On 11/07/2010 11:30 PM, peiman khosravi wrote: > Thanks jörn for your enlightening notes. > > Firstly, the higher frequencies are there as the byproduct of certain > FFT processes (e.g. spectral stretching). I am transposing or stretching > the spectrum and so some of the lower spectral materials move up the > spectrum. I could filter them out I guess but then why filter them if > they are not audible.
i see. well, maybe do filter them :) > One thing that I have noticed when experimenting with spectral > processing is that very high frequencies, even if inaudible or almost > inaudible can cause distortions in the speakers, and that such > distortions are usually heard at lower frequencies as a sort of > saturation or interference with a particularly unpleasant roughness. you mean high frequencies at very high levels (enough to overload the amp), or even at low volume? >> or maybe you have power amps which are susceptible to hf oscillation >> (which might saturate them and then introduce audible distortion in the >> 20-20k band). watch your level meters to check. >> > > Could you expand on this please? i've seen p.a. power amps die after starting to oscillate in the ultrasonic range. only a month ago, a qsc amp died on me this way. it had 3 12" monitors connected to it (and yes, it's being marketed as 2ohm-safe), and all of a sudden, its clipping leds went full on. funnily enough, it even *created* a signal at its inputs - when connected to a cheap small yamaha mixer, the mixer's master meter would max out (and fall back to zero when the amp was disconnected, so i can be sure it wasn't the mixer that generated the problem signal). all this time, there was only a very inconspicuous hum to be heard in the monitors. i was able to break the oscillation by removing one or two monitors, but some permanent damage had been done. now my uneducated guess is, if an amp-plus-speakers circuit is susceptible to such oscillation, maybe excessive hf content triggers it, and drives the amp into saturation. (even if it's not a runaway oscillation but just long ringing at loop gains < 1.) a clipping amp might produce undertones as well as overtones, so that might actually produce artefacts in the audible range. best, jörn -- Jörn Nettingsmeier Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487 Meister für Veranstaltungstechnik (Bühne/Studio), Elektrofachkraft Audio and event engineer - Ambisonic surround recordings http://stackingdwarves.net _______________________________________________ Sursound mailing list [email protected] https://mail.music.vt.edu/mailman/listinfo/sursound
