Well it's done and he is burning it in now. I'll have it by the middle of next week. My tech is smart guy an engineer and designer. He has lots of satisfied customers. Is it that outrageous to think this design can be improved upon? I'm not trying to start trouble here I just want to share what I'm doing with people that love music and fine audio. I will be sure to post my impressions after a day or so with the amp.
Here is an email from my tech:
This is the build details.
The stock power supply with factory upgrades produces 3 volts AC ripple. (I use software simulators). My redesign produces .3 volts AC ripple. Downstream the ripple is better than .05 microvolts. No AC ripple to mix with the audio causing minute, but detectable intermodulation products (IMD).
The stock power supply input filter is at the limit of ripple current. My input filter can carry 5x the ripple current. Cpacitors should last 40 years.
The stock power supply with the upgrade capacitors on the preamp board at 385 volts using 400 volt capacitors. I modd'ed the input board voltage from the main power supply with 430 volts and used 450 volt capacitors. The extra voltage is needed to bring the 12BH7s into the more linear part of the plate curves that equals better sound.
The phase splitter 12BH7s are re-biased from 4.5ma to 7ma each further bringing the 12BH7s into a more linear parts of the plate curves. Simply, standard operating voltages & current per the tube manufacture (attention to detail).
The bias supply produced .3 volts AC ripple. I added a series resistor to reduce the bias AC ripple to .03 volts ripple.
The stock coupling caps were K42Y-9 and a 'run of the mill' unidentified film cap. My mod uses K40Y-9 capacitors throughout.
The wiring routing was performed in such a matter as to reduce coupling from other wires that could cause some degrading of the sound.
About my rectifier:
Permanent- SS only. A GZ34 rectifier cannot handle the current of a ST-120. That rectifier tube is pressed as is with a ST-70 amplifier. Not only is the rectifier socket not wired for the 5 volts as (why stress the 5 volt winding with 535 volts DC riding on it), a tube rectifier will not sound better nor produce the peak watt output of the ST-120 capacity. The only GZ34 that has a chance is old Mullard GZ34 at huge cost. The Sovtek GZ34 would fail quickly. Perhaps immediately.
250ma maximum current with power transformers having a secondary winding of 125 ohms (peak current limiting by design)
60uF maximum capacitor size (first filter)
Dynaco ST-120 secondary winding of approximately 50 ohms violates the GZ34 250ma current rating reducing it to 160ma maximum in a ST-120
250ma current (average)
60uF first filter (limited by a rectifier tube)
155uF second filter (with upgrade option)
227uf first filter. I installed a super filter. Same as Jim McShane performs in the H&K Citation II.
110uF second filter
Result- no tube allowed.
The new SS rectifiers are quiet (no switching noise) and the same as tube rectifiers that are quiet. That is, a quiet power supply does not add IMD products in the sound. SS rectifiers have no series resistance (unlike tube rectifiers) and provide much higher peak to average current, This how a 60 watt H&K Citation II makes over 100 watts peak output.