View Single Post
  #25   Report Post  
Old January 7th 07, 10:11 AM posted to rec.radio.cb
Frank Gilliland Frank Gilliland is offline
external usenet poster
 
First recorded activity by RadioBanter: Jul 2006
Posts: 432
Default Brian's non-linear linear

On Sun, 7 Jan 2007 02:04:05 -0500, "Jimmie D"
wrote in
:

snip

I must admit I never built a single transistor PA before SO I will have to
look at the numbers again. Granted the push pull amp is more efficent than
the single ended amp



Not necessarily....


but Im still not sure about the gain. Il have to crank
the #s when I dont have a snoot full of cold medicine. Please correct me if
I am wrong but you are saying that two amps biased for say classB operation
that a 2 transistor anp will have 6 db more output than a 1 transistor amp
with the same drive power.



No, just 3dB. Double your pleasure, double your fun.


He probably does have an impedance matching problem ,



Yes he does. And his amp probably does, too.


almost all amps of
this type do to some degree. The impedance
transformation of the transformers can almost always be improved upon by the
inclusion of an adjustable L or Pi network in the output of the amp.. I know
the last amp I tested had an output impedance of about 35 ohms. A matching
network added to this amp increased power out, reduced harmonics and made
the transistors stop popping.



Impedance matching in an amp isn't a plug-n-play application. It's not
as simple as reading the specs off the data sheet and punching numbers
into a calculator. Since the big-signal gain of the transistors varies
so much, each transistor pair much be physically measured, matched,
and the impedance adjusted accordingly for maximum gain. The biggest
problem with these amps is that the transformers have so few turns
that they are very impedance-specific, as you discovered with your 35
ohm amp. There are only two solutions: either use transistors within a
very narrow range of gain -or- change the gain (and therefore the
impedance) of the transistors with negative feedback. I like the
latter because it increases linearity. It does require each amp to be
adjusted for any given pair of transistors, but almost all of the
transistors in stock can be used because the gain range isn't nearly
as limited.

Yes, you can transform the impedance with a tuning network prior to
the input transformer and after the output transformer, but it will
limit the frequency response of the amp to a narrow bandwidth. This is
how most of the so-called "broadband" CB amps are built. Most of the
people who have used them know what a bitch it is to retune the amp
every time you want to use it on a different part of the band. That's
not "broadband" -- it's false advertising.