COOL IT, MAN! By M. B. ("Molt") Taylor, EAA 14794 AEROCAR, Inc. Box 546, Longview, Wash.

The cooling fan is mounted on a

Continental 0-200-B pusher engine.

f?/^OOL IT, MAN," says the well-versed teenager, with \^t no thought and little meaning. "Cool it," says the homebuilder, realizing that he has a major problem which may require a great deal of thought, and may mean success or failure of his creation. Engine cooling is as important as wings, for even gyro-copters without wings have cooling problems. Our work with the Aerocars and other experimental aircraft over the past 25 years has required solutions to many cooling problems. The following ideas and suggestions, taken from this experience, will assure the homebuilder of a cool engine installation, whether the arrangement is something very sophisticated or completely conventional, pusher or tractor. The ideas are presented as applied to the usual light four-cylinder engine, although they will work equally well with a single-cylinder or an eight-cylinder air-cooled power plant. The usual critical cooling condition is the low-speed climb attitude on a hot day. The FAA wants the engine to operate well below (at least 10 deg. F. below) the manufacturer's recommended maximum head temperature on a 100 deg. day at minimum speed and full throttle. This requirement took some doing for even some of the commer-

cial designs of yesteryear. In most modern engines, however, recent design refinements, some of them quite subtle, put this requirement within the designer's reach. Some of these subtleties are inadvertently disturbed by the homebuilder when he tinkers with carburetor jets, changes carburetor type, or changes cylinders from one engine to another. A very critical item in engine cooling, of which the homebuilder may not be aware, is the distribution of the fuel-air mixture from the carburetor to the cylinders. The manufacturer has tried to get the various intake tubes as equal in length as possible, so that each cylinder gets an equal charge. The zone area just above the carburetor where the fuel-air mixture is divided off to the cylinders is given very careful design attention. Often an experimental installation can be cooled satisfactorily by moving the carburetor a bit one way or the other so that the stratified fuel-air mixture hits the distribution zone a bit differently. In some engines this stratification can be controlled with straightening vanes just below the carburetor where the air comes into it from the carburetor-heat air box. If your engine had such vanes on it originally, leave them on! Equal fuel distribution to the cylinders should be checked by measuring the temperature of each cylinder under identical conditions. It is possible that with one thermocouple, the coolest cylinder is being measured, giving a false indication that all is well. At any rate, a difference in temperature from one cylinder to another may indicate poorly distributed fuel mixture.

Engines are commonly cooled by using a rich mixture. Pouring more cool gasoline into the engine helps cool it.

The cooling fan is so mounted as to suck cooling air over the cylinders on this pusher arrangement. The aircraft is a prototype of a small flying boat. 12

AUGUST 1965

Fig. 1—Typical cooling la>out. Outlet should be behind maximum cowl depth. Avoid turbulence downstream, such as is caused by cowl flare or flap (shown dotted). Note rolled lip on inlet and air box at rear of engine.

It is well known that a lean engine runs hot, and the converse is also true. The most economical mixture is not necessarily the coolest. A cool air supply to the carburetor aids in engine cooling. Air intakes out of the free airstream are to be avoided except for the carburetor-heat source. We have

seen homebuilts with unconventional lightplane engines

where the carburetor air intake was just downstream from the hot cylinders. Besides causing a power loss from running on hot air, this installation would amplify heating problems.

With proper mixture and equal fuel distribution assured, the next consideration is putting the cooling air coming into the cowl to the most efficient use. Many homebuilts suffer from hot operation due to insufficiently tight engine baffles, or leaks around the baffles where they meet the cowl. A piece of felt or other sealing material can sometimes bring the head temperature down to limits with very little effort. Getting cooling air routed to the cooling fins is not enough; distribution to the fins is also necessary, as shown in the sketch. Moving the air closely around the fins will cause it to be progressively heated until it is overly hot before it has completed its course past the fins. This is particularly true of some of the Lycoming engines. It is desirable that the cold air be routed well around the cylinder, particularly in the exhaust valve area, before it is introduced to the fins. The box idea shown in the sketch is common practice, and the writer has seen 30 deg. temperature drops on the hot cylinder by merely adding such a box to the exhaust valve fin area. It is particularly important that the engine be cooled with air moving in the direction intended by the designer. An engine designed for downdraft cooling is often difficult, if not impossible to cool with an updraft arrrangement, due to shape, depth and placement of the fins. In addition, the shroud tubes for the valv.' push roas can affect cooling, giving destructive hot spots which the soark plug thermocouple will not detect.

A word of caution about spark plug thermocouples: they should be regarded as a check of relative heat and not as an exact source of measurement. They are not precise enough to be depended upon if the engine temperature is close to the limit. Gasket thermocouples being reused should be annealed before installation, and the plug torqued properly. Failure to do either may cause COMi t» BffFLE

-

Fig. 2 Air box, baffles a n d thermocouple location.

improper seating which can result in temperature errors as high as 50 or 60 deg. F. Much preferred are bayonet type thermocouples or, as second choice, termination thermocouples with drive screws to hold them in place near the downwind plug. Taking the temperature at an upwind location is meaningless. The remaining problem is to assure an adequate sup-

ply of cooling air. The cooling air inlet opening is very important; it must be out of a turbulent area and must be presented as squarely as possible to the free airstream. Screens or louvers at the inlet should be avoided. A simple decorative grill can cause cooling problems. Removing the screen or grill should be a first step in investigating a hot engine problem. Very important also is the shape of the inlet opening where it is presented to the airstream. A rolled or round edge can increase the flow by as much as 40 percent over that of a sharp or squareedged opening. Those sharp edges on sophisticated installations such as jets have boundary-layer control at the duct lip! A one inch radius roll is adequate.

