Impedance Matching

First, I have no idea what PSD stands for.

You do not explain why you uses such a high impedance preamplifier. There are many made to work with low impedance sources such as magnetic phonograph pickups or dynamic microphones. To truly match the two devices you need a circuit between the components that make your 500 ohm resistor look like 100 Megohms. When you do that, less than a picoFarad stray capacitance will screw up your broadband match.

My conclusion is 1) You do not know what you are doing; 2) You do not really want to match; 3) You do not know what you want.

Bill

-- Ferme le Bush

PSD = power spectral density

I use a high impedance preamplifier because that is what is available to me, it is a Stanford low noise preamplifier, model SR560.

Your conclusions are fairly accurate, I don't really know what I am doing. I want to measure noise across a resistor and have been using a preamplifier + spectrum analyser, I should be getting white noise (a flat PSD) but I am not getting anything close to this and have been told that this is due to a lack of impedance matching in the circuit (maybe this is incorrect?). Now I have looked into impedance matching and found that an inductor capacitor combination will do the trick but will only match at one frequency, for broadband matching i need

2capcitors and 2 inductors but I do not know how to calculate their values for my specific case.

As I said earlier any advice would be appreciated.

Karl

you might want to read this article

then this one

here is almost a book one the subject

there is quite a bit more out there.

the question i have is why do you want to do it?

Thank you for the articles, I will give them a look.

I want to do this because I am part of a research group that is working on thin film devices and we need to carry out noise measurements on them. Measuring white noise across a resistor is only the first step, my ultimate goal is to measure 1/f noise in a tunnel junction but I need to be able to do the basics first.

It looks like you are trying to look at resistor noise voltage. Instead, try looking at the noise current into a short circuit. Emulate the short circuit with a transimpedance amplifier with much feedback. That feedback allows the amplifier to cool its feedback resistor (active refrigeration) so that you can measure the Norton equivalent noise current of the resistor.

To me, it seems that you are in way over your head. You really should invest the time NECESSARY to understand noise, of components like resistors and amplifiers with feedback. Otherwise, you will never understand what you are doing.

Bill

-- Ferme le Bush

A lot of the noise that a receiver or other instrument detects is caused internally. Thus would be different with each detector. Or you might be looking for noise at the output from some device and what you actually are looking at is noise generated in your test equipment. Keep that in mind before you spend a lot of time chasing your tail.

Josh