Tanis Aircraft Products - Articles

Articles & Research

Heating Aircraft Reciprocating Engines

Peter G. Tanis

Tanis Aircraft Service, Inc.

ABSTRACT

Aircraft engines need preheating to be able to start without damage in cold weather. It is possible to do this efficiently by placing a small amount of heat in the proper places. An installed electric system is very convenient and easy to use. A good design will be lightweight and energy efficient. Corrosion in engines that operate in cold weather can be reduced or eliminated by the right preheater design and the proper operation of the engine. Some designs of preheater systems aggravate other problems. A certain amount of caution is needed in selecting the system. The unit should be specifically matched to the aircraft and engine.

INTRODUCTION

When engines are designed they must meet certain minimum standards. Here in the USA they are designed to meet FAR Part 33. This standard does not address cold weather operation, nor does the airframe standard FAR Part 23. Since this issue is not addressed in the FAR's, when a preheater system is approved by the FAA it may be subject only to the ideas of the approving engineer. It may not have to demonstrate that it even operates. Other countries have more stringent standards in this area. I will attempt to show practical standards and ways to meet them.

Since everyone believes that it gets cold where they live, we must define when preheating is necessary:

If the goal is the prevention of engine damage on the start and the oils recommended by the engine manufacturer are used, then preheating should be done below 20 Degrees F (-7 Degrees C).
If the goal is easy starting, then preheating below 40 Degrees F (5 Degees C) is beneficial.
If the goal is reducing stresses, wear in the cylinders on the start and to reduce "run up" time, then preheating below 60 Degrees F (15 Degrees C) can help.

Air-cooled aircraft engines have always been hard to start in cold weather. Preheating has been used to start them almost as long as there have been aircraft. One early method was to build a charcoal fire, park the aircraft over it and cover it with an engine tent. The WWII "Herman Nelson" was a gasoline-burning portable furnace. A single cylinder engine ran a fan and air ducts were placed in the cowling. Soon after this a propane torch with an electric air blower appeared on the scene. All of these Systems were external to the aircraft.

Early attempts at installed heaters were made by mounting car warmer electric heaters or by putting light bulbs in the engine cowling. In the early 1970's I patented an electric preheating system installed on the engine itself and obtained the various FAA approvals needed to install it on certified aircraft.

There are many misconceptions about starting cold engines. There is the idea that the engine is difficult to turn over because the "oil is stiff". There is the idea that warm oil is equal to a warm engine. Some think that there is some benefit from having oil warm even beyond that of obtaining proper viscosity. Some believe that preheating is responsible for moisture and corrosion in engines. I will attempt to address these issues.

One must understand what happens in each part of the engine during a cold start in order to know how to address proper preheating. There is an idea afield that it's all a matter of total BTU's delivered to the cowling and that it doesn't matter where the heat is applied. I will attempt to show that a properly designed preheater can be effective by placing a small amount of heat in the right places.

MAIN SECTION

WHAT COULD POSSIBLY GO WRONG?

This discussion mentions many "worst case" situations. Not every engine will exhibit the same type of failures. New engines seem to be more likely to be damaged on a cold start due to close internal clearances, whereas old "high time" engines are more tolerant of a cold start.

In the Cylinder

It all starts with combustion. If you don't have combustion then you don't have anything. The basics of combustion have always been fuel, air and ignition and a cold start can interfere with two of these. It requires heat from somewhere to vaporize fuel. If the cylinder head and induction tubes are cold the fuel will not vaporize into the fuel/air mixture needed to start. After repeated compression strokes, due to long term cranking, you may raise the temperature enough to start. Once the fuel is vaporized enough for starting you're in for another problem - Ignition.

One of the by-products of combustion is moisture. The spark plug will fire the fuel/air mixture one time and then go dead. When the spark plug fires, this moisture from the combustion condenses on the cold spark plug electrodes and freezes, forming an ice bridge across the gap. Further starting attempts are useless, because heat is needed in the cylinder head to melt the ice.

An aircraft cylinder is constructed with an aluminum head tightly fit on a steel barrel. The barrel is usually ground so that the top end is about .003" smaller than the base. This is called "choke" and allows the cylinder wall to be "straight" at operating temperature. The coefficient of thermal expansion of aluminum is 12.9 x 10 to the -6th per Degrees F. Steel is 6.5 x 10 to the -6th per Degrees F, almost exactly one half the rate. When the cylinder is cooled, the head casting shrinks twice as fast as the barrel. Tests that I conducted In February 1984 showed a measurable .0057" increase in choke at -15 Degrees F (-26 Degrees C), which means that this cylinder could be .009" smaller at the top end than at the base at this temperature. As temperatures drop this choke will increase.

