Phillip Jockell wrote in message ...

Check out this guys site. He builds everything from scratch and
shows schematics for exaclty what you asked about, including 3 transistor
SSB transceivers!!!

http://www.intio.or.jp/jf10zl/index.html


i have seen this site a number of times and spent some time reviewing
it in detail. his designs are pretty simple. and i would say, too
simple to be replicated and used.

for instance, he has a transceiver with 50mW output. while this might
be a good experiment, don't expect to work all the grids using this
one. similarly, his receivers (like the one SSB transceiver with three
transistors) might work across the lab bench. but it will hardly
suffice as the main rig in a station. in anycase, i think we are past
the point of measuring rig complexities with the number of active
devices. these days the active devices often cost less than some
passive components. I get decent RF low power transistors for less
about 3 cents each.

This week, i finished an 2M transceiver too (The schematics are still
being drawn out, i will post the circuit soon.) It uses (i just
counted) 19 transistors, two ICs and two diodes. The total cost of
material is still less than 10 dollars (at least in india). Quite
often, using active devices often reduces the cost rather than
increasing it. For instance, a phasing type of SSB transceiver cost
far less than a single collins mechanical filter. Yet, the performance
is as good if not better than a that of a filter type ssb generator
and all most all circuit complexity is at audio frequencies.

too often, we tend to look for the simplest (read least active device
count) system only to find that it doesn't work as advertised. It is
more likely that a slightly more complex circuit turns out to be more
stable and easier to get going that the least count winner.

easy replication of a design means that if you build it again, you
should expect the same performance. this usually means that you
extract a little less out of each stage than it is capable of leaving
some head-room for device variation and component value tolerance.
That is the reason why many of us here like to reuse certain blocks
(like broad-band feedback amplifiers) over and over again.

finally, i would say, that it is not enough (at least for me
personally) to build something that makes a contact across the city.
one should aspire for the best quality in terms of whatever your
personal design goals are. Often, unrestricted by the need to include
all the features (like multiband, PLL, etc.) and with the ability to
use one off components, the amateur can spend more effort in building
and aligning his rig that a commercial counterpart. That is a goal
worth acheiving. it is not enough to match a commercial rig in terms
of performance, it is just a moderate challenge to build something
that that surpasses it by your own criterion.