Comment by yetihehe
Finally making a simulation of my heat engine. It's kind of like stirling engine, but working on entirely novel cycle, theoretically should be able to achieve over 80-90% efficiency (yes, more than Carnot theorem says). I had this idea for 15 years already, but didn't have enough free time to make prototype. Now I don't have much time, but thanks to AI I have kickstarted simulation program (it told me which equations to use and how, I do most of my programming myself), turns out it's not as hard as I've imagined when you know which equations to use. Still working on it at max several hours a week because I have two other programming jobs. If it works, it will noticeably bump humanity on Kardashev scale.
carnot theorem does not put an absolute limit , a carnot cycle can have any efficiency as long as you have reservoirs hot and cold enough ,
besides that if you could be more efficient than carnots cycle at two temperature you can spontaneously extract heat from a cold object and out it in a hot object without any input energy , this would mean you can have perpetual motion . we are not just going to bump a kardashev scale , were going to truly max out and go beyond.
fun fact the U S patent office has been bogged down by so many perpetual motion applications they have made it policy to outright reject such applications without a working model satisfactorily demonstrating it