|

One of the
biggest obstacles to overcome in the quest for accurate measurements
is excess noise and interference that obstructs the signal from your
sample. In other areas of this site, there are explanations of the
many steps that can be taken during the experiment to reduce noise.
Note, however, that the connections between the magnet and data
acquisition devices are arranged to help reduce as much noise as
possible. A central component of this arrangement is the Lock-In
Amplifier.
Lock-In
Amplifiers detect the signal and any noise associated with it
(including thermal, shot and 1/f noise). Because of this, it is
necessary to take noise into consideration when choosing a frequency
for the measurement. There is substantial 1/f noise at zero
frequency, making DC undesirable. Measuring at a higher frequency
usually results in a less noisy environment. However, if you go too
high in frequency, cable capacitance and inductance may obscure the
signal. Therefore, it is best to test your system at chosen
frequencies to determine the appropriate one for your particular
sample and measurement.
Taking an
average of the signal is the first way to eliminate noise and
isolate the signal. Averaging over a time t enables you to measure
noise over a bandwidth ~1/t, no matter what the excitation frequency
is. The noise you will see is
just

Where is the
noise density at that frequency.
Once an excitation frequency
is chosen, a mixer is used to isolate the signal. A mixer is an
electronic gadget that multiplies, instantaneously, two voltages.
The two common modes of operation of the mixer and related signals
and excitation frequency are homodyne detection and heterodyne
detection.
Heterodyne
detection is when . This is
better. Homodyne detection is what lock-ins use and is when . If
then
then
The mixer
output has components at 2f. Also, at DC the amplitude of the DC
component is dependent on the phase of the signal relative to the
reference, and the size of the signal.
It is important
to know that the averaging time determines the bandwidth. This is
crucial if the signal is changing. For example, if the signal varies
at 100 Hz, then the averaging time has to be less than 10ms.
Conversely, if
there is an interference source at 60 Hz, and you want to measure at
200 Hz, while the signal varies at 10 Hz, then:
If you choose a
filter time constant of 50 ms for
averaging:
Then the filter
bandwidth is

|