[Amps] SS amps and auto-tune. Is it even necessary?

[Manfred Mornhinweg]

Alex,

> I think the ARF15XX line has 300 V devices.at the kW level involved,
> the output impedance is disgustingly close to 50 ohms.

Do you mean to say "delightfully close"? I would see it that way - if 
only those transistors had low enough capacitances and/or would work 
well enough in linear mode.

> Actually when
> designing for high VWSR operation, you have to take into
> consideration the TOTAL power involved complete with the reflected
> power from the antenna? . The amp doesn't care whether it receives
> power from the power supply, or from antenna.reflection. It has to
> dissipate it, besides generating fresh RF to pump into the antenna. 
> It turns out that an amplifier able to handle a 5:1 swr and still put
> out 1 kW has to be able to output about four kW.? And that goes for
> everything: power supply, cooling, antenna matching of whatever
> type.....

No, no, no...!  You got that wrong. The power dissipated in the 
amplifier stage (mostly in the active devices) is equal to the amount 
the power supply is putting out, minus the amount that the antenna is 
effectively accepting. And not any sum of powers.

A high SWR indeed means that the amplifier components have to handle a 
higher _reactive_ power, but _not_ necessarily a higher real power. And 
the power supply definitely only needs to supply the power delivered to 
the antenna plus the power being dissipated.

If an amp has to put 1kW into a 5:1 SWR load, it needs to put out a real 
power of 1kW, no more. It may need to handle more reactive power, but 
not necessarily. It depends on the actual impedance offered by that 
antenna/feedline combination. Then the design of the amplifier, with all 
of its matching, comes into play, to determine how many of the 
components need to handle the additional current, voltage, or both. 
Usually that will be some relays, measuring circuits, and the tuner or 
matching section. In a typical amplifier designed to operate into such 
poor loads, there will be some sort of tuning/matching circuit that 
takes care of it, and the amplifier stages proper will run just the same 
way, regardless of how high the SWR is, and what the load impedance is.

Thinking about a tuner-less amplifier intended for driving poor loads 
directly, it would need to be a switchmode amplifier that can handle 
certain variations of voltage, current and load phase angles without a 
big increase in losses. Plain common motor controllers (called "drives" 
in the industry) routinely do this up to a frequency of a few hundred 
Hz, and achieve efficiencies close to 100% over a very wide range of 
motor impedances. There is no fundamental reason why it couldn't be done 
at RF. Only practical reasons, like the unavailability at this time of 
suitable active devices...

An amplifier intended for operation into a poor SWR, that needs 5kW of 
DC to deliver 1kW into a 5:1 load, would be a mighty bad design!

Manfred


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