I decided to review my ECAT simulation model to see if it were reasonable to achieve a COP of around 3.5 while operating within a non thermal runaway region under steady state conditions. The earlier runs and model tended to indicate that it is quite precarious to operate the ECAT at a COP of greater than 2 without the pulse wide modulation input power waveform.
Once a decision is made to operate within a potentially unstable region, it becomes necessary to turn the input power on and off periodically to prevent thermal run away. To the best of my knowledge, Rossi has used this type of operation until the latest test. In that demonstration the input drive is relatively constant and operation in the so called SSM mode not used. The new HotCat expels the internal heat through a combination of radiated, convected and conducted paths. The radiation path is quite useful when one attempts to prevent thermal run away conditions since a small increase in surface temperature results in a large increase in thermal radiation. Everyone by now has seen that the radiation goes up proportional to the forth power of the temperature and that puts the brakes upon increases in extra power generation due to internal temperature increases. My main question was related to understanding how he now can operate without having to worry so much about overheating and thermal run away while that was such a problem before. The trick apparently is in the geometry of the device. A large surface area is available to radiate away the escaping heat at a manageable surface temperature. Also, the surface of the main cylinder is specially treated with grooves to enhance thermal escape due to convection. This carefully constructed design is capable of removing enough heat to quench the positive feedback action that the internal core would normally encounter at the elevated operating temperatures. My model needed to take into account the new geometry features that were not present in the earlier devices. When I first ran a simulation of the new device I noticed that it was easy to limit the maximum temperature since the radiation was so efficient at handling the extra internal heat energy generated by any moderate increase in core temperature. I model the core heat generation by means of a polynomial power series and as long as the main terms contributing to the core heating are below forth order, a stable operating point is obtained. It would be useful to have the actual power series from an operating device, but that is apparently too much to expect at this time. A problem appeared when the input power was removed. As expected the temperature dropped a large amount in the core, but it reached a point of stable continuous output. This situation would not be tolerable and fortunately not seen within the test. I scratched my head and then realized that a cure to the problem was available. I adjusted the coefficient of the linear term that represented the convection heat emission and found that a value could be chosen that allowed the output temperature to continue downwards when input drive is removed. This adjustment very much falls into line with the real device since a lot of effort was expended in designing the groove structure. When the dust settled I had an opportunity to figure out exactly what was required to achieve a stable system. The surface area of the device must be designed so that convection currents carry away more heat than is generated within the lower temperature regions. This is needed to ensure that a low temperature latching performance is not obtained. Also, the surface areas must be able to radiate the correct amount of heat at the desired operation point. In that case, the sum of the drive power and the internally generated core power has to match the power that is emitted due to radiation, convection and conduction. This new model is amazingly simple in structure but demonstrates interesting insight into operation of the new CAT. Operation with a COP of approximately 3.5 did not seem to be too difficult with the optimum parameters according to the latest model. I plan to continue to evaluate my model as time permits and new data and questions arise. Dave
