Kauai Inventor said:
RE 48: The point of having a demonstration is beyond the point of needing true knowlege seeking investment. I live at the lower edge of financial comfort in a near perfect climate. A lot of people paying a little collectively could pay for a prototype so it should be done that way. Setting it up is next.
RE 49: Please visulize the diodes sitting in their place inertly with Johnson noise inside which flows out preferentally making the common anode buss negative which can be tapped for electrical power. Johnson noise supplies the electrical power while cooling the diode array.
I Thank all the contributors for their views.
Aloha
Oh i understand what you
want to happen, but you're dreaming if you think it
can happen. If you really want i'll go back into my notes and tell you what sort of temperatures you're looking at to generate that level of noise, i'll bet you it's higher than the silicone can handle.
For the posters who don't know their quantum mechanics and/or micro electronics what KI intends to do can be described as follows.
Half fill a metal tank and heat it, the water in the tank will boil and start to boil over (like a pan on the hob), by putting a tube near the top of the pan you can use the water that makes it high enough to flow round your system and to drop back into your tank, as it flows round the sytem it might drive a water wheel for example. The water being electrons and the tank being our power source. (There's a valve so the water can only go round the loop one way)
Now this tank gets topped up regulary so it can't run out of water, but unless you get it hot enough it won't be able to reach the tube. For various reasons the height of the tube above the water is fixed (due to the energy bands of the relevant materials, note a material limitation), as such the only way to make it work is to heat up the liquid to get it to boil over.
Notice the term boil, as you need to put a lot of energy into the system for this to work. In
theory (*note, see next paragraph) if you were to find the right materials it could work as the tube would be low enough for just a teensy weensy bit of heat might get the water to rise up and flow out the tube.
If you didn't know your theory that is. Let's say that a single electron manages to make the transition from n to p side, that then makes the potential V0 even greater, increasing the field (raising the tube) and shifting another electron back into the n side. In terms of our analogy this is the same as making the tube go even higher up the tank, and tilting it so the water that has made it in will tend to flow right back down again.
So, why can't it work?
1) To get a current flow you'd have to create a noise power
on one side of the diode only greater than V0 where v0 = kT/e ln(Pn/Pp) that's not going to happen. Material science can't do it, run the numbers if you want to check.
2) Once you get a curent flow you increase V0 as you've put the diode into reverse bias, meaning it will not conduct.