cooled by expansion and refrigeration and thus drop to a temperature so low as to prevent the ice sticking to the sides of the venturi and causing carburettor icing. Fitting an ice guard ahead of the intake mouth will prevent intake blockage and overcome the threat of impact icing on engine performance. If your aircraft does not have an ice guard (or some other mean of preventing intake icing) it will not be cleared for operation in known icing conditions. You should have decided upon a clear and carefully thought out plan of action to follow if you encounter airframe icing. Is it better to climb or descend? That’s probably your first decision.



The Mechanisms of Carb Icing
There are three mechanisms involved in the formation of carburettor icing. The first is
throttle icing. The constrictions at the throttle valve and the choke venturi in the induction system cause a local increase in the speed of the airflow. This causes a drop in both temperature and pressure and may cause ice to begin to form inside the venturi and around the throttle butterfly valve. Each grain of ice that forms will further reduce the size of the venturi. This increases the speed of the airflow and further reduces the temperature, thus accelerating the rate of ice accretion inside the carburettor. The temperature reduction may be as great as 25C, but differs in different types of carburettor.
The second mechanism involved is
refrigeration icing. When fuel is injected into the venturi a certain amount evaporates. The heat required for evaporation is taken from the surrounding air and metal, thus further reducing the temperature of the mixture passing the throttle butterfly. This temperature drop due to refrigeration can be considerable and can, alone, cause ice formation in the carburettor when the OAT is well above freezing.
The final mechanism is
water evaporation. When flying through visible moisture like cloud or rain, some of this moisture will evaporate in the carburettor, further reducing the temperature of the airstream inside the carburettor. The actual temperature drop caused by water evaporation is quite small and not a great hazard by itself, but the additional drop due to water evaporation may be enough to bring the temperature in the venturi to below freezing point.

It is worth mentioning in passing that MOGAS is more prone to carburettor icing than AVGAS.


Anticipating Carburettor Icing

Now that we know why carburettor icing occurs we need to examine when it will be most likely to occur. The amount of carburettor icing encountered depends on the relative humidity of the outside air. The higher the relative humidity, the more likely is the occurrence of carburettor icing, because of the amount of water vapour in the air. The actual temperature of the air has very little to do with the risk of carb icing. For example, air having a relative humidity of 50% at +10C will have a relative humidity of 100% (saturated) at 0C. Any further drop in air temperature will cause condensation to take place below freezing point, and ice will form. In general, the lower the temperature of the outside air, the smaller will be the amount of moisture it contains. That is, cold air is dry --- and warm air is moist. Therefore, you are more likely to suffer carburettor icing when flying in clear, cloudless warm air than you are in clear, cloudless cold air. That is an important point not always appreciated by private pilots. Perverse as it may seem, you are more likely to suffer carb icing on a warm day than on a cold day.
If the relative humidity is high (i.e. the dewpoint and OAT are close together) large amounts of ice may form in the carburettor when no visible moisture is present. This can occur in outside air temperatures as high as +25C to +30C, although dewpoints above +20C are unlikely in North West Europe. If the overall temperature drop in the carburettor due to the mechanisms discussed above is greater than the OAT, then carb icing is likely. That is, carb icing can be encountered when the OAT is as high as +25C if the air is close to saturation and the temperature drop inside the carburettor exceeds 25C. There are some types of carburettor having a possible temperature drop of 45C.



Dealing With Carb Icing
A slow drop in boost or a gradual decrease in engine RPM indicates the formation of carburettor icing, depending on the engine/propeller combination on the aircraft. If an automatic boost control is fitted, the throttle will be progressively opened until it is fully open and there will probably be no indication of carb ice formation until it has reached a serious level. The engine governor on the Robinson R22 helicopter can mask carburettor icing in this manner — we will be covering this specific problem in a later issue of Gremline.
The longer it takes the pilot to recognise the beginning of carb ice formation the more difficult it will be to get rid of the ice before it causes the engine to lose a significant amount of power.
The vulnerable parts of many carburettors in high-powered engines are heated by the circulation of engine oil or coolant to raise the temperature around the throttle valve and around the carburettor barrel in the area of the venturi. As long as the engine controls are handled to maintain the heating medium to above +60C this system is efficient and the slight heating of the charge has little effect on the power output of the engine. In the earlier float type carburettors, as fitted to most GA aircraft, fuel is injected into the airstream before it passes the throttle venturi and valve. Few of these carburettors have barrel heating and the only means of overcoming ice formation is the hot air intake control. Even with barrel heating, it is worth bearing in mind that selecting carb air to
HOT soon after start-up may not be effective until the engine has warmed up sufficiently to heat the warming medium to above about +60C. Do you know the source of hot ‘air’ for carburettor heating on your aircraft? Is it provided by air from around the cylinders or do the exhaust pipes warm the air? How long after start does it take for the ‘hot air’ to become hot enough? Is that another good reason to ensure a minimum engine oil temperature during your after start checks, or is cylinder head temperature more significant? The application of ‘hot air’ for a short period soon after start may not be enough to heat the carburettor sufficiently to avoid icing, particularly when the outside air is close to saturation and you are parked on wet grass.
Most engines are fitted with a hot/cold air intake control. Hot air should be selected whenever carb icing is suspected, or when flying through sleet, snow or heavy rain, even if the OAT is as high as +20C. These conditions may apply during cruise, descent, approach and while taxiing, particularly when taxiing across wet grass. Pilots should refer to the operating instructions for the engine installed in their particular aircraft for the correct use of the air intake control. These may vary considerably from type to type and using an inappropriate technique learned on another type of aircraft, or even the same engine in a different aircraft model, may well lead you into danger. Don’t be shy about asking a qualified engineer to show you the carburettor hot air system on YOUR aircraft and to explain its operation.

