Lynn,
I have Fs=300Hz, was using the collector to collect 512 samples, this should give sample size 1.7seconds, anddf = 0.6Hz. Am using the hanning window. Whenthe DC component peaks, it merges with the 1Hz. As you suggest, this probably is not enough time, and depending on how the window syncs up with the sine wave would change the mean or DC component. Changing the collector window to 1024 gives separations between DC and the 1Hz spectra and mostly settles the DC. At 2048, slightly better results, slower convergence. Good lesson to learn, thanks, but I still do not want to see the DC component, and even at 2048 eliminating FFT(0) is not enough. Eliminating FFT(0) and (1) does the trick. At 1024 samples this would eliminate up to 0.6Hz, at 2048 up to 0.3Hz. We'll see how it goes. Thanks again. Bob
