Can regulated B+ act like a CCS?

[dgta]

... So in the '50's this was put to the test. As with most things involving perceived sonic performance, the test was inconclusive. Some preferred the performance with bypassed cathode resistors, others liked it without.

Was it put to the test (as in scientific measurement) or subjected to an opinion poll? I would hardly call it a "test" if the result was 'some preferred" and "others liked"

I believe what SY did was a real distortion comparison test, not a subjective listening session. So was there such a test documented in the 50s?

 

[SY]

Ran the test again last night (yes, measurement of the distortion spectrum) using an ECC83 and a cascode bipolar CCS, cathode resistor bypassed and unbypassed, 0.8mA, 1k5 cathode resistor. I want to repeat it one more time, then post the results to see if someone can find an obvious mistake.

Like I said, this odd result came as a surprise to me. If I made an error, I'll learn something from the correction. If I didn't, we have a fun puzzle. I *think* there's a clue in the distortion spectrum, but... more later.

 

[thermo]

Winegamd, Don't forget, using an LED in the cathode of a preamp tube gives you a constant VOLTAGE source, not CC. I use them in my custom designed tube preamp and found that for some reason, Orange, Clear LED's give the lowest distortion (using a spectrum analyzer).

 

[dgta]

...for some reason, Orange, Clear LED's give the lowest distortion (using a spectrum analyzer).

How did you compare orange and clear with other colors? Typically different color = different voltage.

 

[thermo]

dgta, I used a 6DJ8, set up for 20dB of gain, first using a 47u cap in para. with an 820 Ohm to Gnd. Then ran a 1v, 1kHz sine wave into it and took a spectrum graph with True RTA, 24th oct.
Repeated same test using a white LED in the cathode. The results were bad, too much
3rd harmonic. The Blue LED did better. But the Orange LED (with 1n4148 in series to make up the voltage drop) added extra 2nd harmonic. The octave. Which I prefer, especially in Push-Pull amps. that strip some of the 2nd harm. out in the output transformer.
Try in in a tube preamp, just make sure to match the same DC voltage to maintain proper bias.

 

[winegamd]

What is the voltage drop for the clear orange LED?

 

[dgta]

dgta, I used a 6DJ8, set up for 20dB of gain, first using a 47u cap in para. with an 820 Ohm to Gnd. Then ran a 1v, 1kHz sine wave into it and took a spectrum graph with True RTA, 24th oct.
Repeated same test using a white LED in the cathode. The results were bad, too much
3rd harmonic. The Blue LED did better. But the Orange LED (with 1n4148 in series to make up the voltage drop) added extra 2nd harmonic. The octave. Which I prefer, especially in Push-Pull amps. that strip some of the 2nd harm. out in the output transformer.
Try in in a tube preamp, just make sure to match the same DC voltage to maintain proper bias.

What was the cathode voltage for the C/R, white, blue and orange? Did you make all 4 the same? How? Can you post the circuit diagram so we can see what we're talking about? Very curious about this.