[BC] FM Stereo invention

[WFIFeng@aol.com]

In a message dated 02/19/2007 01:58:03 PM Eastern Standard Time, 
Robertm at broadcast.net writes:

> Look at it in these terms. In a switching system where the left and 
>  right channels are sampled at 38 kHz, if the signal is pure mono the 
>  left and right channels are identical and there is no switching 
>  component in a perfect world of zero rise time switches. If there is a 
>  difference between left and right, then that difference appears as alias 
>  of 38 kHz which is the sum and difference of the baseband frequences and 
>  38 kHz. In other words a 38 kHz DSB signal. That is what makes the 
>  switching and matrix system mathematically equivalent.

Right, and I have observed exactly this on my oscilloscope. However, the 
deficiency of the switching menthod becomes immediately apparent when there is a 
large difference between the channels, such as one channel being silent. The 
audio is chopped on/off at 38Khz, at nearly a 50% duty cycle. (Perfect switches 
would make it 50%) That's gotta add distortion along the way. Someone else 
mentioned the harmonics, which can intermod with subcarriers (if present) or do 
other nasty IMD tricks.

While the switching method may be *effective*, I don't think it's the best 
way. Someone else pointed out the decoder flaws, and again, those are good 
points... and may, in reality, negate the effects of the "cheaper" switching method.

Willie...