Searching over 5,500,000 cases.

Buy This Entire Record For $7.95

Official citation and/or docket number and footnotes (if any) for this case available with purchase.

Learn more about what you receive with purchase of this case.

Tropic-Aire Inc. v. Auto Radiator Mfg. Co.


April 1, 1938


Appeal from the District Court of the United States for the Northern District of Illinois, Eastern Division; Philip L. Sullivan, Judge.

Author: Sparks

Before SPARKS and MAJOR, Circuit Judges, and LINDLEY, District Judge.

SPARKS, Circuit Judge.

Appellant charged appellee with infringement of its United States patents to Waters, No. 1,684,900, and to Butterfield, No. 1,834,141. The former was issued September 18, 1928, on an application filed November 20, 1926; the latter was issued December 1, 1931, on an application filed February 20, 1929. The defense was invalidity and non-infringement. At the conclusion of the evidence the court dismissed the bill of complaint for want of equity. Claim 5 of the first patent and claims 1 and 5 of the latter patent were involved. The court concluded that each claim, if valid, was so limited in scope, in view of the prior art, as not to be infringed by appellee; and conversely, if the claims were construed broadly enough to be infringed by appellee, then each claim invalid in view of the prior art.

The Waters patent covers a hot water heating system for automobiles, in which a small percentage of the water from the engine cooling system is passed through the car heater in such a way as to efficiently heat the car, yet not interfere with the cooling system of the engine. Efficient heating is said to be provided by the interior design of the heater, by means of the volume and velocity of the water through the heater, and also by its exterior general design.The lack of interference with the cooling system is said to be obtained by restricting the percentage of the flow from the cooling system through the car heater, and not through the cooling radiator. That restriction is accomplished by designing the inlet and outlet connections of small size, and also by the method of connecting them to the cooling system. Waters stated that there must not be a diversion of water from the cooling system greater than 20 per cent; that that amount or less will effectively heat the car if the water be made to circulate rapidly and in the form of a thin sheet.For this purpose he designed his heater with large radiating surfaces in proportion to the volume of water space, which he says will function properly if its volumetric capacity is as low as one pint of water, and where the connections are of such size that the conduits for an automobile with two such heaters have an approximate capacity of one quart. He specifies the size and proportion of other elements of the system with great particularity, for instance, that the couplings by which the heater hose are connected to the cooling system should be of three-eighths inch pipe, or less. He suggests the use of a thermostatically operated valve for the purpose of confining the circulation of the cooling water to the engine jacket and the heater until the water has reached a predetermined temperature, whereupon the valve will open to permit the water to flow to the cooling system. However, it is then necessary that the diversion from the cooling system of the hot water for heating purposes shall not be so great as to impair its function of cooling the engine cylinders, and herein lies the point of controversy between the parties, appellant claiming that its patented system, when the valve is open in any degree, will at all times prevent more than 20 per cent of the water from being diverted from the cooling system through the heater.

Waters, in his drawings and specifications, discloses a number of slightly different arrangements of the elements of his claim, but for the purposes of this case he relies on Fig. 13.


The engine 1 is provided with the usual water jacket which communicates through the usual radiator hose 2 with the usual cooling radiator 3. The water passing through 3, goes by way of pipe 4 to a pump 5, where it is re-introduced into the engine jacket. In the radiator hose connection 2 there is provided a fitting 66, having an outlet 67 which is connected with hose 22, leading to the heat radiator 25. After passing through 25 the water is returned to the cooling system through hose 26, which is connected to pipe 4 on the suction side of the pump 5. A thermostatically operated valve 44" is located between the engine jacket and the diverter coupling 66. In its argument and brief appellant suggests that this valve may be located between 66 and the cooling radiator 3, as exhibited in Figs. 10 and 11. However, these figures disclose other forms of construction of appellant's system, and it is not stated in the specifications that the location of the thermostatic valve, as shown in Fig. 13, may be changed except in the constructions shown in Figs. 10 and 11.

In Fig. 13, the control valve 44" is constructed so that it can not completely shut off the flow through conduit 2. This is provided for in the usual type of thermostatic valve used in automobile circulating systems by drilling a three-sixteenths inch hole through the valve disc, or by the use of a valve that will not completely close by the area of a three-sixteenths inch hole. Claim 5, the only one in issue, is in the margin.*fn1

The condition of the art was disclosed by prior United States patents 1,124,403 to Faruolo, and 1,581,761 to Muir; and French patents 384,583 to Bayractarow, and 587,872 to Nanteuil. It is admitted that each of these patents discloses a hot water heating system connected to the cooling system of an automobile.

