CURRENT AND SIGNAL TO NOISE RATIO:
The easiest way to increase the signal to noise ratio is to increase the current going through the sample. The current can be adjusted by adjusting the amplitude of the reference sine wave generated by the lock-in amplifier. The excitation current through the sample can be measured by measuring the voltage drop across a resistor placed in series with the sample current wires.
Current through the sample must be chosen carefully, however. The higher the current through the sample, the higher the signal to noise ratio, but also the more the sample will heat. When the sample heats up, it will be at a higher temperature than what the thermometer reads. This obviously will render your data inaccurate. Check carefully, don't just make the current low. Sometimes the signal to noise ratio can be improved just by changing the frequency of the excitation current, finding a "quieter" region in frequency space.
This will be determined primarily by sample impedance R and cable capacitance C. It is usually desirable that:
FREQUENCY OF EXCITATION CURRENT:
Sometimes, the signal to noise ratio can be dramatically improved simply by changing the excitation frequency. Which frequency works best is often a characteristic of the sample. You can get this information by referring to past experiments that have worked with a particular sample.
Signal voltage should not be frequency dependent. If a certain signal characteristic seems unusual or unexpected the first thing to do is to see if it is repeatable at a range of frequencies. Again, you should compare the readings to the resistance measured by a low frequency bridge, such as the LR700, and try to find a higher frequency that produces the same value. Sometimes, the only way to determine excitation frequency is by trial and error, one shot at a time, until the signal to noise ratio and the time response match what is needed.
For quasi 1D charge
density wave systems, or spin density wave systems, use as high a frequency
as possible. Very high resistance samples need a lower frequency, because
cable and other reactance begins to have a higher effect at higher frequencies.