The piston in the cylinder is made of aluminum and has a tightly fit wrist pin made of steel. This fit is typically .0001 to .0007 loose at room temperature. As the temperature drops the wrist pin is locked in place by the shrinking piston. In my 1984 tests I documented a .001" decrease in this clearance at -20 Degrees F (-29 Degrees C). Obviously this piston pin is locked in place by the interference fit. If the piston stopped near the bottom of its stroke the pin may be extended out one side of the piston against the wall. Consequently, as the engine cools down, the pin is locked in position against the cylinder wall. When a cold start is attempted, the piston pin button and the opposite side of the piston are forced into the smaller diameter top end of the cylinder bore. This start can produce scoring and possible ring breakage. If the cylinder has been cold started you can see the track left by the piston pin button on the cylinder wall. Should the engine have a "summer grade" of oil in it then the piston rings could be expanded in their groove with stiff oil under them. These rings could be subject to breakage when forced into the choked portion of the cylinder bore.

Once combustion occurs then another unexpected process starts in the cold cylinder. Combustion produces a 1200-1400 DegreesF (456-532 Degrees C) flame in direct contact with the aluminum piston, which produces exponential growth of the piston while the cylinder wall warms and grows more slowly. The result is more piston and ring scuffing.

Some people are very concerned with sudden cooling changes called "shock cooling" in an operating engine. If this is perceived as a problem, then when you start an engine with cold cylinders the sudden temperature and dimensional changes in the cylinders should be a concern.

In the Crankcase

The crankcase of an aircraft engine is made of aluminum. The rotating pans within it are made of steel. When you consider the coefficient of expansion of the two metals, it becomes apparent that the crankcase shrinks more than the metal parts within it as it cools. This affects many things: the crankshaft main bearing clearance decreases, accessory drive gear clearances reduce, cam bearing clearance decreases, and cam followers fit more tightly in their bores. An IO-52O Continental engine has a crankshaft to main bearing clearance of .0005 to .0035" loose at room temperature. At -110F (-240C) all bearing clearance is gone, so in temperatures below this, the bearing is in interference fit. Because at these temperatures the engine is difficult to "turn over", people attempt to "prop" them and often think that they turn hard because the "oil is stiff". It isn't the oil. The fit of the bearings, gears and pistons is the problem. It's possible to rotate a main bearing insert by hand propping in these temperatures. Should the engine start at temperatures close to this it could be operating with no bearing lubrication due to lack of clearance.

An engine may shear the bearing-locating pin which allows the bearing insert to rotate. As the insert rotates it closes off the hole that supplies oil to the bearing. Some of these bearings are part of the oil passage from one oil galley to the other, so when this bearing rotates one half of the engine is without oil. A massive failure follows shortly. An insidious variation of this type of failure occurs when the bearing insert moves only slightly. The metal of the bearing "piles up" on the rotation side of the locating pin, which causes a hot spot due to contact with the crankshaft. As a result the shaft may be worn or the bearing may adhere to the shaft and rotate further. This partial movement may also restrict oil flow due to the hole in it being partially restricted. This type of failure will not occur for many hours, possibly into the next summer when no one even remembers that it was cold started.

Some engines that are cold started only wear the bearing surface away. The nose case bearing is a common offender for this type of failure because the propeller is a large heat sink attached to the crankshaft near this bearing. In cold weather propeller losses make the nose case bearing hard to warm up, and engines with friction bearings are the worst offenders. Engines such as radials which have ball or roller nose case bearings are not affected as much. On engines with constant speed propellers the propeller governor supplies oil pressure through this bearing. An engine that has been cold started will have a large bearing to crankcase clearance and will leak prop governor oil back into the crankcase. Twin engine planes will have trouble keeping props in sync since the prop governor oil is being lost.

In the Oil Sump

I once thought that you could preheat an engine by heating the oil. My early experiments and attempts at this convinced me that there were better ways. The oil in an aircraft engine is held in the oil sump, which is well below the rotating parts, or in the case of the dry sump engine, in a separate oil tank. The hard cranking of a cold engine is more related to clearances and fits than to "hard oil". After the engine starts it needs oil flow for lubrication. This flow is the result of having the right viscosity oil available to the intake of the oil pump and the proper clearances in the bearings and parts that it lubricates.

