Has Anyone tried to manipulate a mp3 or other file using Adobe Audition (Cool Edit) to get 125 on the Pos Peaks and 95 on the Neg Peaks?
I must admit to not feeling I really understanding what people are talking about with the concept of 125% modulation, since I was always taught that applying power to the modulator in excess of what it takes to achieve 100% modulation of the carrier would result in distortion of the audio and probable spurious emissions from the modulated carrier.
I've read many posts here referring to things like '125% positive peaks and 95% negative peaks' and that still sounds like an average of 120% modulation to me.
Now, it's the "flats" on the waveform when it gets to "past zero amplitude" on the negative peaks that cause most of the distortion. Or at least it's that side that will cause trouble first. But I would have thought that applying too much modulation even on the positive peaks would also distort that edge of the waveform because the carrier simply doesn't have enough to go beyond 100%?
The only thing I've been able to think of would be *maybe* (depending on how the transmitter circuit is set up) one could add a DC offest voltage to the signal right before it enters the transmitter so the negative peaks would never actually make it to zero and the positive peaks would basically cause the transmitter to have more than 100 milliwatts in the final rf on the positive peaks by drawing more current than usual from the power supply. Which could possibly get a signal with no clipping on either end of the waveform, but the average power (at least when fully modulated with a test tone) would exceed 100 milliwatts. Not sure what it would do to the sidebands.
Now, as I recall, testing of AM part15 for power is done off the unmodulated carrier.. So that might be a "gray area" or "wiggle room". Or for all I know it might be a standard practice in professional broadcast, where I understand they use quite a few interesting tricks.
Now, adding 20 milliwatts to a 100 milliwatt signal wouldn't sizably increase the range of the RF carrier, especially considering the low efficiency of the short antenna. But adding what amounts to 20% of "headroom" for the audio from the modulator could result in a signal that sounds louder without distorting (even more so with a bit of careful compression) so that within the range of that bit of rf it could maybe be heard better on recievers.
But anything you can do to an audio digital recording, "0 db" is (at least theoretically) the point past which making it louder will gain you nothing but the beginnings of distortion. That's "right to the rails". Brief transients won't usually be noticed, but other than that you're basically adding mud into your sound. So I don't think it could be done by just editting the waveform of a digital file. Might be possible to do it by adding in a small amount of DC voltage between the playback of that clean file that doesn't go past 0 db could "push" the side of the carrier that's away from the bottom trough of the modulation peak further, *if* the transmitter was able to actually provide a bit more than 100 milliwatts when that was done, even though it's unmodulated power was 100 mw.
In practice, you'd probably use a capacitor to keep the DC from flowing back into the audio chain and I'm not sure what one would want on the transmitter end or exactly how it would need to be set
up to take advantage of it. That'd be assuming my theory on how it might be being done is actually somewhere near correct, though.
Come to think of it, if it *does* amount to a DC offset, it *might* be possible to do to an audio file. AnalogX (a musician and man of many talents) wrote a DirectX plugin you can find HERE that might do the trick if you have audio software that can use DirectX plugins well. I'm not familiar with the new version that he says can handle realtime audio, but at the least it might allow adjusting the oddset on a test file to see if it works.. Assuming my guesswork is anywhere near the mark as to how the whole 125%+/95%- modulation thing works. And assuming the soundcard wouldn't automatically correct the offset at the output. If it did, adding the offset at the end of the audio chain would be about all I could think of trying.
I don't know a lot about adding a DCC offset on the output. To be honest, my audio background in recording usually is more focussed on eliminating such offsets if they're present on a file because someone used a crappy soundcard/mic to record it.
Anyway, there's my best guesses and thoughts on it at the moment.
Daniel
While I believe maximum transmitter modulation is a function of hardware, I do know that using a peak limiter and normalizing application will make a tremendous difference in maximizing audio levels delivered to the transmitter. This helps get more of the audio source modulated at a consistently higher level no matter what percentage of modulation your transmitter delivers.
The commercial packages have some capabilities in this area, as does the free application Audacity. Audacity plug-ins also will perform this function.
One of my favorites is the free stand-alone app called Peak Limiter.
From the author's site:
"Peak Limiter processes your 16-bit sound files using a sophisticated algorithm. Single peaks exceeding a user-defined level are softly compressed in such a way that the result cannot be distinguished from the original by the human ear. This permits rising the main volume of the sound file considerably without causing clipping or distortion! "
The free version of Peak Limiter plays an audio nag when you open the application but the basic functionality is unlimited. The pay version offers batch processing.
Peak Limiter is here:
www.x-ways.net/peaklimiter/index-m.html
Audacity is here:
Our Part15.us software list is here:
Experimental broadcasting for a better tomorrow!
Try Volume logic, there is a free trial period, then it is $20. I am impressed with it as that it seems any file will be processed by it unlike the Tomas limiter which seems to only process higher quality certain bit rate MP3's. So you can broadcast those lower bit rate talk radio archives with lower bit rates with that added punch.
I have the winamp plug in and use the "spoken word" setting for talk material and "FM magic" setting for MP3's at maximum drive. It really brings up the sound and gives it a good punch.
I cannot say if it will give 125% modulation, but it is worth a try. I do not know if one can use it for automated software unless they somehow patch it through winamp.