Hello Everyone,
I am in desperate need of the collective wisdom of this forum
Basically, over the last couple of weekends I put together an 833A based amplifier (actually, the tubes are an old russian GU-48 and a chinese 833C, but that will change). I had to use what I had on hand so the design is both non original and slightly unusual.
In short:
The driver is couple via a 0.47uf 630v cap to Pete Millet's A2 follower A2 Buffer powered by a +/- 150v rail
The 833A are powered by a 1200v rail (but this is part of the issue I am having)
As for bias on the 833As I basically rigged a voltage divider capable of delivering POSITIVE bias, from 0.2v to 150v. The reason for this is that Pete Millet's A2 follower is ungrounded and runs -4v. I wanted to experiment with
positive bias and go full A2 mode.
The good
The amplifier works and the sound is excellent. I hear a slight hum if I put my ear right to speaker despite running all filaments on DC. I guess there is room for improvement there and I will look into changing a couple of things.
The bad
Now comes the mystery. So I powered everything and at 220v mains I was reading good voltages on every rail. 1200v B+, +/-150v for the follower and +450v for the 12hg7. I was reading uneven bias on the 833A with the Chinese 833C running lower (90/100mA) than the GU-48 (130mA). This is not, however, the most troubling thing. What had me worried was the incredibly low gain of the amp. All the way up (source and volume knob) I could barely here sound from the speakers.
So I started to play around and tried something different: I set 110v on my mains (as you can imagine all my rails were cut in half) and played with the positive bias on the 833A to reach 120mA bias on both tubes (had to run +4v/+5v on the grids of 833A). My goodness, what a difference! Nothing that would get the cops knocking on my door but the sound was amazing, the clarity was there and so was gain. Ecstatic!
But then I went back to 1200v, 0v bias and gain disappeared. I scratched my head and tried to come up with possible answers albeit I suspect all incorrect:
- the A2 follower is not working properly (but I would get no sound at all at any bias level and distorted sound);
- the way I measure current draw of the 833A is unreliable. I am using chinese meters and measuring draw of current from 0v heater to ground. This is by definition not precise as I will see plate current and grid current as well, but reliable or not this doesn't seem the culprit;
- the driver is not working properly. Again this doesn't really make any sense since I get clear undistorted sound.
I am rather lost but I am sure you will all have excellent advice. I've been always impressed with the knowledge, wisdom and generosity of all its users.
Thanks!
I am in desperate need of the collective wisdom of this forum
Basically, over the last couple of weekends I put together an 833A based amplifier (actually, the tubes are an old russian GU-48 and a chinese 833C, but that will change). I had to use what I had on hand so the design is both non original and slightly unusual.
In short:
- 800VA transformer (power) custom wound
- 628SE Hammond 5k OPTs
- 12hg7, cathode bypassed, CCS loaded single driver (35mA load) powered from a B+ 450v, cathode resistor 47ohm (bypassed 50uf 50v) and 10kohm screen drop resistor. The driver is based on Pete Millet's driver schematic for his 813 amplifier (813 SE triode amps)
The driver is couple via a 0.47uf 630v cap to Pete Millet's A2 follower A2 Buffer powered by a +/- 150v rail
The 833A are powered by a 1200v rail (but this is part of the issue I am having)
As for bias on the 833As I basically rigged a voltage divider capable of delivering POSITIVE bias, from 0.2v to 150v. The reason for this is that Pete Millet's A2 follower is ungrounded and runs -4v. I wanted to experiment with
positive bias and go full A2 mode.
The good
The amplifier works and the sound is excellent. I hear a slight hum if I put my ear right to speaker despite running all filaments on DC. I guess there is room for improvement there and I will look into changing a couple of things.
The bad
Now comes the mystery. So I powered everything and at 220v mains I was reading good voltages on every rail. 1200v B+, +/-150v for the follower and +450v for the 12hg7. I was reading uneven bias on the 833A with the Chinese 833C running lower (90/100mA) than the GU-48 (130mA). This is not, however, the most troubling thing. What had me worried was the incredibly low gain of the amp. All the way up (source and volume knob) I could barely here sound from the speakers.
So I started to play around and tried something different: I set 110v on my mains (as you can imagine all my rails were cut in half) and played with the positive bias on the 833A to reach 120mA bias on both tubes (had to run +4v/+5v on the grids of 833A). My goodness, what a difference! Nothing that would get the cops knocking on my door but the sound was amazing, the clarity was there and so was gain. Ecstatic!
But then I went back to 1200v, 0v bias and gain disappeared. I scratched my head and tried to come up with possible answers albeit I suspect all incorrect:
- the A2 follower is not working properly (but I would get no sound at all at any bias level and distorted sound);
- the way I measure current draw of the 833A is unreliable. I am using chinese meters and measuring draw of current from 0v heater to ground. This is by definition not precise as I will see plate current and grid current as well, but reliable or not this doesn't seem the culprit;
- the driver is not working properly. Again this doesn't really make any sense since I get clear undistorted sound.
I am rather lost but I am sure you will all have excellent advice. I've been always impressed with the knowledge, wisdom and generosity of all its users.
Thanks!
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.