[Amps] Price per Watt Conversation

[Manfred Mornhinweg]

Jim,

> ##  I was afraid of this.   We are back to square one.   50%  eff and
>  lousy imd, or better IMD and lousy 35% eff..take your pick.

Yes. You see? That's why I think that we need to look forward, forget
conventional old-fashioned class AB linear amplifiers, and develop a
better, higher efficiency, cleaner concept. The broadband push-pull low
impedance class AB amplifiers, with all of their problems, have been the
workhorse since about 1975. That's over 40 years. Even the general move
from bipolar transistors to MOSFETs has changed very little about their
basic design and performance. It's high time that we move on to
something better. It's not a matter of doctoring around the old design 
trying to fix its unfixable shortcomings. It's about designing something 
new! And it's about not being afraid of doing something new!

> ##  You may well be able to put 2 x  50-60 vdc switching  supplies in
>  series to obtain your 100-120 vdc  for your proposed

And why on earth should I use two power supplies in series, instead of
using a single one of the proper voltage? I fail to see the point! Maybe 
if I had two 50-60V supplies lying around... But if I have to build or 
buy them, I can as well go for a single one of the proper rating! 100V 
switching supplies are slightly simpler and cheaper to build than 50V 
ones of the same power.

> ##  baffles me how SS  marine xcvrs all have superb IMD..and also run
> on 11-15 vdc.

The ones I have worked on are designed much like ham equipment, and have 
much the same basic performance. Except for one important difference: 
While ham transceivers are full of knobs to give hams something to play 
with, marine and other professional radios have as few knobs as 
possible, to simplify operation. After all they are intended to be used 
by largely radio-ignorant people, who want to get communication, not to 
play with radios. So the microphone gain, drive level, etc, are fixed, 
or internally adjusted at the factory, to those values that give a 
correct and clean signal when speaking quite loudly into the microphone. 
After all, most operators of such radios tend to shout, and the 
environment is often noisy.

Hams instead get to play with mic gain, compression level, drive level, 
and often they add non-original mikes having further level and 
compression adjustments. And most hams just LOVE to deflect the needles 
as far as they possibly can. The result is: All knobs fully to the 
right, resulting in tremendously excessive gain, with the ALC fighting 
it. Add a poorly designed ALC, and many are, and that's it.

Some hams go another step further, take the cover off their radios and 
defeat the ALC. That goes a good way toward deflecting the needles even 
further. And toward splattering all over the band and then some, of course.

Hams should be asked to pass a test, both about technical knowledge and 
mental sanity, before being allowed to operate radios that have external 
controls for mic gain and compression level. Those who haven't passed 
that test should only operate radios that don't have these controls, and 
only operate them with the stock mikes. And the covers of their radios 
should be welded shut, just in case! :-)


Joe,

just a small correction:

 > The PA must be over driven in order to produce any ALC!

That depends on how it's implemented. An ALC that requires overdriving 
is a badly designed ALC. In solid state transceivers the ALC signal is 
usually derived from a comparator that starts giving an output when 
either the power output reaches a certain level, or the reflected power 
does so. If this comparator is properly adjusted so that it starts 
giving an output a little before the PA is overdriven, then the ALC 
prevents overdriving. Of course if a monkey with a screwdriver fiddles 
with that internal adjustment and sets it above the overload level, the 
ALC can no longer do its job of preventing overdrive.

Another problem is that the ALC usually doesn't know what's the actual 
saturation threshold of the PA. And this level changes with the supply 
voltage. So, if at the factory the ALC threshold is adjusted so that at 
13.8V supply voltage it works fine and triggers just before overload, 
then when running that same radio at 12.0V (long cables, battery 
operation) the PA will saturate and splatter. That's a very real problem 
we can hear on the bands every day. Only radios with internal, good 
voltage regulation for the PA are free from this problem. All the 13.8V 
radios are not, and that's what most hams use.

And then, many ALC circuits are designed with inadequate time constants. 
These will produce a single burst of splatter at every strong voice 
attack, but not during continuous talking. The higher the operator sets 
the gain controls, the worse this gets. But only on the attacks.

Manfred

========================
Visit my hobby homepage!
http://ludens.cl
========================