Amps
[Top] [All Lists]

Re: [Amps] SS amps watercooling - was PowerGenius XL

To: amps@contesting.com
Subject: Re: [Amps] SS amps watercooling - was PowerGenius XL
From: Manfred Mornhinweg <manfred@ludens.cl>
Date: Wed, 22 Feb 2017 16:44:07 +0000
List-post: <amps@contesting.com">mailto:amps@contesting.com>
Jim,

## which broadcasters are using SSB ???

Only a few do, on short wave, most of them just outside ham bands. Most broadcasters use AM or FM, of course.

I do take your point. High efficiency amplification is trivially simple on FM, and on AM it's still simpler than on SSB. But high efficiency amplification of SSB signals is far from impossible, and when using modern technology it's actually practical. By "practical" I mean that the added circuit complexity is more than outweighed by the simpler cooling and lower power consumption, and that the added cost for additional small signal components is _far_ outweighed by the lower cost for the high power components, which have less demanding specs in high efficiency amps! That applies to the required power dissipation of the RF devices, the entire cooling setup, and the power supplies.

David,

Are you speaking of Class-E amps?

Indeed class E is one of the circuit configurations that can be used, and in various different ways.

A basic, bare-bones class E amplifier without any additions is only suitable for constant-amplitude modes, such as FM or RTTY. With a simple shaping circuit it could be used in CW too. But with either predistortion or with envelope elimination and restoration it can be used in SSB. The former approach is very simple these days, using a Software Defined Radio as the driver, and results in improved but not spectacular efficiency, while the latter is somewhat more complex and can produce nasty wideband noise if poorly implemented, but results in best efficiency.

http://www.classeradio.com/sokal2corrected.pdf

Yes, that's a good article on basic class-E amplification.

Here's a switching design from the late 80's
http://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/radcom/1990/02/page30/index.html

The second circuit on that page actually operates as class E amplifier, even if it's not explicitly mentioned. The term "class-E" became widely used only a relative short time ago, but class E amplifiers have been in widespread use for a long time. For example many VHF radio final stages actually operate in class E, even if it's said that they are running class C. As soon as you drive a class C amplifier into saturation, and this amplifier has the proper output configuration (enough capacitance in parallel with the transistor, and a matching circuit that presents a high impedance to harmonics) you get class-E operation.

Such an amplifier is the basic building block. It can be used directly for FM or RTTY, and stuff can be added around it to make it usable for SSB. That "stuff" is either a simple RF feedback to the driving SDR that implements predistortion, or it's a modulated power supply controlled by the SDR.

It's also possible in principle to take a standard drive signal from any 100W transceiver and split it into phase and envelope components, then time-delay the phase signal and drive the amplifier from it, while using the envelope signal to control the switching power supply. But this method is prone to various distortions that can't be easily corrected by analog circuitry, and result in a low quality transmitted signal. Because of the higher circuit complexity and the lower quality result, this method really isn't convenient these days, with the SDR-based methods being far better.

Fascinating ...

I agree. At this time I'm eagerly awaiting my SDR board, already ordered, to finally start working on my pet project: A high efficiency, legal limit Software Defined Radio!

Manfred

========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>