One of the "claims to fame" of the modem multi-viscosity oils is the ability to maintain a constant viscosity over a wide range of temperatures. Consider the common automobile engine: it starts well down to -40 Degrees F (-40 DegreesC) with the use of an "in block" heater which heats the cooling liquid surrounding the cylinders. The engine has no oil heat.

Aircraft engines do have some minimum temperatures below which they will no longer lubricate properly. These oil temperatures are related to the particular type of oil used. The oil viscosity needs to be liquid enough to flow through the oil pump inlet screen and the small passages in the lifters and push rods. In extremely low temperatures the oil filter is usually bypassing oil, thus allowing the viscous oil to lubricate the engine even though it can't pass through the small passages in the filter. If the engine should start with cold cylinder heads it may be as much as fifteen minutes before the valve guides get lubrication, even if the oil sump is warm and has liquid oil. This is due to the small oil passages in the push rods, lifters and rocker arms that supply oil to the rocker area. It takes some heat in the head to allow this oil flow. Some valve guides require lubrication immediately upon starting to avoid damage so fifteen minutes is much too long.

The maximum viscosity at which oil will lubricate properly is 33,000 centistokes. As oil approaches this number then it needs thinning and a little heat is a good way to do it. This viscosity is reached at different temperatures by different oils. Examples of these temperatures are as follows:

15W50-----------	0 Degrees F (-18 Degrees C)
65W---------------	10 Degrees F (-12 Degrees C)
20W50-----------	15 Degrees F (-10 Degrees C)
100W-------------	35 Degrees F (2 Degrees C)

When starting at -30 Degree F (-34 Degrees C) with the most viscous oil (100W), you need to raise the oil temperature at least 65 Degrees F (36 Degrees C) above the outside air temperature. This problem can be taken care of by applying a very small amount of heat to the oil sump. But this being done, then the other parts of the engine still need attention before starting because cold starting can cause other problems.

In the Accessory Section

One of these problems can be the gear fit of the accessory drive. There is also the chance of a magneto driven by an impulse coupling not firing. This impulse pawl may not move due to being coated with stiff oil. There is also a higher incidence of vacuum pump drive failure and mechanical tachometer cable breakage.

HOW TO PREHEAT THE COLD ENGINE

There are several areas that need to be warm with any preheating. First are the cylinder heads, since combustion and so many problems occur there. Secondly the crankcase and the accessory section need to be warm enough to provide bearing clearances. The "nose case" just behind the propeller is the hardest part of the crankcase to warm. Lastly the oil sump needs some heat at very low temperatures.

It's easy to overheat the oil sump thereby producing changes in the oil that are not beneficial. In an engine with a thin aluminum oil sump it's possible to heat the oil until it is quite hot but still have no significant heat transfer to the crankcase. On engines with a thick cast aluminum sump, much of the heat applied to the surface of the casting is transferred into the crankcase. On some of these engines with cast sumps the bottom of the casting is an induction housing and the oil is in a chamber above. On these sumps the bottom provides an excellent location on which to apply heat. Heat applied here will improve fuel vaporization, transfer heat to the crankcase and provide a small amount of heat to the oil.

Very little of the heat placed in either the oil sump or the crankcase migrates to the cylinder head. Some low priced single point sump heaters claim to heat the whole engine, but to be effective at all they need to insulate the cowl with a very good insulated engine cover and to operate for long periods of time. Convection in the air within the cowl and limited conduction will eventually transfer some heat to the cylinder heads. Extremely cold temperatures or severe wind chills usually render these systems ineffective.

Air Blowing Heaters

It is possible to preheat using one of the hot air blowing heaters. To do a proper job it normally takes 100,000 BTU's for 15 to 20 minutes to preheat an engine. There are small propane units that claim to produce 100,000 BTU's but they are fueled by a small disposable bottle that has 6,000 BTU's available. These units are too small to he effective in all but mild conditions. Another thing about propane powered preheaters is that at low temperatures propane will not flow from the bottle. A heater of the proper size and a propane bottle of sufficient capacity will still not produce enough heat if the bottle is cold. If preheating this way care is needed as to where the ducts are placed. Putting them in the cowl flap opening is usually the best place. Improper placement of the ducts can cause blistering of the paint on the cowling, melting of plastic push-rod tubes or damage to the ignition harness.

