I spent a bit of time adjusting some of the parameters of my thermal control ECAT model and have a few observations.
I suppose that it seems obvious that the greater the non linearity of the function binding core power generation to core temperature, the more critical will be the required control needed for stable high COP operation. I toyed with forth, third, second and the really wonderful linear order functions to establish the trend. An example of the type of function would be P=k*T^4. I prefer to use polynomial functions for this analysis since it is possible to construct most other functions from a series of polynomials and it is very easy to work with them. My model can use any function that I choose for this purpose with the exception of delay mechanisms. I suppose that they could also be incorporated if there is a good reason to believe that they are present and important. The linear function has the characteristic of allowing the ECAT to have a constant gain over any input temperature range somewhat like what Rossi used to say. Recently, that has been put aside as it should have been long ago to be more realistic. Any higher order relationship exhibits what I refer to as the two temperature effect. The first temperature of the pair is determined where positive feedback with a gain greater than unity takes over and the device begins its journey toward melt down. This path is followed toward an ever higher core temperature until one of two things happen. If the drive power is removed soon enough, the core will cease to increase in temperature and begin to cool down. This constitutes the normal operation function that I suspect Rossi uses. The trick is to remove the drive power at the correct time before the second temperature of the pair is reached. Unfortunately, the closer to that point you allow the device to reach, the higher the overall COP. And it appears that this behavior is fairly critical if reasonable COP is to be achieved. If you are too slow in removing the drive, then meltdown will result. Unless Rossi includes some method of rapidly extracting heat from the core region, the device can not be brought back to safe operation. I leave open the possibility of a design that is extremely well adjusted such that the functional relationship between core power generation and temperature drops rapidly before core destruction occurs. This seems unlikely, but the NAE might have some characteristic that ceases their activity while not being melted, or perhaps they regenerate after some delay and cooling. This is totally speculative. Some of the temperature waveforms that I observed as I adjusted the model parameters bear a remarkable resemblance to those observed during the recent tests. Particular attention should be applied to the exponentially rising waveform observed during the active drive pulse and the drawn out delay seen with the falling temperature. These waveforms are the signature of a system that exhibits gain as a result of the positive feedback effect I have mentioned. They become more pronounced as the gain is increased. Dave
