I think we are reaching a point where we will have to simply agree to disagree again.
As mentioned before I will post the incubator design here so people can critique but please be kind to me and do your homework first!
OK. Sounds good!
I would like to point out that we are both heading in the same direction. In particular, considerations of thermal runaway and hysteresis (hysteresis means that after runaway you can remove all input and the high temperature reaction is self sustaining). I am also in favour of looking for inflection points in the temperature time series because these are indications of incipient runaway. Early on I began to wonder why these had not been a feature of your results and mentioned it to you. You did not see the relevance at that point. I hold that they are relevant now and were relevant then too.
So I began to construct a model. The simplest model possible consonant with the data you had released. And that is what you have seen so far -- a model based on the data you have released. Having done this, I realized last Spring that further progress (the relative importance of radiative cooling) required more data so I began asking if you would release more. You have chosen not to do that yet, which is fine. It is your decision. But to say I have not not doing my homework is wrong. My results are just deductions from the partial information you have released.
Last thing. Do you still have confidence in the temperature/power plot you released in 2021 (which I replicate below where x-axis is input power in Watts and y-axis is temperature in degrees C)? If so, how do you account for its being so linear if radiative cooling is a major factor involved? Have I been overly influenced by the fitted straight line in the plot?