Hi Stefan,
Class E may be efficient but, as you already noted it has some drawbacks.
One you didn't mention is the rather large voltage swings on the Drain
(typically 4X VDD) which provides less than optimum (price/power) utilization
of the active device.
Driving either a Class D or E "linear" amplifier (employing either EER or
Polar Loop) from a typical (100W) transceiver is problematic as the phase
portion of the signal needs to be at a constant level.
I've been tinkering together a variable mode design employing a tube that is
compatible with a standard XCVR. It operates in Class A on low level signals.
Then, it transitions, as the input level is increased, through Class AB, Class
B and finally deeply into Class C with envelope restoration at the highest
power level. Device dissipation thus remains more or less constant over the
operating range. The plate supply is a switching supply which varies from
1-6KV. The screen voltage and grid bias are also dynamically adjusted in
concert. The operating point (i.e. supply voltages) need only vary at the rate
of the envelope of the input signal: 3 KHz. The delay in the plate supply is
minimized by means of a dual loop scheme and operating the main switching
regulator at 1 MHz with a 100KHz LPF.
This methodology might be adapted to a (typical push-pull) transistor design
with the addition of suitable (gate) protection. One can allow excursions on a
grid, that would prove catastrophic on a gate. It is also easier to drive a
swamped grid over a wide operating range than MOSFET gates with their large
(and varying!) Drain-Gate feedback C.
73 & Good afternoon,
Marv WC6W
**********************************
-- "Stefan" <ea5fy@ure.es> wrote:
Dear friends
Let me share a different point of view to Class E. I come from the Power
Supply business (switching power supplies), where efficiency is one of the
main design goal (but not always). Please forgive me for all bad
appreciations due to the lack of knowledge on RF PA.
- Efficiency is not only important for the energy saving but it also reduces
mass and volume (less heat sink). It should be indeed the main goal for all
these portable TX/RX (FT817).
- Class E is also a one band amp, but it would be nice if we could change
the tuning elements to adjust them for the different bands (I know, that is
not very easy in this kind of amps).
- I already had the feeling that driving this amps with the output signal of
a normal transceiver is not very easy (impossible?). I have no idea about
this subject, sorry. But I see that the discussion goes into this direction.
- Class E is a step forward of Class D (PWM). Every PS engineer knows that
with MOSFET the switching losses increase with frequency and soft switching
has to be provided to decrease these switching losses. Class E provides this
soft switching means. In the PS business higher frequencies were always
looked after to decrease the size of the reactive elements (L, C and
trafos).
I hope my comments are not completely wrong and I will be looking forward to
see the how the driving discussion goes on. I would appreciate some
solutions on driving this amps directly from a transceiver (please, if
somebody could send to me the references, I would apreciate it very much).
Kind regards,
Stefan
EA5FY
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
___________________________________________________________________
Speed up your surfing with Juno SpeedBand.
Now includes pop-up blocker!
Only $14.95/month -visit http://www.juno.com/surf to sign up today!
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|