Precision about Lock-In Amplifier

[Archer]

I was doing some works on Lock-In Amplifier of weak signal detection at
video frequency.
It is reported that some Lock-In Amplifier could do with deta frequency
about 0.0004Hz,
I think it is incredible. Could the oscillator and low pass filter can work
in such low tolerance?
How could they get it?
With Wein Bridge Oscillator and Chebyshev filter?
Help would be appreciated!

 

[Bob Masta]

On Wed, 12 May 2004 21:36:41 +0800, "Archer" <xydarcher@126.com>
wrote:

I was doing some works on Lock-In Amplifier of weak signal detection at
video frequency.
It is reported that some Lock-In Amplifier could do with deta frequency
about 0.0004Hz,
I think it is incredible. Could the oscillator and low pass filter can work
in such low tolerance?
How could they get it?
With Wein Bridge Oscillator and Chebyshev filter?
Help would be appreciated!

First, for those who haven't yet encountered the
mysterious lock-in amplifier, a quick tutorial:

A lock-in amplifier essentially just multiplies the input
signal times a reference frequency. The output of
the multiplier will contain sum and difference products
as when you multiply any two sinusoids. This is
followed by a low-pass filter, so all you read on the
output is the DC component, which gives you the
measure of any inpu component that was at the
exact phase and frequency of the reference. (You
use quadrature references for arbitrary phase.)

The "lock-in" part of the name comes from the
phase-locked loop that syncs the reference oscillator
to some reference input you supply. A typical example
is an optics experiment where you have a chopper
wheel that interrupts a beam. One photodetector can
pick up the chopped beam and drive the reference,
while a sensitive detector picks up the results of the
beam after it has passed through the experiemental
setup.

Since the output of the multiplier is DC, you only need
a simple low-pass filter to get ultra-fine resolution.
Many lock-ins used to use a simple single-pole R-C
filter and just used a very long time constant. More
modern ones typically use a couple of poles. They
don't normally use anything more exotic, since that
just sharpens the band edges... the real selectivity
is all from the ultra-low cutoff, so a single-pole is
really OK for most everything.

Of course, the problem with this or *any* system
that is measuring with this high of frequency resolution
is that it requires a long time constant, so you get
lousy response to sudden changes. (Even digital
methods have exactly the same limitations.)

Note that the multiplier and low-pass filter is
exactly what you have in an FFT or DFT, just that
they have lots of 'em and lots of reference frequecies.

Note also that the early lock-ins used switching methods
as the multipliers, so they were effectively multiplying
by a square wave. The drawback was that they were
sensitive to odd harmonics of the reference as well
as the reference itself, but often that wasn't a problem.

Hope this hleps!





Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com