ckleung17 wrote:
I would like to reproduce your results with my unit, but I don't know what equipments should I need? I have a multi-test meter, an audio generator and an oscilloscope.
If you mean the gain and phase plots, what you really need is a signal generator, an oscilloscope, and, if possible, a good AC voltmeter. Your multi-test meter will not really help because they will generally only produce accurate AC readings at 50Hz to 60Hz. Even the True RMS volt meters like the Fluke 179 are only specified to measure AC voltages from 45Hz to 500Hz. If you look at the picture of the unit under test on the project page, you'll see the Leader LMV-185A Dual AC voltmeter that I use. This unit is specified to produce readings accurate to 2% from 20Hz to 200kHz
The best way to take the gain measurements if you don't have a wide-band AC voltmeter is to read voltages directly off of the oscilloscope. Just make sure to periodically zero the inputs to make sure that the traces are still baselined to zero volts. The phase measurements are slightly more problematic. The way I did it was to use a plotting compass and take measurements of the relative zero crossings between the input and output waveforms accounting for the 180deg mid-band phase shift. If you take a measurement, divide it by the length of one full cycle, and multiply by 360, you get the relative phase shift in degrees. Then just repeat both procedures at a number of frequencies to generate the response plots.