enahs wrote:
One tech firm tested a number of the iPhone apps against very sophisticated meters. They found only the SPLnFFT app to be accurate. In fact, that app was reported routinely as accurate as a very expensive meter. But in all of this discussion, the suggestion to measure my Hondas against a new one is the only reasonable approach. The actual decibel count is not the issue. It's the change that matters. The "Noise and Directivity" test would also provide a reference point. IMHO, I would guess that the muffler packing deteriorates with run time.
i have SPLnFFT along with a separate folder of others. They are all basically the same out of the box. Since we retain Acoustical Engineers for studio work so it was easy to cal mine, but the big issue is the response of the mic on the iPhone that is tailored for voice, not generator noise. The analyzer apps tweak the app to appear flat, but the spectrum WE are interested in is really 200Hz and lower, and that's where the mic starts rolling off. Even the cheap $50 Radio Shack meter is better for these measurements IMO