This subject is treated in a paper in two parts. Part I, by Alex Taub, deals with laboratory tests to prove by comparative data that the higher average operating-temperatures maintained in the engine by the constant-temperature, or evaporation, system of cooling have negligible detrimental effects. Part II, by L. P. Saunders, gives the results of road-tests of cars operated under the same conditions when fitted with a standard water-cooling radiator-core and with a constant-temperature cross-flow condenser-core.
Although contamination of the crankcase oil by heavy ends of the fuel is not prevented by the higher temperature of constant-temperature operation, it is asserted that this higher temperature is effective in striking an acceptable balance in such contamination and results of the tests show that the cylinder-walls are maintained at temperatures sufficiently above the vaporization point of water to reduce the condensation of water vapor to the minimum. Water in the crankcase is the objectionable element. Oil dilution by fuel up to a certain amount is not detrimental; in fact, experience shows that about 16 per cent of such dilution is necessary to facilitate starting a cold engine. Even when an anti-freeze solution containing 50 per cent of alcohol is used and the boiling temperature reduced to 184 deg. fahr., the cylinder-wall temperatures are maintained at 212 deg. or more.
Since a boiling liquid does not change its temperature, it affords the simplest means of maintaining a constant operating-temperature and also the simplest, least expensive and lightest means of providing for quick warming-up of the engine and slow cooling-down, because there is no circulation of water except when steam is passing from the engine-block to the radiator or condenser.
Test runs were made in the laboratory with an engine fitted with a Muir constant-temperature system which could be converted to water-cooling by blocking-off the circulation through the cylinder-head with a special gasket to provide for concentrated circulation around the exhaust-valves. Outlet-water temperatures were controlled by admitting more or less cold water. Results of the tests indicate that fuel consumption is approximately the same for constant-temperature cooling at 212-deg. outlet temperature and water-cooling at 170-deg. outlet temperature; that with both systems the spark-lever advance for maximum torque is safely below the degrees of advance at which detonation, or spark knock, begins; that the falling-off in torque with reduction in richness of the fuel mixture is virtually parallel for the two systems; that the difference in volumetric efficiency of the engine when operated on the two systems amounts to only 2 per cent, which is within the allowable error of the air-meter used; that the brake engine-pull is nearly identical; that the temperature of the lubricating oil is not affected by the system of cooling but by the temperature of the cooling medium, and that the temperature of the walls and inlet and exhaust-valve seats of the No. 1 and No. 6 cylinders of a six-cylinder engine is much more uniform with constant-temperature cooling than with water-cooling.
Cylinder-head formation and spark-plug location are important factors as regards detonation. A compact head with spark-plug carefully located to allow the maximum spark-advance before detonation starts provides sufficient leeway for the use of higher operating-temperatures.
With constant-temperature cooling it is advisable that the normal water-level be such that, in operation, the water flowing from the radiator to the engine-block will fill the pipe only about half full and allow air to escape above it, thereby eliminating the possibility of an air-trap in the water-pump.
A steam-dome capacity equal to 21 per cent of the normal quantity of water in the engine-block gives the proper proportion of water and steam passing to the radiator. The smallest possible quantity of water is the proper quantity to use, as the quantity of water in the block controls the warming-up period. The normal water-level is raised between 12 and 15 per cent by expansion and volcanic action of the water when the engine is running, and the capacity of the steam dome is thereby reduced 6 or 9 per cent. If the steam dome is too small, excess water will pass to the radiator.
That the high operating-temperatures that develop with constant-temperature cooling are safe is indicated by the much higher operating-temperatures in air-cooled engines.
In Part II, after pointing out the general recognition of over-cooling by the water cooling-system in winter, as made evident by the use of air shields on the radiators, and describing the operation of the Muir cross-flow condenser-core, L. P. Saunders gives the results of many road-tests of cars with water-cooling and constant-temperature cooling. It is shown that the cross-flow core, when used as a water-cooler, maintains a lower temperature of the outlet water from the radiator than the conventional core and still lower temperatures when used as a constant-temperature system.
Miles per gallon of fuel consumed are increased by constant-temperature cooling with the cross-flow core as compared with water-cooling with the standard core. Acceleration tests showed a slight advantage for the former system, while deceleration times were slightly longer than with water-cooling. The constant-temperature system showed higher speeds in hill-climbing.
Better ventilation of the engine hood may be necessary with the constant-temperature system to avoid uncomfortable heat in the driving portion of the car body. Size of the radiator-core cannot be decreased, as many cars are now inadequately cooled under certain extreme conditions of driving and air temperature and density. The fan size should remain as large, at least, as at present.
The paper is concluded with a chart showing the effects of temperatures in the water-jacket and in the lower part of the radiator caused by starting and stopping of the car in cold weather.