[Elecraft] ALC and the K3

[Bob McGraw K4TAX]

As I understand and was taught, the amplifier ALC came about as the 
result of exciters having the ability to excessively overdrive the 
linear amp.  Case and point, the Collins 32S1 transmitter would produce 
about 150 watts output when correctly tuned and loaded.  That was about 
twice the required power to drive the matching 30L1 amplifier.  Many 
ask, why not just turn the power down?  Well the 32S1 is a tube 
transmitter which has lowest IMD and best efficiency when tuned and 
loaded to the correct value. In order to prevent overdrive of the 30L1 
amp, the amp ALC signal was fed back to the 32S1 to reduce the power to 
the correct level via the intermediate stages , while the tuning of the 
exciter PA, 32S1, remained correct.  And thus today we do have amps 
which have the ability to provide ALC back to the exciter.

It is of note that many modern day transceivers have ALC circuits which 
do prevent overshoot at rated output.  Unfortunately some of these, when 
operated at less than rated output, have a faulty ALC system which does 
briefly allow ALC overshoot.   Using ALC from the amp back to the 
exciter has been found not to be successful in controlling this 
overshoot issue as the control element is too far down stream in the 
system.  Just look at the spectrum of many signals using an SDR receiver 
and spectrum display.  You'll observe a bit of splatter on the first 
word or syllable of many signals.  This is attributed to ALC overshoot 
in the respective transceiver causing brief over drive of the amplifier.

73

Bob, K4TAX



On 12/21/2017 3:13 PM, K9MA wrote:
> Sometime back around 1970, I built an amplifier for my college radio 
> club. I spent many, many hours trying to get ALC to work, and failed 
> completely.  Fortunately, the exciter really couldn't overdrive a pair 
> of 3-500Z's, so we ended up running it just as Elecraft recommends, 
> using only the exciter's internal ALC.  I'm sure amplifier designers 
> no know a lot more about it than I did back then, but apparently 
> closing the ALC feedback loop around an external amplifier is still 
> difficult to do.
>
> 73,
> Scott K9MA
>