G0RUZ Conrad wrote:
>> I can't see where there's a difference.
>>
>> When searching for references to the srticle I remembered, I was
>> struck by how many top end broadcast audio processors use split band
>> clipping and limiting.
>>
>> 73, Steve
>
>
> This is certainly true, some of the more up to date ones split the
> frequency into five bands and then perform adaptive compression with
> different time constants for each frequency band. They are able to
> give 12dB of increased average level without audible artefacts on
> full bandwidth music signals. They are implemented in DSP and have a
> slight 'look ahead' delay (<4ms)incorporated to prevent peak
> overshoots. As a square wave can be described as a Fourier series of
> sine waves then it seems logical that removal of harmonics by a
> filter can indeed restore the original waveform, if the odd harmonics
> are removed then it's no longer a square wave. Because of the limited
> bandwidth of an SSB transmission a two band approach is more than
> adequate and should provide excellent results if implemented
> properly. I may just make one using a DSP sometime soon, it's easy
> enough these days using Matlab. No programming required, just select
> the appropriate DSP modules, set the parameters, see how it behaves
> and if you like it tell simulink to write it to a TI DSP. You can
> tailor the response to exactly what you want. There is no
> disadvantage to doing it at audio or baseband provided it's
> implemented properly.
I found my copy of the Ham Radio article, which uses four bands. The heart
of it is just two quad opamps. I'd imagined that the pro units used more
bands, but DSP can be clever stuff.... 'Look ahead' always struck me as a
good idea - I've an article by G4BWE describing a compressor using a digital
delay line so that peaks spotted on the way in are attenuated when they come
out.
Cheers,
Steve
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.7.0 - Release Date: 17/01/05
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|