So I couldnt like wait a day for the proper tubes to arrive for a vari-mu circuit, so I thought I'd just jam a big negative transient down 12AX7's tailed pair throat, in that "cold clipper" position. 12AT7 and 12AV7 worked a little better, to where you could at least hear some compression really happening from the grid reference ties being driven negative by the envelope voltage. Which turns out to be the interstage potentiometer's ground connection now tied instead to the envelope voltage control.
Same DC envelope circuit essentially as Fairchild, only I'm using a spare 70V winding on the opt my amp happens to have to drive the full wave bridge. It only takes about 1/5 of the voltage that can make, so I'd have some confidence in a 16 Ohm winding being able to work also (25V) to generate a negative going envelope for upstream diff pair tube gain control.
I bought some 150V Zeners on the 12th still not here, probably lost in the mail. Re-bought different ones from a different vendor...
Same DC envelope circuit essentially as Fairchild, only I'm using a spare 70V winding on the opt my amp happens to have to drive the full wave bridge. It only takes about 1/5 of the voltage that can make, so I'd have some confidence in a 16 Ohm winding being able to work also (25V) to generate a negative going envelope for upstream diff pair tube gain control.
I bought some 150V Zeners on the 12th still not here, probably lost in the mail. Re-bought different ones from a different vendor...
After experimenting today, I've come to the conclusion that even a tube made for vari-mu or AGC circuits cannot work in an overdriven situation. You do get some gain variation, but not nearly enough to be effective as a compressor. Any such tube must go into the first position in the amplifier gain-chain, well, as depicted in all the compressor circuits I've seen and those I've posted here. A day wasted trying to put it in the cold-clipper position, aside from learning it cant go there.
Before I try a vari-mu tube as my input phase splitter, I'm going to play around with the Vactrol method of attenuation again. I also obtained some H11F1 parts; the LED - photoFET optoisolator. The FET part breaks at just 30V, while the photoresistor in the Vactrol is a bit more forgiving when it comes to the voltage you can put across it, in the high Z state. We'll see if I can make use of the H11F1 parts as well, considering solutions to their relative fragility wrt the voltage levels I'm seeing in these tube stages. Maybe, there's no way...
The cold clipper makes a nice "full wave rectified" looking signal across its 10K common cathode resistor and I'd like to use it to generate the DC control envelope for the LED part of these opto-isolators. One aspect I'd like to retain is having the effect not be dependent on an output to the speaker, as it would be when using a secondary OPT winding as a signal source. I noticed it doesnt take much current to the Vactrol LED to attenuate all the way to zero - heck just the DMM's diode test current will do it.
I have a gang of 3, 25 Ohm pots I pulled from some home theater passive surround box I snagged at the thrift store. I've wired all of them in parallel to make an ~8 Ohm resistive load, then connect the speaker to the wiper. Turned toward the ground side, it beats listening to my failing experiments with the speaker connected straight to the amps output. I'm going to have to find a place for that to mount permanently in my cabinet, along with a switch to disable the loading and let the speaker instead have 0 - 8 Ohms in series. It'll be great to simply reach under the bench and turn it down to a tolerable SPL, with the amp still running full blast and "properly" loaded.
Before I try a vari-mu tube as my input phase splitter, I'm going to play around with the Vactrol method of attenuation again. I also obtained some H11F1 parts; the LED - photoFET optoisolator. The FET part breaks at just 30V, while the photoresistor in the Vactrol is a bit more forgiving when it comes to the voltage you can put across it, in the high Z state. We'll see if I can make use of the H11F1 parts as well, considering solutions to their relative fragility wrt the voltage levels I'm seeing in these tube stages. Maybe, there's no way...
The cold clipper makes a nice "full wave rectified" looking signal across its 10K common cathode resistor and I'd like to use it to generate the DC control envelope for the LED part of these opto-isolators. One aspect I'd like to retain is having the effect not be dependent on an output to the speaker, as it would be when using a secondary OPT winding as a signal source. I noticed it doesnt take much current to the Vactrol LED to attenuate all the way to zero - heck just the DMM's diode test current will do it.
