I recently finished this Tubelab SPP and have a question about feedback.
https://www.diyaudio.com/forums/tubelab/157491-pictures-tubelab-amp-59.html#post6855247
The output transformers are 8K primary to 4/8/16.
The feedback is taken from the 16ohm secondary tap going through 5.1K resistor and 240pF compensation capacitor.
Using either 4/8 taps for output.
Is there a simple way to measure how much feedback is applied ?
The input sensitivity is currently 0.455V input for about 9.2V output (10.5W) which is right before clipping.
I wouldn't mind increasing the feedback by 6dB so that input sensitivity is closer to 1V (is that a correct assumption?), but I would want to know how much is being applied currently before making changes.
https://www.diyaudio.com/forums/tubelab/157491-pictures-tubelab-amp-59.html#post6855247
The output transformers are 8K primary to 4/8/16.
The feedback is taken from the 16ohm secondary tap going through 5.1K resistor and 240pF compensation capacitor.
Using either 4/8 taps for output.
Is there a simple way to measure how much feedback is applied ?
The input sensitivity is currently 0.455V input for about 9.2V output (10.5W) which is right before clipping.
I wouldn't mind increasing the feedback by 6dB so that input sensitivity is closer to 1V (is that a correct assumption?), but I would want to know how much is being applied currently before making changes.
Interested in your responses!
I had real problems in how to determine this theoretically in my EL86 version. In the end I fitted a 100K pot and tried to tune it by ear, but I think the value I have (around 50k) is too conservative. I am away from home and I miss my oscilloscope, which would have introduced a modicum of science into the process.
I see examples where the cathode resistor of the driver tube is split in some ratio, and half is bypassed, but that could be the part including the feedback, or not. I wish I understood the rationales behind these design decisions better.
I had real problems in how to determine this theoretically in my EL86 version. In the end I fitted a 100K pot and tried to tune it by ear, but I think the value I have (around 50k) is too conservative. I am away from home and I miss my oscilloscope, which would have introduced a modicum of science into the process.
I see examples where the cathode resistor of the driver tube is split in some ratio, and half is bypassed, but that could be the part including the feedback, or not. I wish I understood the rationales behind these design decisions better.