Faruolo discloses a water cooled engine which sends hot water through a conduit, consisting of a metal pipe section and a flexible or hose section to the top portion of the cooling radiator. The water returns to the engine by way of a pipe connecting near the base of the cooling radiator, a pump adjacent thereto, and a return pipe or manifold. The heater supply tube or pipe extends downward from the hose section bending laterally and extending underneath the floor of the car to the heater positioned in front of the rear seat and practically flush with the floor. A valve is placed in the vertical portion of the heater supply pipe, which is manually controlled, and it is worthy of note that the claim in suit is not restricted to a thermostatic control. A restriction is used in the hose section which deflects some of the water through the valve when open, to the heater. When the valve is wide open, substantial amounts of water may be forced to the heater by the restriction in the hose section; if, however, the valve is closed, then no water flows to the heater. The water deflected to the heater retunrs through a pipe extending from the heater to a pipe where it mingles with the water returning from the cooling radiator. The circuit through the heater is said to be "in parallel" with the cooling circuit passing through the cooling radiator.

The Muir patent discloses a hot water heating system connected to a cooling system. It is a series system in which all the water which goes through the heater returns to the automobile by way of the main cooling radiator. The heater has a plurality of water tubes, as does appellee's. Muir convects the air around the hot water tubes of the heater by means of a fan adjacent the cooling radiator, which blows the air through a pipe to the heater. Appellee likewise uses a fan, except that its fan is in the interior of the car, and instead of the water passing through the heater in a wide sheet or stream, it passes through it in a plurality of parallel channels which mutually interfere with one another, so far as free radiation of heat in all directions is concerned. In other words, in Muir, as in appellee's system, heat is delivered by convection of air currents around the tubes, rather than by free radiation in all directions.

Bayractarow discloses an engine with a water outlet through a pipe to the cooling radiator, and a return to the engine through a pipe. At a valve in the outlet pipe to the cooling radiator is a passage leading to the heater, and there is a return passage from the heater to the cooling radiator. The cock of the valve is provided with two buttons, or with a double manipulating lever whose two ends are connected by pull cords or rods to another double lever fixed to a handle disposed within reach of the driver. When the valve is set for cooling conditions, and the heater is not in use, the circulation is directly through the outlet pipe and valve from the engine to the cooling radiator. When heat is desired the valve is turned so as to direct the hot water through the passage leading to the heater, and in turn through the return passage to the cooling radiator. The valve can be set to any position between its extreme off and on positions by manipulation of the handle near the driver.

Nanteuil discloses an engine jacket with a water connection, or outlet, leading directly to the upper part of the cooling radiator, with a return pipe from the lower part of the cooling radiator to the engine. A valve is shown in the outlet pipe leading from the engine. Between this valve and the engine is connected a pipe which leads hot water from the engine outlet pipe to the heater. There is a return pipe which returns the water from the heater to the return pipe of the cooling radiator, thus completing the heating circuit. No pump is shown as the disclosure refers to a thermalsyphon system. It is manually operated, and may be regulated and operated from the seat of the vehicle.

Nanteuil refers to his valve as an obstructing element, and in his specification he states:

"When the obstructing element completely closes the principal circuit, the water circulation takes place entirely through the heat exchanger of the vehicle. By opening more or less the said obstructing element, the circulation of water in this exchanger can readily be regulated. When the principal circuit is completely open, the water circulation in the circuit of the heat exchanger becomes practically zero, the resistance presented by this circuit branching from the main flow of the water being greater than that presented by the principal circuit.

"Experiments have shown that after having completely closed the principal circuit, the heating of the heat exchanger (for example a kettle) takes place very rapidly. From this instant on, the obstruction element may be partly opened. The temperature of the heat exchanger remains high while the cooling of the cylinders is effected under ordinary conditions. It should also be noted that the device is used solely during cold periods. During these periods the cooling effect due to the radiator is frequently too great, to such an extent that the drivers are compelled to cover up a portion of the surface of radiation. This inconvenience is completely avoided by the use of the device according to the invention."

It is practically admitted that 160 degrees is about the water temperature which a thermostat of the character here involved is to maintain. It begins to open at 158 or 159 degrees and is not fully opened until it reaches about 175 or 180 degrees. The purpose of a thermostat is to maintain, as far as possible, a suitable temperature in the engine jacket. This is predetermined by the setting of the thermostat. It regulates because at 160 degrees it slightly opens and permits a small amount of water to pass into the cooling radiator. On a winter day this water cools rapidly and returns to the water in the engine jacket to hold the desired temperature. When the weather is very cold the thermostat may remain closed at all times. The normal operation of the cooling system in winter is not to flow as much water into the cooling radiator as in summer. To do so would result in too low an engine temperature. For this reason the thermostatic valve must be shut more closely in winter than in summer. Hence, a maximum amount of water that can go to the cooling radiator occurs when the valve is wide open, as in the summertime, or when the temperature in the engine jacket is at about 175 or 180 degrees, or in excess of that. During winter opertion the only way in which a thermostat can be made to open wide is to exceed the desired engine jacket temperature of 160 degrees.