When the engine is warm enough to start, the cylinders will feel warm to the touch; oil will drip freely from the dipstick and the oil temp gauge will have some indication. When preheating in this way the oil is usually the last thing to be warmed, therefore if the oil is warm the engine is ready to start. This is not necessarily true with installed electric systems.

Electric Preheaters

An installed electric preheating system can be very effective and easy to use. These are normally left on the engine year round and do not operate in flight. They can be connected to readily available electric power. They are designed to heat the engine more slowly than the air blowers but they do not have to be attended and can be left on for prolonged periods.

In The Cylinder

Many of the problems that occur are in the cylinder head. Heating the cylinder head casting addresses all of the cylinder problems such as fuel vaporization, spark plug frosting, excessive choke, and piston ring scuffing.

To raise the temperature of the cylinder head casting it's most practical to place a heater directly in contact with the cylinder head. There is some heat transfer from one cylinder head to another by conduction through the crankshaft, rods and pistons. It is possible to heat five cylinder heads on a six cylinder engine and to have the sixth cylinder warmed by conduction. However temperature rise on the sixth cylinder is slower, and it doesn't get as warm as the others. Heating all of the cylinder heads is the best way to do it. To heat the cylinder head to approximately 100 Degrees F (55 Degrees C) above the outside air temperature is the ideal situation and all the problems have been addressed. The cylinder wall is then relatively straight and the engine is ready to start.

There is one other variation of heating for the cylinder. Heat has been placed at the base of the cylinder barrel. This has the same problem as heating the sump or the case in an attempt to raise cylinder head temperature. It isn't very effective in heating the cylinder due to the heat path to the head. It uses the cylinder barrel, which has cooling fins, as a path for heat conduction. Placing the heat here is more effective in raising the case temperature than it is in raising the temperature of the cylinder head and it also works counter to the effort to reduce choke since it expands the base of the cylinder with little expansion of the top end. This may be better than a sump heater alone but is not as effective as heating the head casting directly.

The following types of heaters have been used to heat the cylinders:

A heater placed in the thermocouple well on the cylinder head. This device is a brass body with a heating element inside. The brass and the aluminum of the head have almost the same coefficient of expansion so that as heat is produced they both expand at the same rate. This location is very effective in producing a fast start since it is close to the lower spark plug. The problems of plug frosting and fuel vaporization are easily corrected by this element. Since the aluminum head is a very good conductor of heat the entire head quickly comes up in temperature. Some people question if this "localized heating" can crack a cylinder. In approximately 175,000 of these cylinder heater installations I have never seen even one with cracking caused by them. I believe this is due to the common coefficient of expansion and the good conduction of the head.
A heater attached to the head casting. Some cylinders do not have a thermocouple well. For these cylinders heaters have been made that am extruded aluminum blocks with a heating element inside of them. These are usually attached by bolting to some part of the head casting. Some have been made that fasten behind the rocker box area mounted with a rocker cover bolt. Others have been used on radial engines bolted to the "speed ring ears". These operate by transferring heat by conduction to the head casting and depending on the excellent heat transfer of the casting to get heat to the combustion chamber.
A heater placed in an intake bolt. Where the above methods do not work there are other ways. Heated bolts have been used in place of the bolts that hold the induction tubes to the engine. These do a particularly nice job of heating the induction port and they help fuel vaporization. They are slightly slower than the thermocouple well heaters in heating the combustion chamber.
A heater under the rocker box cover. Some installations have the thermocouple wells used by other systems such as CHT gages. A recent development has been the gasket heater. This device is shaped like a rocker cover gasket and has a heating element inside. One surface of this heater is aluminum and the other is silicone rubber. The silicone side is bonded to the cylinder head rocker cover parting surface. An oil-proof high-temp RTV is used. The aluminum side becomes the new parting surface for installing the normal gasket and rocker cover. This allows for heating the cylinder head and normal removal and installation of the rocker cover. The power lead attachment of this device is particularly rugged and it should have a very long life. It can be removed and reused. It is very effective in heating. The thermocouple well heater is the fastest in producing a start but this device is a close second.
A heater on the base of the cylinder. As mentioned earlier one type of heater has been developed which installs on the base of the cylinder. It is a modified worm clamp with a silicone heater bonded to it. It is the least effective of any of the above in raising the cylinder head temperature but it does nicely raise the case temperature. The installation is around the base of the cylinder just below the cooling fins. This is a highly stressed area with a radiused transition to the bolting flange. The worm clamp is tightened against this area. It holds the inter cylinder baffles slightly away from the cylinder. The worm clamp is stainless steel which has a different coefficient of expansion than the steel barrel. As the engine operates this band tightens and loosens.

