GoodSoundClub - Romy the Cat's Audio Site

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

Post Subject: 2A3 voltages, loads and realized output powerPosted by floobydust on: 2/14/2009

Adrian, Romy, (Dima),

First, I most certainly agree that the textbook operating point for the 2A3 is not optimal and can be improved on, hence my own design from a few years ago. If you are sticking with a 2.5K load, you will simply not get the most from the 2A3, distortion will be higher and waveform symmetry will suffer. This is validated by hundreds of hours of research/testing which included calculated load lines, software simulations and most importantly, bench time with over 50 different NOS 2A3 tubes using different load impedances, plate voltages and bias currents. Measurements were done using a Fluke digital meter (for DC voltage measurements), Tek 2215A dual-trace scopes, HP 400FL True RMS meters and a 334A Distortion analyzer with a clean sine/square oscillator.

From all of my research on this tube, I feel quite comfortable stating that 3.5K is the optimum load. Going lower increases distortion and going higher results in a loss of power with no tangible benefit. At 275 volts and a 4.1K load (15-watt Pd) you'll be lucky to get much more than 3-watts. You simply don't have the voltage to swing. I prefer ~290 volts on the plate (plate to cathode measurement). I also run the plate dissipation up a bit higher to 17-watts. Note: I don't see this as any sort of risk provided you have adequate ventilation around the tube. What is critical is NOT to exceed the rated cathode current of 60ma, as this will decrease the usable lifespan of the tube.

In my existing (implemented) design, I have a 350-volt plate supply, a 3.5K primary OPT, and an 866 ohm cathode bias resistor (bypassed). Pd is 17-watts, cathode bias current is 58ma and it puts out a solid 4.5-watts with good NOS tubes before any clipping occurs, and the clipping is symmetrical. The performance is also very consistent with the majority of the 2A3 tubes I have. There are of course some NOS tubes which are not so good... I even have two with grid leak problems despite using a 249K grid resistor.

Still, I'd be very interested to hear what Dima finds in his testing.... and of course which tubes he winds up using for his tests.

Regards, KM

Rerurn to Romy the Cat's Site