How to monitor a Sine wave for noise

[johnsold]

FFTs in general and the Tone.vi which uses the FFT internally work best with relatively large numbers of cycles of frequencies which are not changing during the calculation. 1000 samples at 500 kS/s is about 10 cycles of the 5.2 kHz. That should be enough. The one cycle at 100 samples probably will not give good results. Also, the frequency resolution depends on the number of samples fed to the FFT VI. Look at the help for details. If the frequency will be changing during the measurements you may want to look at the STFT or other time-frequency spectrogram techniques. They can be tricky to use and interpret, so plan on some time with test VIs and simulated data where you know what is going in to the measurement system.

Do you have a good theoretical model for how the system should behave? Sometimes other techniques based on knowledge of the system can be more effective.

Lynn

[dnlwthrn]

Good theoretical model?  Yeah right.  That would mean I had all the information.  In my job, we never get all the info, so I'm kind of used to it.

Thanks for the tips on the FFT.  I'll try to play around with the Tone.vi and FFT.vi and see what I can do.  I'm currently using a function generator to give me an input signal, so I can tweak almost anything.