On 9/26/2016 7:47 AM, jimlux wrote:
This is sort of an academic discussion - but one of some importance in
the long run as more sophisticated radio systems start to be built. The
idea of amplifier per element hasn't got a lot of traction yet, but it's
coming. - SW broadcasters do it, why not hams. Particularly with solid
state devices, the losses in a combining network in a SSPA with a single
output jack are substantial, particularly with microwaves. Spatial
combining (which has essentially no loss) is pretty attractive, when
you're ganging up 20 100 watt modules to get 1500W out. Why burn the
extra 5 modules just to overcome the combiner loss.
There are many reasons why this does not make sense for HF ham
radio arrays (at least above the QRP level). (I will leave aside the
question of how to measure power for legal purposes).
1. Combining networks in SSPA's are only difficult if you
want to cover 1.8 to 54 MHz in one combiner. If you are dedicated
to just one band, or maybe several low bands, it gets fairly easy.
In any event, we know that substantial losses are not present else
the combiner would burn up.
2. In steerable arrays (4 square, various "N-circle" arrays), it
is well known that the the impedance and power level varies drastically
from element to element. Typically, the rear elements have negative
power. This would simply be a nightmare to implement with an
amplifier per element. Amplifiers want to work into a 50 ohm
load and put out a fixed amount of power. Anything else results
in a drastic loss of efficiency.
3. You need to force current, not force power.
etc. etc.
The architecture that works in a practical sense is:
Solid state amplifier that always sees a 50 ohm load
and always puts out 1500 watts followed by an antennna
tuner followed by the usual 1/4 wave current forcing networks.
Rick N6RK
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|