"Question- If the goal is to make a full solid state replica of the RCA, are we still talking big antenna coils and inter-stage audio tx's"..
IMO, the goal should be to make a REAL solid state replica - as in, something which sounds and behaves as closely to the RCA as is achievable with easily available components.
Audio transformers are a potential problem - the transformers will colour the sound, and obtaining 'sound alike' parts - probably impossible.
What we really need to get a true clone, is to have a competent engineer with access to an RCA, and for signals on the plate of the mixer, the grid of the preamplifier, and the grid of the power amplifier, to be accurately recorded over the whole pitch range - We also need the volume changed as each of these signals are recorded, so as to be able to determine the harmonic variations (if any) resuting from pitch and volume variations, and by comparing signals at these different test points, determine the audio transformers charactaristics and 'coloration' attributes.
But obtaining the above data is most unlikely.
We know that there must be a roll-off quite a way down below 172kHz, caused by the tubes and / or transformers, as there is no actual filter in the circuit to remove thye HF components from the mixer - I suspect T7 alone removes the bulk, if not all of the HF..
I am inclined to think that most of the tonal charactaristics will be determined by the mixer and its input waveforms.. And I am inclined to think that, if we get this section right, we can tailor the harmonics using non inductive mechanisms - as in, standard filters or perhaps a more complex graphic or parametric equaliser circuit.
The volume circuit (as in, the antenna / detector) seems to me, at this time, to be non-critical.. But I am not going to assume anything, and will analyse this - However, at the present time I think that this does not need to tightly follow the original design.. I am happy to evaluate the existing design, but I cannot participate in the design of a solid state replacement.
I think that a H11F1 opto isolated FET would probably be the ideal VCA, having wonderful 'tube-like' qualities.
We probably cannot get samples from inside the RCA - But at least we need samples of the RCA output signals, so that we can compare our output against these - Once the mixer has been built, there will be the (probably long) job of fine tuning the amplitudes of the signals going into the mixer, possibly tweeking the oscillators, and then possibly developing complex post-mixer equalisation - and all this will need to be done while actually listening to RCA samples and comparing these, both by ear and with spectrum analysis, against what the clone is producing..
It aint over until the tiny fat lady really sings!
Fred.
" I have also always been curious about the interaction (if any) between the 2 windings on the (lg) volume antenna coil. "
I havent studied this section yet - it looks even wierder than the pitch oscillator did before I understood it - but it is a similar configuration, except there are both series and parrallel components .. No, I dont yet understand it, sorry - T3 3-4 looks like some sort of rectification / integration using C13 - but I dont understand why C13 doesnt add so much capacitance through T3 to the antenna, that the antenna becomes useless - Nope - That circuit aint clicking.. needs perhaps an hour or even a simulation (wish I had a good triode model..)