Tweet
I'm was wondering about how to bring this up as it has its roots in the silvertone 1474 - 6U8A reverb drive/recovery project here:
https://music-electronics-forum.com/...ad.php?t=47164
but, you don't really need all the history for the basic question I have and answers I'm looking for. Bottom line is this:
from what I understand, even at full drive, the signal voltages at the reverb tank output transducer is well below 10mV and somewhere around 5mV. The triode section of the 6U8A which I'm using as the recovery amplifier has a mu of 40 and I'm able to get a max gain of about 35(≈31dB). So, optimistically at full drive I should get a 175mV wet signal from the recovery amp. Ideally, I would probably cascade the stage into another gain stage to give me more to work with, but I don't really have one to work with. At least without really changing the function and operation of the front end of the amp completely.
So, my thought was to maybe take advantage of using a JFET/triode hybrid cascode to provide a higher gain out of the single recovery stage with the smallest footprint. I know how to design a tube cascode, but I don't have much experience (any, actually) designing with FETS. So, I'm looking for a good primer and maybe some suggestions on what JFETs would be best suited?
Another thought was to use use this idea in the 12AX7 gain stage before the Cathodyne phase inverter. This is the mixing stage for both preamp channels and I find the 330k mixing resistors heavily load each stage when adjusting the unused channel's volume/tone controls. This causes quite a bit of interaction between channels, and I think by eliminating the miller capacitance of the triode stage, you can increase the series resistance creating more isolation from interstage loading without the fear of the rolling off significant high end. If enough gain were achievable here, perhaps sacrificing a bit and implementing some local inverting feedback would further improve the mixing stage. I would be interested to hear some opinions on this. But I'm getting ahead of myself. First things first, educate myself on designing a hybrid cascode!
Here are the curves for the triode section of a 6U8A and a schematic of the circuit I'm working with:
okay.. It wont let me upload the schematic jpeg for some reason, so I'll add it to the following post
https://music-electronics-forum.com/...ad.php?t=47164
but, you don't really need all the history for the basic question I have and answers I'm looking for. Bottom line is this:
from what I understand, even at full drive, the signal voltages at the reverb tank output transducer is well below 10mV and somewhere around 5mV. The triode section of the 6U8A which I'm using as the recovery amplifier has a mu of 40 and I'm able to get a max gain of about 35(≈31dB). So, optimistically at full drive I should get a 175mV wet signal from the recovery amp. Ideally, I would probably cascade the stage into another gain stage to give me more to work with, but I don't really have one to work with. At least without really changing the function and operation of the front end of the amp completely.
So, my thought was to maybe take advantage of using a JFET/triode hybrid cascode to provide a higher gain out of the single recovery stage with the smallest footprint. I know how to design a tube cascode, but I don't have much experience (any, actually) designing with FETS. So, I'm looking for a good primer and maybe some suggestions on what JFETs would be best suited?
Another thought was to use use this idea in the 12AX7 gain stage before the Cathodyne phase inverter. This is the mixing stage for both preamp channels and I find the 330k mixing resistors heavily load each stage when adjusting the unused channel's volume/tone controls. This causes quite a bit of interaction between channels, and I think by eliminating the miller capacitance of the triode stage, you can increase the series resistance creating more isolation from interstage loading without the fear of the rolling off significant high end. If enough gain were achievable here, perhaps sacrificing a bit and implementing some local inverting feedback would further improve the mixing stage. I would be interested to hear some opinions on this. But I'm getting ahead of myself. First things first, educate myself on designing a hybrid cascode!
Here are the curves for the triode section of a 6U8A and a schematic of the circuit I'm working with:
okay.. It wont let me upload the schematic jpeg for some reason, so I'll add it to the following post
Comment