In the Crankcase

An engine with a cast aluminum sump will transfer much of the heat placed on the sump casting to the crankcase. In many cases there will be enough heat to address the needs of the crankcase. On engines that do not transfer heat well from the oil sump or that have large propeller losses, heat needs to be applied to the crankcase directly.

A practical design standard for a preheater is to have the "nose case" and the top of the crankcase at least 30 Degrees F (-1 Degrees C) upon starting. In the past there has been some concern that propeller driven air in the cooling air box of the engine would drop the nose case temperature upon the start. The concern was that this drop would temporarily reduce the bearing clearance. I have run tests on this and found it not to be a factor, as the nose case temperature tends to stay about the same until it begins to rise. I believe this to be due to the large mass of the engine.

Crankcase heaters are usually a silicone pad heater which is made of a layered fiberglass mat that is impregnated with silicone rubber. Between these layers is a heating element. These elements are flexible and can be made in practically any shape and in a wide range of voltages and wattages. They can be made to fit complex shapes. These adhere to the crankcase casting and they transfer heat by conduction. These heaters can be installed in three ways: they may be attached to the casting by curing in place in an autoclave; they may be attached by using RTV as an adhesive: or they may be made with a pressure sensitive adhesive which allows installation by pressing them in place.

The autoclave method produces the best bond and the best heat transfer. This method requires that the engine be apart and is best suited to new construction. The RTV method produces a good bond and has good heat transfer if the installer uses the right technique. Heaters installed in this way can be removed and reinstalled. This method is well suited to "field installation". The third method using pressure sensitive adhesive has the poorest bond and the poorest heat transfer. They are very easy to install but they may not be reused and if too much time elapses between manufacture and installation they will not adhere.

In the Oil Sump

In heating the oil sump the main purpose is to allow oil flow at those temperatures where the viscosity of the oil is thicker than the 33,000 centistokes. A practical design standard is to raise oil temperature to at least 40 Degrees F(22 Degrees C) before starting. If oil is overheated detrimental things can result and extreme overheating can cause a carbon build up on the heated surface. This carbon insulates the heating element from the oil and causes an even higher temperature and more carbon formation. Upon engine teardown this carbon build up can be observed on the back side of the piston. If exposed to temperatures that are lower, over a period of time, oil still deteriorates but in different ways.

Typically oil is changed at 25 or 50 hours of operation. The reason that oil is changed is contamination and its exposure to heat over time. Long term exposure of oil to heat can shorten the useful life of oil.

One type of sump heater is a heater that replaces the oil pump inlet suction screen retaining nut which is located on the tower portion of a cast aluminum sump. On first examination this appears to be an immersion heater but its design transfers much of its heat to the casting. This is very effective in that the wattage can be kept low and it heats the first oil available to the pump. It also produces small oil flow since the oil passage above it rises vertically. As the oil is heated it rises and works past the small clearances in the oil pump. This flow can be demonstrated by checking the temperatures in the oil galley above the pump. They rise as the heater operates. This flow varies between individual engines). Because of this flow tendency oil pressure comes up rapidly upon the start with these engines. The oil temperature in the sump of the main oil supply will be lower than the initial oil supplied to the pump. This is not of any consequence since the oil pressure and temperature come up nicely on the start. On small engines this heater alone is enough heat to raise oil sump and crankcase temperatures to a safe level.
Some engines require more heat in the oil sump. A second type of element can be used either alone or in addition to the one above. Certain engines have two drain bosses cast into the oil sump and they use one of these for a quick drain. A heater element is made to fit into one of these openings. This element also appears to be an immersion heater but it works differently. It also transfers much of its heat into the casting.
On engines with thin aluminum oil sumps the above heaters will not work well since there is no good heat path away from the element. To heat this type of sump the silicone pad type heater works well. It needs to be made so that it has enough surface area to avoid high surface temperatures. This type of element may also be used on cast sumps but doesn't have the oil circulation advantage of the screen heater element. It may also be used to heat remote oil tanks by bonding on the outside surface. Various manufacturers have made many designs of these pad type elements. Some of these attempt to heat the entire engine by putting high wattage on the oil sump This is an attempt to warm the cylinders by the excess heat from the sump. Since some of these units produce temperatures of over 300 Degrees F (149 Degrees C) they often have a thermostat. This is an attempt to avoid carbon build up on the surface. I tested one model with a thermostat designed to regulate at 160 Degrees F (71 Degrees C) and found it to have temperatures on the surface in excess of 330 Degrees F (166 Degrees C) before it regulated. After long term operation the oil temp was much higher than advertised and this excessive heat will cause problems which we will discuss later in this paper.
One type of sump heater is a calrod element which is designed to fit tightly around the oil sump. This unit runs at a very high temperature and it is similar to a charcoal starter or an electric branding iron. The temperature of both oil and sump gets very high. This too is an attempt to heat an engine from one point and is excessive heat for the oil sump.
Oil sump heaters have been made which are aluminum blocks that are attached to the sump with epoxy. These are high wattage units with the heat in a localized area. They attempt to heat the engine with one element and have the same problems as the higher wattage silicone elements. They also make a rather large protrusion on the surface of the sump.
Oil sump heaters can be made which are gasket type heaters similar to the rocker gasket heater discussed above. This element replaces the gasket at the parting surface of a cast oil sump and the crankcase. It will address both the oil heat needs and the heating of the crankcase. Since the oil sump needs to be removed to install it, it is best suited to new engine or overhaul assembly.
Oil immersion type elements are sometimes used but are usually limited to remote oil tanks made from rubber bladders.

