GoodSoundClub - Romy the Cat's Audio Site

In the Forum: Melquiades Amplifier
In the Thread: Single-stage Melquiades vs. DHT amps

Post Subject: Preferred 2A3 operating pointPosted by floobydust on: 2/13/2009

Adrian, Romy, et al,

Okay, I don't take any of it as an attack... so none of that bothers me.... I'm fine with it... again, it's a view on what you heard based on your test scenario, nothing more and I certainly welcome the review. Now that I understand it... here's some of my (personal) viewpoints on this.

The Moondog design goes back a good 10 years now, is a straight-forward design and there's basically nothing wrong with it. If you do some modeling you're running about 15.5 watts of dissipation on the plate and pushing over 4.5 watts output. Running the higher B+ voltage (350 volts) with a 1K cathode bias resistor and the 2.5K load results in higher distortion than the text-book operating point, which is B+ of 300 volts and a 750 ohm cathode biasing resistor and the 2.5K load.

From my findings with the (NOS) 2A3 and higher voltages, going to a higher plate load impedance makes a large difference by lowering distortion. My basic 2A3 SET design uses a 3.5K load with a 866 ohm cathode bias resistor and a 350 volt B+ supply. This was far more optimal and provided much better (clean) low-frequency response and a noted reduction in measured distortion and better wave-form symmetry. I also tested with a 2.5K load and a 1K cathode bias resistor.... so at least with NOS 2A3 tubes, performance suffered.

I have a "gut feel" that the new manufacture 2A3 versions may perform better than the collection of NOS tubes I used for validating my design (literally over 50 of different brands) at the operating points of the Moondog. Again, a gut feel... nothing more. One last point to consider.... I've not seen any specifications for new tubes which state what the interelectrode capacitance values are. As pretty much all of the new tubes are NOT the same materials and/or internal construction, it's unlikely they are the same in these measurements. This could result in an audible difference based on the driver circuit and available slew rate and current flow. If I were to obtain a pair of KR 2A3 tubes (I'm thinking about them... but they are very pricey!) I would certainly go back and optimize the operating points for them. I had to do this for the EML 45 tube and it paid off in better performance (ad more power).

Regards, KM

Rerurn to Romy the Cat's Site