John Byrns wrote:
In article , Stewart Pinkerton
wrote:
On Sun, 13 Mar 2005 01:57:31 -0500, Jon Yaeger
wrote:
in article , John Byrns
at wrote on 3/12/05 8:17 PM:
I have scored the three current entrants in the KISS amp design
competition
The competition to see if a reasonable solid state facsimile of Andre's
300B KISS amp could be designed.
What 'competition' is this?
And who elected you judge?
Everyone else took a step backwards, so I was elected. Somebody had to
compare the various designs, your comments/criticisms on my judging are
welcome.
For scoring purposes I considered NFB to exist only if two or more stages
were enclosed in a NFB loop, local NFB as in a triode tube or transistor
emitter follower, was not counted as NFB. Stewart's KISASS design failed
the NFB test because two stages, Tr1 & Tr2 were enclosed in the single NFB
loop formed by R7.
That is incorrect.
How so? Are you saying that my judging criterion relative to NFB around
two or more stages is incorrect, or that your KISASS design doesn't have
NFB around two stages?
Tr1 & Tr2 constitute a two stage amplifier with the load consisting of the
150 Ohm resistor R5 paralleled with the input of the output stage. The
load current flows through the 24 Ohm resistor R7 in the emitter circuit
of Tr2, R7 is also in series with the input signal driving the base
emitter circuit of Tr1. This constitutes and obvious case of a NFB loop
encompassing more than a single stage.
There is 70 dB of series voltage NFB in the output stage
from the follower connection alone.
The NFB in the output stage is not 100% effective because
part of the output load current current appears in the bases of the output
transistors, and thus this current is subject to the series current NFB in the
driver stage.
Its not as much as i first assumed though, and really, very
little NFB is applied over more than 1 stage.
I'd still prefer to build a mosfet amp with an OPT
if i wanted simplicity due to a small number of devices.
Patrick Turner.
Regards,
John Byrns
Surf my web pages at, http://users.rcn.com/jbyrns/