[Amps] SS Amps are expensive...why?

[Manfred Mornhinweg]

Hi Roger,

> I honestly do appreciate the clarifications and although I did say 
> capable of 3 KW, I should have created the distinction between rated for 
> 1500W capable of 3KW.? Of course as a SWAG I'd say about 50 to 75? would 
> run an amp at what ever it's capable of doing

That's why manufacturers should put a lock on the internal ALC setting, 
or whatever other power control mechanism a transmitter or amplifier 
uses, and also add a self-destruction circuit that activates if some ham 
tampers with the power setting! ;-)

>> Add to this that it's mighty hard to find ways to correctly handle the
>> very low impedance of power transistors delivering 100W from a 13.8V
>> supply. As a result almost every of those final stages is only a very
>> poor approximation to what a true linear amplifier should be.

> At least going to 50 V helped that

Yes - but only while the power level stays the same. 100 watts at 50V 
gives about 13 times higher drain load impedance than 100W at 13.8V. 
Chokes and transformers need to stand the higher voltage, so they need 
to have 3.6 times larger cores, or 3.6 times more turns, and that makes 
100W 50V amplifiers only about 4 times easier to build than 100W 13.8V 
ones.

But 1500W at 50V gives even slightly lower drain impedance than 100W at 
13.8V, so that with the higher requirements on magnetics, such an 
amplifier is easily 5 times as hard to get right than a 100W 13.8V one. 
  What's really needed is 1500W devices that can be powered from roughly 
300V, along with better magnetic cores that can handle higher voltage 
per turn per size at acceptable loss.

I will keep dreaming.

>> All this can be improved quite dramatically, but hams these days buy
>> radios instead of building them, 
> 
> True, but today's rigs are getting mighty complex.

There is no need for homebrew rigs to have all the bells and whistles 
factory0made rigs have. Also the higher demands and thus higher 
complexity of modern equipment is offset by the wide availability of 
inexpensive, complex components. For example, these days you can build a 
complete frequency synthesizer covering zero to 300MHz in extremely 
small steps, using fewer parts than a 5-5.5MHz VFO took in the tube era! 
And it's probably less expensive too!

And you can build a complete high-performance transceiver by buying an 
FPGA board, adding a power amplifier and some filters and switching, and 
connect it to a computer running free software.

In that sense homebrewing has become simpler.

> OTOH "It's my 
> opinion" that modern rigs and particularly the top end rigs are giving 
> the user far too much control, allowing them to make a pretty decent rig 
> sound like a pile of "fertilizer"?

Very true - I hear such fertilizer every day on the bands. But then, 
hams are not supposed to transmit with broadcast quality! This is a 
hobby, and hams are supposed to have fun, and hopefully learn some 
things, playing with radios. Those modern radios with hundreds of 
user-settable adjustments, fulfill that purpose. Instead professional 
radios, made for people who just need to communicate, tend to have 
little more than a channel selector and a volume control on the front 
panel. And they work great.

>> Forget that. Solder down the transistors.
> Another plus for the new silver solder paste

It can be used, but common rosin-core solder wire works well, if 
correctly used. W6PQL has a video here that shows how it can be done:

https://www.w6pql.com/video/ldmos2copper.wmv

I prefer doing it without extra flux, but squeaky clean surfaces, and 
instead of moving the transistor after the solder melted, I prefer to 
just gently weigh it down during the process, so that excess solder 
squeezes out. I believe this leads to lower risk of having solder voids 
under the transistor.

> That was my goal too, but two strokes and one heart attack

Oh, sorry to hear that. I'm still a lot better off than that, but 
arthritis has made my fingers clumsy, and the autofocus system my eyes 
once had, has ceased to function. Now I need four different glasses and 
an assortment of magnifiers, and half of my work time is wasted 
switching glasses and handling the magnifiers. Oh, to be young again...

>>> Those large displays are often necessary to display many important
>>> functions in advanced SS amps.
>> Like which?? And are they really necessary?
> Depends on the amp and transistors. With the Quadra I'd appreciate more. 
> Particularly Emitter current and temp as well as drive. Some swear by 
> them and I swear at them.

:-)

I instead like a clean black box, with all user interface for the whole 
rig being concentrated on a single screen. For reasons of cost, 
programmability and resolution, this is usually the screen of a plain 
normal PC, rather than something built into equipment, and that's what 
turns radios into black boxes with a power switch.

>> Some rigs are expensive and over priced, but some bring new technology
>> with extremely clean signals.?
>>
>> Like which? I'm not aware of commercial ham equipment having extremely 
>> good IMD performance. I would like to know of any!
>>
> Anan has a group of transceivers and circuit boards from around 10W to 
> 200W using dynamic predistortion with better than -70 db IMD with "Clear 
> Signal" software https://apache-labs.com/1001/Ham-Radio-Products.html 
> and they use "open source" software for most functions.? Pricey but 
> still less than the competition and much cheaper in general than the big 
> three.

Ah, you meant those! I was thinking about mainstream radios, the variety 
with 50 to 100 knobs and buttons.

The Anan transceivers are implementations of the OpenHPSDR project, and 
I understand they are based on the Hermes board. So they sit in almost 
the same boat I do - only that instead of the Hermes, which I find too 
expensive for what it does, I use a Red Pitaya board, and I'm working on 
developing a high power, high efficiency TX amplifier chain for it. On 
the software side, what I use is the exact same used by the Anan 
transceivers.

My own trials using dynamic predistortion have given somewhat mixed 
results. Sometimes I have gotten sensationally low close-in IMD, at 
other times it doesn't seem to help at all, and in yet another cases I 
have seen it reducing the close-in IMD products while at the same time 
significantly worsening those IMD products that lie outside PureSignal's 
correction bandwidth. The result is a signal that's very clean up to 
+/-20kHz, and outside that bandwidth has an abnormally high level of 
"grass" on the spectrum display. I have yet to gain more experience with 
PureSignal to be able to tell much more. My current project is being 
developed so that the hardware is present to use the PureSignal 
predistortion software, but I expect to achive better signal quality 
than normal rigs even without using predistortion.

> Ability too 
> as my left side has limited mobility. That now takes me hours to do what 
> used to take minutes. I could touch type over 70 WPM, now it's one 
> finger typing.

We would complement each other in that regard. I can still type with 9 
fingers, although some are slow and cause more trouble than joy. The 
tenth one, the right index finger, is about as useful as a piece of a 
broomstick glued to my hand. With the difference that a broomstick 
wouldn't hurt all the time. Anyway, your single finger and my nine 
usable ones would combine into a usable set. ;-)

Manfred


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