In a nut shell, what a lock-in amplifier does is measure the amplitude Vo of a sinusoidal
voltage,
Vin(t) = Vo cos(ωot)
where ωo = 2πfo and fo are the angular- and natural frequencies of the signal
respectively. You supply this voltage to the signal input of the lock-in, and its meter tells
you the amplitude Vo, typically calibrated in V-rms. What makes a lock-in different from
a simple AC voltmeter, which is what I just described, is that you must also supply the
lock-in with a reference input, that is, a decent size, say 1 volt p-p, sinusoidal voltage that
is synchronized with the signal whose amplitude you are trying to measure. The lock-in
uses this signal (like an external trigger for an oscilloscope) to "find" the signal to be
measured, while ignoring anything that is not synchronized with the reference. In
practice, the lock-in can measure voltage amplitudes as small as a few nano volts, while
ignoring signals even thousands of times larger. In contrast, an AC voltmeter would
measure the sum of all of the voltages at its input.
Let's consider an example. Suppose the signal is a 10 nV sine wave at 10 kHz. Clearly
some amplification is required to bring the signal above the noise. A good low noise
amplifier will have about 5 nV/ √Hz of input noise. If the amplifier bandwidth is 100 kHz
and the gain is 1000, then we can expect our output to be 10μV of signal (10 nV x 1000)
and 1.6 mV of broadband noise (5 nV/√Hz x √100 kHz x 1000). We won't have much
luck measuring the output signal unless we single out the frequency of interest.
If we follow the amplifier with a bandpass filter with a Q=100 (a VERY good filter)
entered at 10 kHz, any signal in a 100 Hz bandwidth will be detected (10 kHz/Q). The
noise in the filter pass band will be 50 μV (5 nV/√Hz x √100 Hz x 1000) and the signal
will still be 10 μV. The output noise is much greater than the signal and an accurate
measurement can not be made. Further gain will not help the signal to noise problem.
Now try following the amplifier with a phase-sensitive detector (PSD). The PSD can
detect the signal at 10 kHz with a bandwidth as narrow as 0.01 Hz! In this case, the noise
in the detection bandwidth will be only 0.5 μV (5 nV/√Hz x √.01 Hz x 1000) while the
signal is still 10 μV. The signal to noise ratio is now 20 and an accurate measurement of
the signal is possible.