I finally found time to track down my low power problem. I double checked all the coils and connections they seemed to be OK. There was one connection on T3 that seemed a little flaky, so I removed T3 and rewound it. I traced the TX signal path with the scope and made good use of the excellent information on Alan's site https://sites.google.com/site/g4zfqradio/levels-measured-in-a-softrock-tx.
I found that using the PTT test in CFGSR and the IQOUT program makes signal tracing reasonably easy.
Ultimately, I discovered that the voltages across R48 and R49 were not even close to spec (.79mv and 0 mv respectively). It turned out that R48 was open. Maybe it failed when I toasted Q3 during my earlier testing, I don't know. After replacing R48 I am getting a nice TX signal, and when I put the Pennywhistle amp in line I am getting a forward power reading of around 7-8 watts on the low power setting of the tuner into a dummy load. Not sure how accurate this is, but it is definitely progress.
I still plan to do some tweaking as the voltages across R48 and R49 seem low (around 34 mv). I'm going to experiment with matching the BS170's when I get time, or may play with different values of R41 as there seems to be some good information on these subjects available.
Thanks for the help everyone.
--- In firstname.lastname@example.org, "Alan" <alan4alan@...> wrote:
> ----- Original Message -----
> Subject: [softrock40] Ensemble RXTX Power Output
> > I am unsure that the RXTX is developing the rated amount of power though.
> >It appears to me that the RXTX is not developing full power. Am I mistaken
> >about that? Are my measurement techniques faulty? Could this be a sound
> >card issue?
> As has been said, most times this sort of issue can be resolved by careful
> inspection, especially of the coil connections. First you should be certain
> that everything on Robby's pages has been checked and that component values
> are correct.
> If you have a way of measuring levels, you do, the method is to go through
> the circuit methodically, starting at the audio levels and through the
> stages. Levels should increase at each amplifying stage.
> I did this and recorded my results here
> All RXTXs are much the same so provided your measurement device is capable
> at the frequency you are testing then you should get results something like
> mine. Driven harder most Softrocks will deliver more than 1 watt, perhaps a
> little less on 10/12 metres.
> 73 Alan G4ZFQ