[Elecraft] SSB transmit audio - Where's the punch?

[Wes Stewart]

With that introduction I guess I have to respond.

Actually, credit for the original concept belongs to Walt Schreuer, K1YZW.  He developed the "SPB-3" sold by Maximilian Associates in about 1976.  No schematics were provided, of course, so I and my late friend Jim Metzer, W7TKR, produced our own way to do it.

See: http://k6mhe.com/n7ws/S_PROC.pdf

Our simplified method for accommodating the phase shift through filters does lead to more passband ripple than more elegant phase correction, but operationally it's never been an issue.

IMHO, the split-band approach is better than rf clipping. Split-band clipping was the choice of the engineers of one of the arguably best sounding radios around, the Kenwood TS-870.

I personally know of several hundred of these that were built around the world, not counting however many kits were sold after the magazine editor founded a kit company on my design, the RADIOKIT DX-1.

Wes  N7WS


--- On Tue, 10/5/10, Joe Subich, W4TV <lists at subich.com> wrote:


Jim,

 > This only works with full bandwidth audio (that is, 20-20,000 Hz);
 > the bandwidth we transmit would fit into one of those individually
 > processed bands.

Wes Stewart, N7WS had a very successful split band approach for amateur
use years ago ... it was butchered by Alpha (the Vomax) but three bands
(300 - 600 Hz, 1200 - 2000 Hz and 2000 - 3500 Hz) is a very good choice
for "communications/talk."  By keeping each band less than one octave
in width, IMD and harmonic products from clipping can be minimized even
with analog techniques.

With modern DSP technology one can add phase scrambling (rotation),
per band noise gating, pre- and post-clipping equalization and use
half octave bands with minimal hardware cost to achieve very clean
compression/clipping with very tight dynamic range even in noisy 
environments.

73,

    ... Joe, W4TV