It is conceded that appellee's thermostatically operated valve is positioned between nipple 67 and the cooling radiator 3, and not between the engine jacket and the nipple or fitting 66 as shown in Waters' Fig. 13. Appellee recommends that its heater be used with the valve opening at 160 degrees, and at a permissible range of from 150 to 165 degrees. It is thus apparent that when used under normal operating conditions the valve is closed or only partially open, so as to bring the temperature up into the range of 150 to 165 degrees.This range is never exceeded. Hence, it is obvious that when used according to directions, and it is not claimed that they are used otherwise, appellee's heaters are used in a system with a thermostat operating with a slightly open valve to maintain a 160 degree water temperature, and never with the valve wide open. The heaters consist of two water chambers connected by forty-eight vertical tubes arranged at an angle of about 10 degrees and in four rows. Water is brought to and from these by a supply pipe and a return pipe. This organization is placed in a housing, which is to direct air through the core. These heaters, strictly speaking, are not of the radiator type, but of the convective type wherein heat is distributed by a current of air flow induced by a fan. When heat is desired the fan is started; when it is not wanted the fan is stopped. These heaters do not have large radiating surfaces and they give off very little heat by radiation as compared to that given off by convection. They are what is known quite generally as unit heaters, wherein air passed through the casing is warmed by convection.

The principal question here involved, relative to infringement, is whether when appellee's heating systems are installed as instructed by it, the result is that not more than 20 per cent of the water circulating in the cooling system is diverted to the heating system. If the valve is placed as shown in Fig. 13 of the patent, it is obvious that no water will pass through either the heater or cooling radiator until the valve begins to open. If, however, appellant's valve is positioned between the diverter coupling 66 and the cooling radiator, as suggested in argument, then, as with appellee's device, the amount of hot water diverted to the heater will depend upon the open or closed position of the valve. If the valve is wide open, a relatively small part of the water is deflected to the heater circuit and a larger part to the cooling radiator. If, on the other hand, the valve is nearly closed, a much smaller part of the water will pass through the cooling radiator, and a relatively large part will be deflected to the heater.

Each party constructed and tested the other's device, as well as its own, and, singularly enough, the respective results seemed to be quite satisfactory to each party making the tests. In other words, appellant says that its tests disclosed that appellee's device, as well as its own, did not permit more than 20 per cent of the water circulating in the system to enter the heat radiator; while, on the other hand, appellee says that its tests disclosed that each device permitted much more than 20 per cent of such water to enter the heater.

In testing appellee's device appellant employed both a calibration and a road test. In making the calibration test, the operator used a vacuum bellows thermostat and punctured the bellows with a knife or other sharp instrument. The punctured bellows caused the valve to open permanently, and it would not close thereafter, either partially or entirely, in response to any change in temperature. The tests, therefore, were made with the valve wide open, when the least amount of water was passing to the heater.The results disclosed a slight variation of percentages of water flowing through the heating system, and at no time did it exceed slightly more than seven per cent of the total flow circulating through the cooling system.

Likewise, the road tests were made at a temperature slightly above 180 degrees, which means that the valve was wide open, and the least amount of water was passing to the heater. Such temperature was never recommended by appellee in the use of its device, for it specifically limited its use to temperatures ranging from 150 to 165 degrees. It is conceded that when the thermostat is set at about 160 degrees for the beginning of the opening of the valve, which both parties recommended, the valve will be entirely open at 180 degrees. It is obvious, therefore, that above 180 degrees the thermostat is not functioning any more than when the temperature is below the point set for the beginning of the opening of the valve, which is not at all. Moreover, when the temperature is 180 degrees, the alcohol used as an anti-freeze begins to boil from the solution, and neither party contemplated that the temperature should be permitted to reach that point.

Appellee also made tests of its device to determine the percentage of diversion of water from the cooling to the heating system. These were laboratory tests with the automobile stationary, its rear wheels jacked up, and the front wheels blocked. The heating system was connected in accordance with appellee's instructions. A pressure gauge was connected to the discharge hose from the heater, and the pressures indicated thereon were noted. In the second test the discharge hose from the heater was discharged into a tank. A replacement tank was located above the heater and provided with a hose connection to the suction side of the pump. The water which would normally go to the cooling radiator was discharged into a weighing tank, and the cooling radiator was supplied with additional hot water from a separate source. In this actual flow test the discharge hose from the heater was manipulated until the same pressure readings were indicated on the pressure gauge as those which had been noted in appellee's first test. In both of these tests the temperature recommended by appellee was maintained in the engine jacket, and the valve was not wide open. The temperature was maintained at values prevailing in actual winter operation of a car, in order to demonstrate the effect on the flow ratio. At the temperature of 160 degrees, there was a 96.9 per cent ratio of flow of water through the heater to the water circulating in the cooling radiator. This was obtained by making the number of pounds of water circulating in the cooling radiator or system, which was 65, the denominator of a fraction, and the number of pounds of water circulating through the heater which was 63, the numerator. When the temperature was 163 degrees the ratio was 81.4 per cent. A test was also made with the temperature of the cooling water at 134 degrees, when the thermostat was closed and the only water flowing to the cooling radiator was a slight trickle through the small hole drilled through the valve disc for that purpose. In that case the denominator was 12 and the numerator was 59.5, and the ratio was 496 per cent. In other words, there was almost five times as much water going through the heater as there was passing to the cooling radiator.