Other Places

There are other special heaters made for addressing special problems. Specially shaped silicone heaters have been placed on accessory drives, on air/oil separators, on oil coolers and on breather tubes. Oil lines have been wrapped with strip heaters and insulated. Each application depends on the needs of the particular installation.

IS OIL HEAT MORE OF AN ASSET OR A LIABILITY?

Consider again the automobile engine. How can they start at low temperatures without oil heat? In times past oil heat was tried. Remember the dipstick heater? These devices were not very effective at starting and they formed carbon on the element, some times so much carbon that they could not be removed from the engine. After long time usage they changed the engine oil to a milky appearing emulsion as moisture in the sump mixed with the oil. These engines today depend on multi-viscosity oil instead of heat.

Aircraft engines need a little thinning of the oil at very low temperatures due to the types of oil used. In fact, some aircraft engine text books say that heat is always detrimental to engine oil when it comes to lubricating qualities and cleanliness of the engine. When oil is exposed to heat it begins a deterioration called oxidation. The higher the heat and the longer the time of exposure, the more oxidation occurs. Higher temperatures cause accelerated rates of oxidation, which is a chemical change in the oil. This change occurs at lower temperatures than the forming of carbon on the heated surface discussed above. When oxidized oil is mixed with moisture an acid is formed that can attack the engine.

Moisture is a byproduct of combustion. It has been said that if you burn a pound of fuel one of the byproducts is a pound of moisture. This moisture normally goes out of the exhaust stacks. Some of the byproducts of combustion also go by the piston rings in the form of blowby. If you observe an aircraft engine in flight in very cold air you can see this happening. There will be contrails from the exhaust and an equally large contrail coming from the breather.

One other moisture source is the air drawn in through the breather when the engine cools down after shut down. This is a very minor source and can be ignored for the purposes of this discussion.

Oil has an affinity for moisture and will absorb available moisture. If the oil temperature is high enough the moisture will be released and exit through the breather. Operational experience has shown that if an oil temperature of 180 Degrees F (82 Degrees C) is maintained on the CHT gage, and the flight is long enough, the oil will be dried out and have very little moisture in it. Oil temperature in the engine internally is actually higher than this. If the temperature is below 180 Degrees F (82 Degrees C) or the flights are not long enough the moisture will accumulate in the oil.

I have tested used aircraft oil by letting it stand until the moisture appears on the bottom of the container. The oil was then spooned off and the remaining liquid was tested with litmus paper. The test showed acidity. If this oil with moisture is exposed to heat long enough the rate of acid formation will increase.

THE CORROSION ISSUE

For corrosion to occur inside of the engine several things must be present: a corrosive agent (such as acid, sulfur or chlorine compounds), oxygen, an exposed surface, and moisture. The corrosion may be the result of a direct chemical attack or an electrolysis between two metals with a common electrolyte (moisture). Both moisture and the corrosive agent may be produced by not operating the engine warm enough and by exposing the oil to excessive heat for a length of time. Moisture alone is not enough to produce corrosion. If the engine contains the oxygen, the acid and the moisture, will it corrode? Possibly not. It must also have an exposed surface that is not protected by oil film or a build up of varnish. A new engine with its freshly ground surfaces is most likely to corrode in these circumstances.

