Nick Palmer wrote:
In response to Paul and Michel and Steven I will only repeat what I
wrote. Obviously you did not understand what I said. Read it again
without your knee-jerk prejudices.
If you're including me in the knee-jerk crowd who didn't understand what
you wrote, I take issue with that.
You asserted, among other things, that there is no thermal noise unless
the resistor is carrying current, _and_ that the shot noise depends on
the current.
I agree about the shot noise, and I said so.
However, AFAIK there _is_ thermal noise whether or not there's current,
just as there's Brownian motion whether or not there's current flowing
through the water, and the formula given in an earlier post, with which
you did not disagree, certainly does not describe something which
decreases with current. If it just _cuts_ _off_ at zero current, that
would be very strange behavior indeed! The formula given was:
> <<The thermal noise of a resistor is equal to:
> Vt = SQRT(4kTBR)
>
> where:
>
> Vt = the rms noise voltage
> k = Boltzmann's constant
> T = temperature(Kelvin)
> B = noise bandwidth
> R = resistance >>
Assuming the formula is correct as written, I really don't see a
current term in it anywhere. As far as I can see the RHS does not
depend in any way on applied voltage or current through the resistor,
nor on the resistance, for that matter. And the LHS is a voltage --
_not_ a coefficient to be multiplied into the voltage applied to the
resistor. If you disagree, please explain, because what you've said so
far doesn't elucidate this at all.
And I do not believe that the noise jumps from 0 to its "full-on value"
as soon as the current goes from exactly 0 to, say, 1 fempto-pico-amp.
In any case there's also thermal noise in the diode, as I believe I also
pointed out (though I didn't phrase it that way), and that is surely
where you should be hunting for the flaw in the design.
Besides, in order do work by
extracting energy from ambient heat with no heat sink, Paul's diodes
would need to rectify a current. There won't be one there.
<<you are taking this to mean that the noise voltage is generated solely
by the temperature of the resistor whether or not there is a current
flow and this is what the equation seems to suggest; however, this is a
sound engineer's equation, not a physicist's. I think it means that if
the resistor is ACTUALLY resisting current, then the noise voltage is
dependent upon temperature and the "shot noise" depends upon the
current. I put it to you that when there is no current though the
resistor, there is no electrical noise at all. Still no free lunch.
<<There is free lunch for *intelligent* thinking beings. :-)>>
No, there isn't. Therefore you are not intelligent,
Wrong. Paul is clearly intelligent, and so is Michel, even if both of
them are occasionally rude (and no doubt occasionally wrong). Paul may
very well be an autodidact rather than, say, a PhD solid state
physicist, and may have some soft spots in his knowledge base in this
area (as may I, and Michel, and you) but that is hardly equivalent to
his being unintelligent.
