I am using PXIe 6368 card and am trying to calculate self noise of each channel in this card. For this purpose, i select a differential channel, set input range to +/-5 V, short both differential inputs and performed following steps:
1. Collect 'x' no of samples of voltage signal
2. Perform FFT of size 'x' on collected samples
3. Divide each complex output of FFT by 'x'
4. Find the absolute of each complex output of FFT
5. Multiply the output of above step by 1.414 to get Vrms against each frequency bin
10. Take log10 and multiply by 20 of each output in above step
11. Add 10xlog10(x/2) in each output of step above to compensate FFT gain against noise
The result is self noise floor graph of a channel in terms of dBVrms Vs frequency.I got average noise floor around -68 dBVrms across frequency band.
Now, according to specifications of this card at page-3, it has random noise of 134 uvrms for +/-5V range scale (which in my assumption is self noise/channel of this system) which results in -77dBVrms approximately.
If my method of measurement and calculations are correct and also the assumption that this card has self noise spectrum of -77dBVrms for a frequency band, there is difference of around 9 dB between measured value and theoretical value.
Kindly guide me whether i am doing right calculations and if i am right, what may possible reasons for this difference of 9 dB?
Thanks & Regards
I can't speak to the accuracy of your method, but I use "AC & DC Estimator.vi", which will give you the RMS value of the AC content of the input signal:
For the ideal +/-1 Vpeak sine waveform in the code above, I get an Vrms of 0.707, which matches theory. It's unclear from your post, but it sounds to me like you're just tying AI+ and AI- together. Make sure that you're tying both channels to AI Ground to give externally-coupled noise a path to ground. I usually grab a minimum of 1 full second of data at the maximum sample rate of the card. So, if you're testing one channel at a time, set the sample rate to 2 MS/s and the number of samples to at least 2 M and put the waveform you read with DAQmx Read directly into the AC & DC Estimator VI above. Let us know what happens!
Sorry for delayed response. Unfortunately i am not more familiar with Labview and mostly work in c++ using NI APIs, however i am trying to implement this in labview.
However, can you kindly comment whether my assumption that " random noise of 134 uvrms for +/-5V range scale (from 6368 specifications) " is self noise of PXIe 6368 Card is right or not?
Your assumption is correct. The random noise specification quantifies the typical noise on an AI channel when the AI channel is configured for differential acquisition with both + and - channels shorted to AI Ground. In other words, you're measuring the noise that the DAQ board itself contributes to the measurement.