Some people have accused preheaters of being the cause of corrosion in engines. They claim that leaving them on "produces" moisture which corrodes the engine. This is overly simplistic and inaccurate. If the engine has been operated property with the oil temp high enough, the engine can be preheated for any length of time and no moisture will be released from the oil because the oil has none to release. Even if the preheater over temped the oil and began the oxidation process there would be no moisture released. I know of one engine that had a preheater left on inadvertently for two years. There was no detrimental effect.

By contrast if the engine has been operated in a way which has allowed moisture to accumulate in the oil and then a preheater has been used that overtemps the oil, acid will be formed and moisture will be released. If moisture is in the oil it takes an oil temperature of 100 Degrees F (38 Degrees C) before any significant amount is released. If there is the exposed metal surface then corrosion can begin because there is always oxygen present. Oils with high operational times are more susceptible to this than are new fresh oils. There are also other by products of combustion that can be corrosive and if these are mixed with the acid formed in the oil the mixture can become even more corrosive.

If the oil is saturated with enough moisture when it is heated to high temperatures, it will release moisture, which will condense on the cooler metal pans above the oil and drip back into the oil. This recirculation of moisture can go on for hours or days.

The above graph shows oil temperature, time and humidity in the air above an oil sample. This is 15W50 that had operated about 25 hours in an engine. We tested this oil in our test chamber. It began to release visible moisture as the oil temperature reached 100 Degrees F (38 Degrees C). This release continued for about three hours until the oil temperature leveled off at 180 Degrees F (82 Degrees C). We then heated the oil to 240 Degrees F (116 Degrees C) and moisture again appeared. At this temperature we were able to observe the dripping back and recirculating. Every batch of oil is different when you compare the temperatures at which the moisture becomes visible due to the characteristics of the oil and the amount of moisture it contains. I have not observed this "dripping back" below the temperature of 100 Degrees F (38 Degrees C) hence this is a good maximum temperature to limit oil heat in the design of electric preheaters.

There is a method of dealing with this corrosion which works in conjunction with a preheating system. Using this method, a blower is attached to the engine crankcase breather; this blower introduces fresh air into the engine while the oil filler cap is left off and the preheater operated. As the air enters the engine it is warmed causing it to absorb moisture. The moisture is then dumped overboard as the air leaves the engine. This device, called an engine aerator, allows continuous operation of a preheater with no danger of moisture even if the engine has been improperly operated.

HOW WE TESTED

Temperature Measurement

Engine and oil temperatures were measured by thermocouples that were read by a digital instrument. We have a data-recording device that can read up to 20 thermocouples at once and it prints these readings out on a tape. This device can be installed in an aircraft to read temperatures in flight. Some of the units used are a simple single digital display. Our units were calibrated traceable to the NIST.

Oil temperatures were read by a long thermocouple similar to a rod. This unit was put down the dipstick tube or the oil filler of the engine. Crankcase and nose case temperatures were measured by a washer thermocouple placed under a flange bolt along the top junction of the case halves.

Cylinder head temperatures were measured either at the thermocouple well hole or with a standard spark plug washer thermocouple. I chose to measure at these places because they are the industry standard temperature measurement locations. The temperature of the head casting has many implications for proper preheating. When measuring other surface temperatures on castings, cylinders or tanks a thermocouple was bonded to the surface with RTV then covered with a small insulating patch to keep air flow from affecting the reading.

Bearing Clearances

Tests were done by assembling a crankcase with the bearings installed but without a crankshaft. Both crankcase and crankshaft were placed outside to get cold soaked. Tests were done down to -20 Degrees F (-29 Degrees C). After checking the temperature of the pieces, the bearing diameter and the crankshaft main journal diameter were measured. The micrometers were kept warm and were calibrated traceable to NIST. Then the tests were also done at room temperature.

Oil Heating, Moisture and Humidity

We have a test chamber that is the same in volume as an IO-520 Continental crankcase. There are three silicone pad sump heaters that we use in order to vary and control the amount heat in the oil. The chamber has an oil filler, a crankcase vent tube and Plexiglas windows so one can see inside of the chamber. Oil temperature is measured by a thermocouple probe in the oil. Air temperature and humidity is measured by a humidity meter sensor in the air above the oil. This device is operated at various temperatures and for varying lengths of time to test oil heating. We also did some tests by placing the humidity sensor in the crankcase or in the modified rocker cover of an actual engine.