Appellant's objection to these tests is that because of the short period of testing the temperature was not allowed to reach the degree at which the thermostatic valve would have opened. It is quite true that the valve was never wide open during appellee's tests, and it never would be if appellee's recommendations were followed.Its maximum limitation of temperature is 165 degrees, and that will not completely open the valve when the thermostat is set for beginning the opening at about 160 degrees. In other words, appellee's system does not, and is not intended to operate when the temperature is so high as to cause the valve to open entirely.

Appellant further contends that appellee's ratio is incorrectly written, in this, that the numerator should represent the quantity of water flowing through the heater, and the denominator should represent the amount of water flowing through the heater plus that flowing through the colling radiator. However, if this method be adopted, the deflection of hot water would be much over 20 per cent, if appellee's limitations of temperature are observed. We think appellant's tests are not fair to appellee's system, and that when it is used according to appellee's instructions, it follows rather closely the prior art hereinbefore referred to and does not infringe the Waters patent. It is quite possible that appellee's tests did not disclose results which were absolutely perfect, but to us they seem to be quite logical and approximately correct. However that may be, the burden was upon appellant to establish infringement. For reasons hereinbefore stated we are unwilling to accept its tests as proof of that fact, and though it be conceded that appellee's tests were not proper to prove non-infringement, the fact remains that infringement has not been established, and we so hold.

The Butterfield patent, No. 1,834,141, is for a fitting for automotive vehicle heaters. Fig. 7 is relied upon:


The fitting comprises a threaded nipplereceiving element 28 having an integrally formed part-tube 26'. Into this element is threaded a nipple 22. Upon the outside of the hose 17 there is a shaped washer 31 which conforms to the contour of the outside of the hose 17. There is a clamping or locking nut 32 which is turned down upon the flat surface of the shaped washer 31 to clamp the assembly tightly against the inner and outer walls of the hose. Claims 1 and 5 are in issue.*fn2

Here follows a reproduction of appellee's fitting with the same reference numbers as appear in Fig. 7.


Appellant contends that there is an element-for-element correspondence between appellee's fitting and that of Butterfield. It is quite obvious, however, that appellee's device does not contain the strengthening element 26' as disclosed in Fig. 7. The nut 28 of appellee's construction does not engage any considerable area as does 26' of the patent. Appellant admits that the prior art as to this patent is quite close, and it is well illustrated by United States patent to Christie, No. 1,748,596. However, appellant insists that Christie, as well as all other prior art cited, is in wholly remote arts. It suggests that Christie was not concerned with the problem of connecting an automobile heater in fluidtight relation to the cooling system of an automobile, but rather with attaching a hydrometer casing to the cooling system, and he deemed it necessary to include an additional lock washer with locking prongs adapted to enter recesses formed in the threaded portion of the nipple. This may be admitted. Nevertheless, we are convinced that Christie was not working in a remote art, but on the contrary, his field of operations was quite closely related and allied to that of the patent in suit. Everything in the accused device is found in Christie, or at least could be found with the exercise of a reasonable amount of mechanical skill, and if claim 1 is considered sufficiently broad to read upon appellee's connection, it unquestionably reads upon Christie. On the other hand, if Butterfield fails to read upon Christie because the nipple-receiving element inside the hose is not "a strengthening element, engaging a considerable area," then it follows that appellee's nipple-receiving element 28 is not an equivalent to appellant's element 26'.

Claim 5 does not expressly include the limitation that the nipple-receiving element is "strengthening" or that it engages a "considerable area." In order to save the claim from reading upon Christie there must be read into it these characteristics which are found in claim 1, and when so read, claim 5 fails to read upon either Christie or appellee's device. We accordingly find that the accused device does not infringe Butterfield. The same result was reached in the unreported case of Tropic-Aire, Incorporated, v. E. A. Laboratories, Inc., decided by the District Court for the Eastern District of New York, and affirmed without opinion, 2 Cir., 93 F.2d 1018.

Decree affirmed.

Buy This Entire Record For $7.95

Official citation and/or docket number and footnotes (if any) for this case available with purchase.

Learn more about what you receive with purchase of this case.