The picture below shows a typical system for providing hot air to a carburettor.

Gremline Illustration: A typical system for delivering hot air to an aircraft carburettor.

This is the installation in a Cessna 182 and illustrates the simplicity (crudeness?) of a typical general aviation aircraft. The baffle is wrapped around part of the engine exhaust pipes, thus collecting hot air when the exhaust pipes are hot. This air is then fed through the (red) hot air pipe towards the carburettor. A simple gate valve, manually controlled by the pilot, may be opened or closed to allow either hot or cold air into the carburettor. The weaknesses of this system include the fact that the pilot has to recognise either the possibility of ice formation, or catch the beginning of ice formation, and then take the correct action to either prevent ice forming or to get rid of ice that has already formed. Another obvious disadvantage of this system is the fact that it will only provide air at a temperature dependant on the heat of the exhaust pipes. This means that a system like this may not be efficient for some time after start. It should also alert pilots to the need to select hot air BEFORE reducing power for a descent or approach. It is bad airmanship to close the throttle first and then select hot air to the carburettor.
The sharp-eyed reader will have noticed that the hot air pipe in the picture above is not properly connected to the baffle. The aircraft was being serviced and the pipe was disconnected. I placed it in position temporarily, for the purpose of the picture.
My thanks to Haverfordwest School of Flying for making the aircraft available for photography and for the engineering assistance and discussion.



It was interesting to examine a de Havilland “Domine” in a Haverfordwest Airport hangar. This aircraft was built as a de Havilland “Dragon” in 1929 and later modified to become a “Domine.” It still retains its original 200hp de Havilland Gipsy Queen engines. These are equipped with a fully automatic carburettor icing prevention system. Also, the air driven flight instruments are driven by suction in the engine exhaust system, so there is no possibility of the loss of flight instruments due to impact icing of the instrument system venturi. The “Domine” has almost exactly the same capabilities and performance as the Britten-Norman “Islander” built many decades later. If de Havilland could provide fully automatic ice protection for the carburettors of their Gipsy Queen engines in 1929, why are we still flying about with bits of wire connected to tin ducts that the pilot has to remember to operate? Answers on a ten dollar note, please!

There are some conditions in which the use of hot air may actually start the formation of carburettor icing instead of preventing it. If the refrigeration and throttle icing cooling effects (see above) in the carburettor are high and are assisted by the OAT and dewpoint, the final temperature drop may be so high that the temperature inside the carburettor is so low that ice will not stick. In these conditions, the use of hot air may raise the temperature back into the range at which the ice will stick and thus induce carburettor icing.
Finally, I reproduce the classic OAT, dewpoint, humidity graph below and hope it now makes more sense than the last time you looked at it! Note that on a ‘standard day’ with the OAT at +15C and with a dewpoint at +10C you will be operating in the area of ‘serious icing at any power. If the dewpoint is as low as +5C you will still be on the edge of the area of serious icing at any power.

Gremline Illustration: Outside Air Temperature, Dewpoint and Humidity.

The above diagram shows the degree of carburettor icing likely to be encountered at various combinations of outside air temperature and dewpoint. It may be worth reminding pilots that the closer together are the OAT and the dewpoint the greater is the saturation of the air i.e. the more moisture there is in the air.


The red area on the diagram indicates that there will be SERIOUS ICING AT ANY POWER SETTING.


The orange area indicates that there will be MODERATE ICING AT CRUISE POWER and SERIOUS ICING AT DESCENT POWER.


The blue area indicates that there still will be SERIOUS ICING AT DESCENT POWER.


The green area indicates that there will be LIGHT ICING AT CRUISE OR DESCENT POWER.


The dotted horizontal line at about +22C dewpoint indicates that this is the probable upper limit of dewpoint encountered in NW Europe.


I suggest that your pre-flight planning should always include a careful look at the OAT and dewpoint relationship. Armed with that knowledge and a clear understanding of the carburettor hot air system on YOUR aircraft you should avoid problems with carb icing and we will have fewer GA accidents related to carburettor icing.



Glossary (For our more comprehensive Glossary of aviation and technical terms, click here.)
dewpoint. The air temperature at which atmospheric moisture will condense to form vapour (fog/cloud).
relative humidity. The percentage of moisture in the air compared to that required to produce saturation, which occurs at 100% humidity.
venturi. A narrowed section of a pipe where the speed of any fluid flowing through the pipe is increased and the pressure is reduced. (G.B. Venturi, Italian physicist d.1882)


PS. I was surprised to find a hand-held battery-operated digital combined thermometer and hygrometer that gives instant readings of air temperature and relative humidity, simultaneously. The ranges of this gadget are from 0 to +50C and relative humidities from 25% to 95% so it covers the areas where carburettor icing is likely to be encountered. It is made by Digitron and runs off AAA batteries. Perhaps this would be a useful gadget for your flying club or flying school to have available – so long as the readings are taken outside!



Search Now:
Amazon Logo



Text and Photographs 2008 Gremline & Hill House Publications, unless otherwise stated.


landing page about gremline copyright/conditions/contact information exchange glossary

uk emergency diversions uk links, chirp & gasco global & misc links forum

the gremline cockpit — index of articles the gremline bookshop top of page