Greetings friends. I've put together a small power amplifier around a pair of PCBs sold as Dynaco 6V6 Push-Pull Amp:
This is my first Push-Pull build, and a bit of a Shoebox build, as I already had the power transformer. It has a pair of windings for HV and LV, so each channel has its own dedicated power supply:
I got it all hooked up and the voltages looked good, they're written in red in the diagram above. Hooked up a pair of speakers, and got nothing. Well, a tiny, tiny bit of very distorted signal out of the left channel when I open the volume pot all the way. There is a 100k Log pot on the input as a volume control I forgot to draw in there. Is that 10k input resistor too big? Is this thing meant to be fed a much larger input signal from a Preamp or such? I've no experience with the floating paraphrase - is the cathode bypass cap meant to be polarized?
I'm getting no hum or buzz at all, just a tiny bit of signal coming thru. And when I touch the meter probe to pins 1 and 6 on the preamp tubes, I get very clear taps coming thru the speakers. Voltages look right - what am I missing? should I bypass that pot on the frond end and see what I get, then reduce that input resistor to 1k or less?
thanks!
will
This is my first Push-Pull build, and a bit of a Shoebox build, as I already had the power transformer. It has a pair of windings for HV and LV, so each channel has its own dedicated power supply:
I got it all hooked up and the voltages looked good, they're written in red in the diagram above. Hooked up a pair of speakers, and got nothing. Well, a tiny, tiny bit of very distorted signal out of the left channel when I open the volume pot all the way. There is a 100k Log pot on the input as a volume control I forgot to draw in there. Is that 10k input resistor too big? Is this thing meant to be fed a much larger input signal from a Preamp or such? I've no experience with the floating paraphrase - is the cathode bypass cap meant to be polarized?
I'm getting no hum or buzz at all, just a tiny bit of signal coming thru. And when I touch the meter probe to pins 1 and 6 on the preamp tubes, I get very clear taps coming thru the speakers. Voltages look right - what am I missing? should I bypass that pot on the frond end and see what I get, then reduce that input resistor to 1k or less?
thanks!
will
Last edited:
Seems the only path to ground for input cathodes is via 1K feedback resistor and output transformer. Is the transformer secondary properly grounded?
Here is a link to another build: http://www.r-type.org/articles/art-351.htm
Here is a link to another build: http://www.r-type.org/articles/art-351.htm
I have seen this Dynaco 6v6 circuit many times (and have an unbuilt PCB of the same). What struck me as unusual is the 12ax7 cathode voltage at 5.5 V. Perhaps this is not useful, since you have experienced a “bingo”, but might offer some clues for improvement. BTW, what exactly was wrong? It sounds if the secondary was ungrounded, while @rayma talked about the schematic that did not show the input connector ground. If the secondary was unconnected the 5.5 V bias concern is perhaps not relevant.
At a negative bias of 5.5V the 12ax7 is practically in cutoff. I would have expected the cathode voltage to be around 1.0 - 1.5 volts. Since the combined cathode current for the two 12ax7 triodes travels to ground through the transformer secondary (low resistance) and the feedback resistor of 1 k, it means that the cathode current is around .25 ma per triode. I don’t know if this is expected.
Could a member who had built this circuit perhaps check what cathode voltage you get?
At a negative bias of 5.5V the 12ax7 is practically in cutoff. I would have expected the cathode voltage to be around 1.0 - 1.5 volts. Since the combined cathode current for the two 12ax7 triodes travels to ground through the transformer secondary (low resistance) and the feedback resistor of 1 k, it means that the cathode current is around .25 ma per triode. I don’t know if this is expected.
Could a member who had built this circuit perhaps check what cathode voltage you get?
Last edited:
I've not built it, but I liked its simplicity so I simulated with 320 V B+ and a simple 8k Raa output transformer with 40% UL taps.
The only modification I've applied is the feedback path taken from the 8 Ohm tap with a 1.2 kOhm resistor to bring the 12AX7 cathodes from 1.7 V to around 2.0 V as I usually do.
I get 1 Wrms with 0.07% THD, a DF of 7.6 and around 12 Wrms at clipping.
It is a nice simple circuit, this is a project for a Diy Audio club and I wanted something that first timers could get working. Maybe with a little help along the way.