On installations using a duct to the engine, care must be taken to avoid any protuberance in the duct wall, or any objects, such as cables, running through the duct. These will cause turbulence which may lead to stratification of the airstream and lower the volume delivered. Stratification can be checked by measuring the velocity across the duct, using an airspeed indicator with the pitot tube on a movable probe. Measurements must be made under identical conditions of airspeed, altitude and attitude, and with the probe square with the airstream. A duct installation should have a plenum, or larged area for the duct to empty into ahead of the gine, allowing the air to expand and circulate over entire engine area. Rolled inlet edges are particularly portant with a duct system.

Many air-cooled engines are actually oil-cooled to a great extent, and efficient oil cooling may be an answer to a hot engine problem. Oil radiators should have a good supply of cold air, with rolled air inlets and smooth ducting. Bleeding air from the engine cooling air supply will not seriously impair the efficiency of most installations. In no case should hot air downwind of the engine be used. An oil radiator should be of the full flow type, with oil flowing through it before being introduced to the engine under pressure. Some engines have a thermostatic bypass arrangement so that oil is diverted through the radiator whenever the temperature is high enough to require cooling. A safety valve is provided to allow oil flow to the engine, should the radiator become plugged. This must be installed properly when the engine is overhauled or modified. Unless the radiator is an integral part of the engine, oil hoses must be used to allow for relative motion between the engine and radiator. Hoses must be of ade(Continued on next page)

* TMfGMCCOUPUS

Fig. 3—Placement of fan in shroud. Shroud is a must. Clearance between tip of blade and shroud should be as small as engine movement will allow.

enenthe im-

Fig. 4—Fan installation aft of engine. Note rolled inlet, plenum chambers fore and aft, and baffles. Fan is shrouded.

Fig. 5—Buried engine arrangement with duct and fan ahead of engine. SPORT AVIATION

13

COOL IT, MAN! . . . (Continued frcm preceding page)

quate size with fitting openings at least the diameter of the inside of the hose. Some auto-quality brass fittings have smaller openings than their aircraft counterparts. These should be drilled to size. Hoses must be in good condition, without loose liners, in any aircraft installation. We would be remiss if we did not discuss fan cooling for pusher and buried installations. In fact, cooling fans would be of benefit to most lightplane installations, including tractor arrangements. The common wood propeller found on so many homebuilts is very inefficient at the root area, giving very little acceleration to the oncoming airstream. A spinner will often help cooling by de creasing turbulence at the propeller hub and root area, allowing better flow into the air inlets. Many cooling fan ideas have baen explored, and in our work with the Aerocars we reviewed thoroughly the arrangements used by helicopters and on pushers such as the "Seabee." Since it is almost imperative that a flying automobile have a buried engine, the fan problem had to be resolved in the most efficient way. Power requirements, automobile cooling fan designs, axial vs. centrifugal blowers, shrouds, etc., were all investigated. While a book could be written on this subject alone, we concluded that a multi-blade axial fan such as that shown in the illustrations proved to be the most efficient, lowest cost, and most practical for the Aerocar or most homebuilt lightplanes. Designing the blades was a major project, with blade stiffness, twist, tip velocity, shroud proximity and fan solidarity taken into consideration. Cost, serviceability, balance, flutter, vibration and blade replacement were other deciding factors. The result was a fan which will move an adequate amount of air over any well-cowled engine and should allow full-throttle operation without the aircraft moving, for an indefinite time. The blades are injection-molded Delrin plastic, four inches long, with a

5/16 in. butt thickness. Twelve of them mounted between two Tl/2 in. diameter discs of .080 in. 6061 or 2024 aluminum will make a fan which will cool 100 hp at engine speed, or 180 hp if belt driven at twice engine speed. Blades are installed with two AN3-7 bolts, and can be interchanged without balance, track or vibration problems. Blades were recently supplied to several homebuilders for use on "Sportsman"-type amphibians. One application is using a 20 in. diameter, 24 blade fan to cool a Lycoming 0-320. The blades are now available to other homebuilders who need an engine cooling fan. The fan should be mounted in a shroud, as shown in the illustrations. It makes little difference whether the fan is ahead of the engine, blowing on it, or behind, sucking out the hot air. The purpose of the fan is to accelerate the airstream over the engine, in one way or another. The blades are set at a high angle of attack because they are turning in a moving airstream and are intended to accelerate the air further. They are not designed for thrust, but for acceleration of the air with minimum power use, and these considerations account for the blade section and twist. Disc loading is an important consideration in a fan installation. The upwind side of the fan should be loaded as symmetrically as possible. If the blades meet stratified air at different velocities or at various angles to the plane of the fan, there will be fore and aft loadings which may be destructive. The highly damped plastic material is superior to metal in fatigue resistance. It is finding wide use in small foreign automobiles as a replacement to metal. Replacement fans of plastic for U.S. automobiles are available, but these do not deliver the volume of air required for a lightplane engine. Wide blade spacing allows leakage or recirculation of air between the blades. Cooling of homebuilt engines is complicated, but not too difficult if one knows what to look for. The writer welcomes inquiries from any EAA member who has a particular problem. Be sure to send along a stamped, self-addressed envelope. The last time we invited inquiries, the postage to answer them became a real consideration. •'•

Third Annual New England EAA Banquet Over 150 aviation enthusiasts attended the Third Annual New England Sport Aviation banquet which climaxed a full day of activities which saw a regional EAA chapter meeting at the Holiday Inn in Lawrence, Mass. EAA President Paul Poberezny was the featured speaker. Other notables in attendance included Rev. John MacGillivray of the RCAF, and Re gional Representative Robert Ring and his wife, Hilda. Merrimack Valley (Lawrence, Mass.) Chapter 136 :: was host. ' (John Doyle Photo) 14

AUGUST 1965