Suppose we have a special motor with a rotor made of nickel. The shape of the rotor is such that it has a high surface area for exchanging heat with an applied flame or with a stream of cooling water. The motor will have a single stator in the form of a permanent magnet. The flame is applied to the section of rotor that is clossest to the permanent magnet to cause it to de-magnetise.
We turn on the flame and cooling system. The flame and cooling system is designed to maintain the minimum temperature of the rotor to a few degrees bellow the Curie point and the flame is set to cause the exposed section of rotor to exceed the currie temperature of 354 degree celsius.
We give the rotor a small "kick", and the rotor begins to rotate in the selected direction. The rotor continues to rotate until the flame is removed.
The rotor can only rotate at a very low rpm, since it must heat and cool and temperature is a slow process, but, the torque being produced is dependent on the strength, size and proximity of the stator magnet.
We run our motor and find that we used 1000 watts of fire and cooling to produce 1 watt of mechanical energy. So, we go and get a magnet that's twice as strong/large.
Since the rotor requires the same amount of energy to heat and cool, the applied energy is the same, but now we find we have twice as much torque. Now we used 1000 watts of fire and cooling to make 2 watts of mechanical energy.
We get a bright idea: we call up a magnet enthusiast friend and get a dangerously large neodynium magnet, it's got so much pull it's at the limit of the shaft and bearing tolerances to withstand the pull force.
We apply the flame and find that we used 1000 watts of flame and cooling but we make 2000 watts of mechanical energy since the rotor has so much torque it more than makes up for the slow RPM.
Since the stator magnet determines the maximum torque and that it is a static component and the input energy required to heat and cool the stator is unrelated to the strength of the stator magnet, the torque can always be increased.
Since horsepower is "P = Torque * RPM / 5252" and we can modulate the torque without changing the input requirements with very little theoretical upper limit, it doesn't matter that a Curie temperature motor turns slow, it can be made with enough torque to exceed a COP of 1.
This is because the work being done on the rotor with the flame is not directly related to the work being done on the output shaft: It's tangential, the work being done on output shaft is consequential and unrelated.
Thoughts?
