Today, I made some incredible developments. The Trinity channel strip will have 3 vacuum tubes in it, each one representing a gain stage. Today I finally was successful in developing a working prototype of the first gain stage, in which my sound and performance requirements have been exceeded. The first stage offers 35dB of gain (with a gain knob for the input level). The noise floor of this stage at full volume is -110dB on the spectrum analyzer. It has a very flat response from DC-25kHz, which is the cutoff point of my analyzer. I'll use my o-scope soon to determine how far out the high frequencies extend before rolling off. 25kHz is pretty awesome as it is, but I feel like it probably goes way up. The THD is below 0.1% with a 1v 2kHz sine wave, but I can't give an exact number until I conduct more tests. Each stage will contribute some THD, so it needs to be low... although some THD is ok with me. My limit for what I feel sounds great in normal situations is 0.47% THD. To me, anything above half a percent is too colorful to be an all-around-player. Above .5% it becomes to me, a specific sound and color that is desirable for only certain situations.
This new design for Trinity is going to be very expensive and will probably take a considerable amount of time to work out, but I'm sure I can get it done in under a year! The new layout calls for 4 audio transformers in the signal path, twice as many as before. Its necessary for total isolation between the various circuits. The EQ circuit especially needs to be isolated for noise and stability reasons. I'm planning to go with a push/pull output stage this time around to minimize noise and THD, while providing plenty of volume. I experimented yesterday with using an interstage transformer as a phase inverter and it worked great. That's the plan for the output stage. The main thing I need to figure out is a low noise tube stage for the EQ circuitry. That stage mainly needs to provide about 20dB of low noise gain that's relatively clean. I would be cool with a higher amount of THD with the EQ on vs. bypassed. It would be easier to design and I think no one would complain about more tube tone. More to come!