> I work in the pro audio world, and one the most common problems I see
> is folks running their systems 20 dB below clip and wasting headroom.
Been there - done that too.
> The general rule is that you want system gains to be set so that
> everything hits clip at the same time, and then run just below clip,
> even on the hottest peaks. Since a PSK signal is essentially
> a constant level, you don't have to waste headroom allowing for peaks.
> By contrast, speech needs 14 dB or more between average and peak
levels.
Unfortunately, PSK does not have a constant level envelope ... with
the phase inversion (BPSK) it simply can't. Most sources recommend
running a rig at about 6 dB back-off (ca. 25 watts for a typical 100
watt radio) to prevent distortion, clipping and IMD.
> If you run the PC 10 dB hotter and it's still clean, you've gained 10
> dB with respect to hum, buzz, and RF feedback. Granted not all sound
> cards are. :)
Unfortunately, most sound cards lack sufficient headroom ...
particularly
in the so called line out channel. Designers assume the user will be
using that to drive powered speakers and the gain will be made up
externally. It's not like professional audio or broadcast equipment
where the devices are capable of 10 or 20 dB above the "normal"
operating point ... specifically to stay away from clipping.
There have been several studies of sound cards for amateur digital
use and with great regularity the results are similar to those
reported here: http://www.qsl.net/zl1bpu/MFSK/pdfdoc1.gif
73,
... Joe, K4IK
_______________________________________________
Topband mailing list
Topband@contesting.com
http://lists.contesting.com/mailman/listinfo/topband
|
