On Wed, 2008-03-19 at 10:46 -0400, Rick Denney wrote:
> Paul writes...
>
> > Sounds like the problem here is when someone reads, "Open the TX
> > full" they think that means TX at full output. When I started in PSK
> > I ran at a lower power level (15W on an ArgoV) but as I've been away
> > for about a year I posed the question to the digital newsgroups. The
> > majority said the current trend is to run the TX "full" but reduce
> > the power out by sound input - something about more linearity.
>
> Let me try and explain this in audio terms to make sure I have it
> right. I will be experimenting with PSK31 soon.
>
> When setting up an audio public address system, there will be at least
> two components that handle amplification. One is the mixer board, and
> the other is the PA amplifier. Audio power amplifiers have one or two
> controls: They have an on-off switch, and some (only some) have a
> master level control for each channel. Audio engineers routinely set
> those to full, or zero attenuation of the signal coming into the
> amplifier. Then, they reduce the amplification using the mixer board,
> which attenuates the audio signal going into the amplifier.
>
> The reason is they want the peaks of the audio signal to be farthest
> from the peak output of the amplifier as possible. They call that
> "headroom", and more headroom reduces the risk of pushing the power
> amplifer into overdrive, where it becomes non-linear (at best) or
> where it clips (at worst).
>
> So, my understanding from reading this is that we want as much
> headroom as possible between the audio input signal and the audio/RF
> amplification of the radio. Thus, we set RF output to max, and then
> attenuate audio input to a low value. That provides the maximum
> possible headroom to achieve the desired 30 or 35 watts of output. If
> the radio was 15 watts max RF output, we would set it to max, and then
> attenuate the incoming audio to an output of 5 or 6 watts to achieve
> the same result, it seems to me.
>
> In the end, it seems to be about headroom. Run the final amp and full
> amplification capability, but attenuate the input signal to keep it in
> the loafing range.
>
> With phone, we tolerate more distortion in the signal to get a little
> higher average loudness. That's why we have speech processing, etc.
> Broadcast stations do that, too, and even when fidelity standards are
> really high, they still worry about how loud the signal seems. But
> with digital, we don't need or want any of that distortion.
I hadn't thought of it as a "headroom" issue. The main point is that a
PSK31 signal cannot tolerate any non-linear amplification. Actually,
the intelligence still gets through, but the signal is badly distorted
and takes up unnecessary bandwidth. The fact that the intelligence gets
through can create a problem for some operators. They figure that if
people are copying them, their signal must be okay.
What I have noticed, with TenTec radios (at least the Omniv V and VI) is
that the power output adjustment doesn't adjust the gain of the final
amplifier, it just adjusts the level at which ALC kicks in to reduce
output power. Thus, by running the Power output at max, you have
reduced the chance that ALC will kick in and distort the signal. You
use the input level (i.e. the soundcard output level) to adjust the TX
power.
On other manufacturers radios, it seems that the Power output does
adjust the gain of the final amplifier, so reducing power on them seems
a good idea for PSK31.
Those are my observations.
73,
Mark
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|