Photodiode Question

have to

history.)

that

Here:

You can make a room-temperature amplifier that has very low noise temperature... I think some gaasfet-based microwave amps get down to something like 50K noise temps. And you can use such an amp to simulate a resistor well below room temp. It doesn't violate the laws of thermodynamics because the amp needs power to do its thing.

If you connect two resistors with wires, and they're at different temperatures, the Johnson noise difference becomes a thermal conductivity between them... the hotter one makes more noise, and pumps power into the colder one. I did the math once... it's something like 10 orders of magnitude weaker coupling than the thermal conductivity of any conceivable wires that you'd use to connect them.

John

have to

history.)

that

Here:

If you connect a pin photodiode or a PMT, with a little gain maybe, to a scope, and hit it with a light pulse, the top of the pulse will usually be a lot noisier then the baseline (dark) level. It's pretty obvious if the light is in the roughly 1 microwatt sort of range. It's the quantized photons hitting the detector randomly... shot noise.

I guess it was named for the sound of shot (as in birdshot) falling onto a surface. Hisssssss.

John

Try looking for the patents on Faraday detector amplifier calibration sources. They are typically around 0.1nA with fairly low noise and very high reproducibility. Not sure how much of it is trade secret.

ISTR the resistors for this job are effectively single sourced and in the range 10^9 to 10^11 and some kind of doped conductive glass. They are low noise (at least for low frequencies) but also behave a little bit like they have distributed internal RC and battery behaviour which requires external compensation.

Regards, Martin Brown

Last time I worked with them was about 25 years ago. It was used as a oscillator/ramp generator for a proportional heating control system in a 'Centrifichem' blood chemistry analyzer. Oppie

(Hal

ites:

don't have to

.. =A0(CdSSe's

n history.)

tory?

ate spam.

an

LDRs that

er

AnalogOptoiso.pdf

about

s. =A0Are

DR's?

t's

oise

if

ly get

eries

e. I'm

he

rrent

or

noisy

ssy it's

d on

en

rent)

nt

y...

(R

k' to

hat.

's

ve

ext -

The noise temperature of a FET opamp looking at say a 1 Meg or 10 Meg resistor can be truly tiny. (I guess noise temperatures only really make sense if you stick to a fixed input impedance.)

George H.