A/D Performance Measurement

A/D Performance Measurement with ScopeDSP™, Part 4

Here are the first few items in the Frequency List:

The Frequency List takes a little getting used to, but soon becomes indispensable for this kind of analysis. First it shows the summary statistics that ScopeDSP calculates. ScopeDSP uses the following definitions to calculate them:

Fundamental: The frequency of the component with the largest amplitude.
SFDR (Spurious-Free Dynamic Range): The ratio of the power of the largest component to the power of the second largest component.
SINAD (Signal-to-Noise-And-Distortion): The ratio of the largest component to everything else.
SNR (Signal-to-Noise Ratio): The ratio of the largest component plus its harmonics to everything else. Caveat emptor here!

The power at DC (zero frequency) is excluded from all of the above. (DC is not considered to be signal, noise, or distortion. It's just there. This is a very pragmatic definition: A/D samples often contain DC, which can usually be filtered off or ignored.)

The Frequency List is sorted in terms of decreasing component power. The "Frequency" column just lists component frequencies, rounded to three decimal digits. "Bin" refers to the DFT "bin number", that is, the index into the Frequency Data array. "Harm" is the harmonic number. (The fundamental is defined as harmonic number "1".) "dB Below Peak" indicates the power, in dB, below the fundamental. In this case, the "Bin Power" and "dB Below Peak" columns are identical because the decibel units are referenced to the peak component (dBc). "Total Smaller Power" is the power of all components that are smaller than the given component. Note that according to the definitions above, the "fundamental" is always the first "Frequency" entry, and the "SFDR" will always be the second entry in the "dB Below Peak" column.

Looking in the "Harm" (Harmonic) column, the source of the component "near" 1.1 MHz is now clear! It is actually the second harmonic of the fundamental at 1.094 MHz. It is a little surprising at first that the frequency of the second harmonic could be less than the frequency of the fundamental, but if you work it out you will find that it is indeed the harmonic, after aliasing has been taken into account. Working out harmonic aliases is quite tedious to do manually, so the Frequency List is very helpful that in that regard.

Initial Conclusions...

Home | Previous | Next | Download | Iowegian