On Mon, 2006-07-17 at 21:27 -0700, Ron Castro wrote:
> And just how does one do that on a radio that has the ALC level fixed in the
> firmware and controlled by DSP? This isn't like a Kenmore that has a
> "carrier level" control on the front panel. There is no menu setting to
> control it either. The ALC light blinks on CW at all power levels settings
> from 1 to 100.
There is an ALC setting, its for power output. The manual says ALC sets
the power output whether 1 watt or 100 watts. That compensates for the
varying gain of the transmitter circuits from 160 through 10 meters.
Then that not allowing for a gain setting to hit the ALC limit softly is
the software problem, not as others have described it.
>
> The good news is that since the ALC control is so tight and consistent,
> there is no need for the traditional ALC loop from an external amplifier to
> the exciter (if anyone ever used that--I never did)...just turn down the
> power control.
ALC derived in the PA has some advantages. It protects the PA and it
should keep the PA in a linear region. Suppose the PA used two tubes and
one heater circuit went open. So the PA had only one tube, though
mismatched then it needed only half the drive. The auto tuner in the
exciter took care of the mismatch and without PA voltage derived ALC
drives the PA to full power with one tube, RTTY. Then the owner notices
the air stream above one tube is hot and the other cold and swaps tubes.
And continues to operate full amplifier power from one tube. The tubes
when I got to them had blistered darkened silver plate. Yes the 8874
will run twice rated power for a while. Wasn't something I did, though
when I returned the amplifier to its owner after replacing the tubes and
the metering resistors that had drifted and I used it to boil the oil in
a cantenna checking for balanced tube heating, I insisted on connecting
ALC from PA to exciter and setting that PA ALC level for 90% of
saturation power. Didn't need to replace those PA tubes again.
Or take the situation where the microphone gain was chosen to set the
gain of the radio and the band opens up, the operator gets excited and
shouts... Good ALC preserves the linearity, no PA ALC gets the shouts
splattered.
Its sure that the PA gain is not constant from band to band, so that
setting a drive level on 160 meters where the gain is probably highest
isn't enough on ten meters. Do you KNOW without good test equipment not
normal to the ham shack what drive the PA needs at all frequencies? How
does that change when the PA is loaded a bit differently? It does change
but how much?
Yes a fast acting fast decay ALC introduces distortion. And that fast
acting on every dit causes a click and the shortened rise time from the
rise time setting. If the ALC was fast acting and slow decay there might
be a click on the first dit but the rest transmission might not have any
at all. At least not until after the next pause. Generally the ALC
clipping is less distorting than PA clipping.
Granted the cool, calm, collected and perfect operator will never
overdrive the PA. The excited operator or the neophyte looking for
higher average power may only be calmed by properly set and working ALC.
While checking performance on a VHF receiver recently I discovered that
the audio level across the audio spectrum varied a few dB depending on
the selected AGC speed. On fast, it rolled off highs a bit and as I was
making a curve of audio vs RF input to check out the AGC, the peak
frequency varied with RF input except on the slow AGC position. Things
are not perfect in Utopia here. The AGC time constants on medium and
fast were inside the audio spectrum. ALC time constants can be such a
problem, but in a digital radio attack and delay times can be set
arbitrarily to achieve truly fast attack and extremely long decay.
>
> Ron Castro
> Chief Technical Officer
> Results Radio, LLC
>
--
73, Jerry, K0CQ,
All content copyright Dr. Gerald N. Johnson, electrical engineer
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|