m+p international’s Post

📢 Knowledge that Resonates💡: Why does PSD have g2/Hz units? Another seemingly simple subject this morning, although we find this detail is so often overlooked and never really taught! So long as the test is running, no questions are asked, yet it’s important to know. If you measured the spectrum of a signal of a given RMS with three identical, calibrated systems, but each running with different frequency resolution settings, you would end up with three spectrums all showing pretty much the same RMS.  However, if you overlayed the plots, the traces would confusingly appear to be at different levels, albeit with  the same overall shape. This can be misleading when analyzing and comparing data.  The fact is, those traces taken with higher frequency resolution have more spectral lines of resolution to distribute the overall signal over and therefore each spectral line has a lower amplitude compared to a measurement taken with low resolution which has fewer spectral lines to distribute the signal over. A common analogy is sharing a jug of drink amongst different numbers of cups at a party. The Jug contains a finite RMS of signal but with fewer cups (lines of resolution), each cup gets more signal. The power spectral density plot is the answer to this problem as it normalizes the resolution out of the equation. If we take those three measurements and firstly square them (as is the convention with power measurements) to get an amplitude of g2, then divide the result by the resolution at which the measurement was taken, this gives us g2/Hz. The resultant plot will then show the three measurements as near identical and much easier to compare and analyze, with the area under the line representing overall RMS. Spectrums and PSDs can be compared on an RMS basis but for further information on converting between Sine amplitude and PSD, then please refer to our website: https://lnkd.in/eFPSK9GE   #expertsinvibration #vibrationtesting #vibrationtest #vibration #shakertesting #vibrationcontrol #vibrationcontroller #dataacquisition #digitalsignalanalysis