04-04-2012 11:26 AM
Hi ,
I am trying to figuer out how to calculate thePSD of the ADC output. This is what I have done so far ...
FYI. My proff said the below : I dont know what he meant by taking 100 000 points ...
"Get the power spectral density of the A/D output. Collect big arrays like 100,000 points."
Thank you. Please find the attached file to see my output.
04-04-2012 11:44 AM
that is because you are using an array of the data and not a waveform that contain the time information, therefore you don't obtain the frequency axis. For the 10000 points, you only need calculate the time of your generated signal and the sampling time.
for example, the tool "simulate signal" give for default 100 samples, 1000 samples per second, Therefore you can obtain an array of 100 point every 0.1 sec. and for 10000 points need wait 10 sec.
Perhaps with some of your code we can help you more.
04-04-2012 12:43 PM
Can you please explian this more clearly....
Suppose I want to collect 100,000 Points at 250 ms/sec. what does this mean ?
how long will I have to wait ?
Abhi
04-04-2012 02:17 PM
Abhi,
We will not do your assignement for you, but we can try to help you understand how to do it.
Why large arrays? What is your ADC resolution? If you do not have any data or only have a very small number of points in some bins, how meaningful with the PSD be?
Axis labels are the default labels for a graph. Most likely no one (you) changed them. The labels do not adapt to the source of the data. The programmer needs to change them. You can build a waveform and put the dt in that way or use property nodes to scale the graph axes according to the values you specify.
Suppose I want to collect 100,000 Points at 250 ms/sec. what does this mean ?
Usually ms = milliseconds. Milliseconds/second does not make any sense. Could it refer to megasamples? PSD measurements/calculations on high speed ADCs seem more likely. 100,000 points at 250 MS/s only takes 400 microseconds.
I have seen discussions in the literature about the choice of waveform for testing ADCs. The problem with sine waves is the the signals spends different amount of time in different bins. Thus, the PSD may be biased because more samples occur near peaks than near zero crossings. On the other hand it is difficult to make a linear ramp which covers the full scale of the ADC with linearity which is better than the resolution of the ADC.
Lynn