Now that it's working properly, a few questions:
one PCB has both power tubes' plates at 293v, the other one measures 293v and 263v. I'm assuming this is due to mismatched tubes, and since the circuit is cathode biased, I just need to find a different tube?
also, there's a bit of hum, no prob there, but also an interesting, slow rising-and-falling whine. Since the 12AX7 grounds thru the Fb wire, can noise be coupled that way, and should I run the Fb wires to the corners of the chassis?
Now that it's working properly, a few questions:
one PCB has both power tubes' plates at 293v, the other one measures 293v and 263v. I'm assuming this is due to mismatched tubes, and since the circuit is cathode biased, I just need to find a different tube?
also, there's a bit of hum, no prob there, but also an interesting, slow rising-and-falling whine. Since the 12AX7 grounds thru the Fb wire, can noise be coupled that way, and should I run the Fb wires to the corners of the chassis?
I'm starting to think it was foolish to omit the 1000pF caps in the Fb circuit. I read somewhere they were only necessary to tame oscillations, so I didn't buy em. heh. Got some on the way, and I need to re-run the Fb wires. But not today, there's a kids' party to attend. Thanks for all the help!
w
w
Feedback amps are very likely to oscillate unless they are properly tuned. This is the problem with dropping a different output transformer into an existing feedback design. You really don't know what you'll get. The original Dynaco transformers were very well made and capable of enough bandwidth to reduce the chances of oscillation. The Edcors maybe not so much. That doesn't mean they won't work, but they'll need some attention.
Try the 1000pF cap and see if the amp stabilizes. Then you should learn to use the basic functions on that scope. (If I could do it, anyone can!) It will save a lot of trouble in the future, and also open up a whole new world in terms of having the freedom to build a feedback amp from scratch. What you want to look at is the high-frequency square wave response, usually at 10kHz. They make signal generator apps for phones. That would be a good place to start. If you show us the pics we can help tune that feedback cap to get a smoother response and reduce the chances of oscillation.
Over at Audiokarma there are some good threads about feedback tuning and what to look for.
https://audiokarma.org/forums/index.php?threads/oscilloscope-basics.842398/
https://audiokarma.org/forums/index.php?threads/regilding-the-gilded-lily-heaths-w-2m.767851/
Try the 1000pF cap and see if the amp stabilizes. Then you should learn to use the basic functions on that scope. (If I could do it, anyone can!) It will save a lot of trouble in the future, and also open up a whole new world in terms of having the freedom to build a feedback amp from scratch. What you want to look at is the high-frequency square wave response, usually at 10kHz. They make signal generator apps for phones. That would be a good place to start. If you show us the pics we can help tune that feedback cap to get a smoother response and reduce the chances of oscillation.
Over at Audiokarma there are some good threads about feedback tuning and what to look for.
https://audiokarma.org/forums/index.php?threads/oscilloscope-basics.842398/
https://audiokarma.org/forums/index.php?threads/regilding-the-gilded-lily-heaths-w-2m.767851/
Thanks, I wouldn't have thought the OPT plays such a crucial role in all this. Tuning the feedback loop - this will involve swapping different resistors and caps until I get correct readings on the 'scope? As I'm in a PCB my overall circuit choices are limited.
I'll look into Tuning Feedback Amplifiers while I wait for my delivery.
I'll look into Tuning Feedback Amplifiers while I wait for my delivery.
No, the feedback resistor value should stay the same to maintain the appropriate level of feedback. It's the phase-lead cap that will help to tame high-frequency ringing and prevent oscillation.
In a global-feedback amplifier, or one in which feedback is wrapped around the entire circuit, from the ouput back to the input, the quality of the output transformer is critical. It must have a wide-enough frequency response to avoid phase shifts at the frequency extremes that can lead to instability and oscillation. That's not to say the Edcor won't work--lots of folks have built feedback amps with Edcor transformers--but you need to take a look at it on a scope to make sure it's stable.
In a global-feedback amplifier, or one in which feedback is wrapped around the entire circuit, from the ouput back to the input, the quality of the output transformer is critical. It must have a wide-enough frequency response to avoid phase shifts at the frequency extremes that can lead to instability and oscillation. That's not to say the Edcor won't work--lots of folks have built feedback amps with Edcor transformers--but you need to take a look at it on a scope to make sure it's stable.