CONCLUSION

When selecting a system to preheat an engine one must not select something that will introduce other problems. Some of the heaters described above create unexpected problems if not applied with caution. Ideally a system should be matched to the engine and airframe combination. Examples of improper application follow:

Excessive heat applied to the oil sump can promote carbon formation, oxidation, acid formation and release of entrained moisture. The type of element isn't as significant as what it does. The use of an engine aerator will prevent damage from released moisture.
Mechanical interference with other systems can occur. One type of sump heater that protrudes 1" below the sump is eligible on an engine that is installed in a single engine retractable gear aircraft. On this plane the retraction linkage dears the oil sump by ½". It may be possible to retract the gear but not to extend it again.
Some may affect engine operation at other times than when preheating. One cylinder heater that installs on the cylinder base raises the temperature of the cylinder base under the heater during flight when the heater is not operating. I have measured an increase of 100 Degrees F (38 Degrees C) on an engine at this location. Will this cause a problem? Cooling tests should be redone to determine whether this is safe.

The preheating method that is selected should at least do the following to make the engine start safely and easily:

Produce a temperature at the cylinder head of at least 70 Degrees F (21 Degrees C). 100 Degrees F (38 Degrees C) or higher would be desirable.
Produce a crankcase / nosecase temperature of at least 30 Degrees F (-1 Degrees C). Higher temperatures up to 100 Degrees F (38 Degrees C) are acceptable.
Produce an oil temperature of at least 40 Degrees F (-1 Degrees C)with the object being to have at least 33,000 centistokes or less viscosity when the oil in use is considered. It should not exceed 100 Degrees F (38 Degrees C) maximum so that release of moisture and deterioration of the oil is not promoted.

By selecting a properly designed electric preheater that is properly matched to the engine and airframe, then very good results can be expected. The engine life can be extended by limiting the excessive wear produced by cold starts, it will allow convenient use of the aircraft in cold weather, and it can be energy efficient.

I have designed a heating system that uses 770 Watts of electrical power per hour. This system replaced a "car warmer" installation that required 1500 Watts, insulated engine covers and also required draining the oil sump each night. The oil was taken inside where it was warmed and poured back into the engine in the morning prior to starting. The 770 Watt system replaced all that. If this engine had been preheated with the old "air blower method" it would have required about 100,000 BTU's total and keeping the oil indoors overnight. It would have wasted an hour of time. Since this aircraft was a twin the other engine would also have to be done. It can easily be seen that pair of these 770 Watt heaters used with a pair of engine covers makes much more sense.

I have designed preheating systems from 240 Watts up to 2500 Watts. These have covered engines from A-65 Continentals up to engines like the P & W R2800. Similar designs can be applied to turbines and to helicopter gearboxes, but that's beyond the goals of this paper.

Happy Flying!

ACKNOWLEDGMENTS

The information on oil types and viscosity at low temperature is from Shell Oil Company Graphs.

CONTACT

Tanis Aircraft Services
P.O. Box 117
Municipal Airport
Glenwood, MN 56334
320-634-4772
800-443-2136 US & CANADA

REFERENCES

Aircraft Engine Maintenance by James H. Suddeth. John Wiley & Sons Publisher. Chapter on fuels fuel systems and refueling. Chapter on lubrication.

Aircraft Propulsion Powerplants by Lawrence T. Carginino and Clifford H. Karvinen Educational Publishers. Chapter on lubricants and lubrication systems.

ADDITIONAL SOURCES

Aviation Safety Oct. 1, 1984 Vol. IV, No. 19. Chilly starts may cause unreckoned engine damage by Peter G. Tanis.

Corrosion in Engines, Oil Heat - Asset or Liability, and Profile of a Pilot / Owner with a Rusty Engine. Papers by Peter G. Tanis.

DEFINITIONS, ACRONYMS, ABBREVIATIONS

BTU: British Thermal Unit

CHT: Cylinder Head Temperature

FAA: US Department of Transportation - Federal Aviation Administration

FAR: Federal Aviation Regulation (US)

RTV: A silicone rubber in a semi-liquid form that cures to a solid when moisture is present. Used as an adhesive.

NIST: National Institute of Standards and Technology

Oxidation: This term can mean many things. It can refer to digestion of food or rusting of metal. It is used here to describe chemical changes in oil caused by exposure to heat resulting in the formation of acids and sludge if moisture is present.

Sump: The lower portion of an aircraft engine that holds the lubrication oil. Similar to a "pan" on an automobile.