I have a gang of 3, 25 Ohm pots I pulled from some home theater passive surround box I snagged at the thrift store. I've wired all of them in parallel to make an ~8 Ohm resistive load, then connect the speaker to the wiper. Turned toward the ground side, it beats listening to my failing experiments with the speaker connected straight to the amps output. I'm going to have to find a place for that to mount permanently in my cabinet, along with a switch to disable the loading and let the speaker instead have 0 - 8 Ohms in series. It'll be great to simply reach under the bench and turn it down to a tolerable SPL, with the amp still running full blast and "properly" loaded.
Last edited:
Done!
It's hot to the touch, maybe 50 C after a few minutes of thrashing the amp with all 3 volumes at 10. If it fails, I'll use this other one, but have to get 6 ohms in front of it. I'm way up on the attenuation anyway, way down on the SPL. It's a Godsend - and I bet the neighbors would think so too. It's funny to hit that switch with it quiet as a chord is decaying, have it blast into a sustained feedback. Could almost put that into a Foot switch!
It's hot to the touch, maybe 50 C after a few minutes of thrashing the amp with all 3 volumes at 10. If it fails, I'll use this other one, but have to get 6 ohms in front of it. I'm way up on the attenuation anyway, way down on the SPL. It's a Godsend - and I bet the neighbors would think so too. It's funny to hit that switch with it quiet as a chord is decaying, have it blast into a sustained feedback. Could almost put that into a Foot switch!
Last edited:
I was able to coax the Vactrol back into operation today, using the 10K common cathode resistor of the so-called "cold-clipper" stage for a drive signal. It doesnt take much current to get the Vactrol moving its LDR around in value, ~300uA will shutdown the audio signal path I have it connect to, basically across one gang of the 250K "post" volume pot.
The tube of course has some quiescent voltage across that cathode resistor with no signal. I tried to trim that off some using an LED which has a larger forward drop than a regular diode. The diode has a 10K resistor in series with it, feeding a 0.5 uF cap, which discharges across a 2M potentiometer. The wiper of which connects to the Vactrol diode through a 180k resistor.
About mid setting you get the reverse envelope where as the guitar strings decay in amplitude exponentially, the Vactrol attenuated signal is doing the opposite; it's rising in amplitude. So the louder the guitar input is, the more squashed just one side of the drive to the output tubes - as it decays off the drive evens out in amplitude. A dynamic shift between SE and PP operation of the output stage.
The effect isnt so in your face either; it is more subtle than a full on compression.
So I am considering doing a voltage doubler for +/- DC off the existing grounded center tap 6.3V heater. To take it further, I really need op-amp circuitry to do the active analog envelope generation and processing to create a better drive signal for the Vactrol. Is there anything I'm unaware of attempting this power supply off the 6.3V heater - which has a grounded center tap already in place? i.e. ;
I'd be running a few op-amps off these voltages, input signal ground referenced, output signal floating (no current through ground). FWIW, I'm pretty sure my amp's heater winding can do better than 3A...
The tube of course has some quiescent voltage across that cathode resistor with no signal. I tried to trim that off some using an LED which has a larger forward drop than a regular diode. The diode has a 10K resistor in series with it, feeding a 0.5 uF cap, which discharges across a 2M potentiometer. The wiper of which connects to the Vactrol diode through a 180k resistor.
About mid setting you get the reverse envelope where as the guitar strings decay in amplitude exponentially, the Vactrol attenuated signal is doing the opposite; it's rising in amplitude. So the louder the guitar input is, the more squashed just one side of the drive to the output tubes - as it decays off the drive evens out in amplitude. A dynamic shift between SE and PP operation of the output stage.
The effect isnt so in your face either; it is more subtle than a full on compression.
So I am considering doing a voltage doubler for +/- DC off the existing grounded center tap 6.3V heater. To take it further, I really need op-amp circuitry to do the active analog envelope generation and processing to create a better drive signal for the Vactrol. Is there anything I'm unaware of attempting this power supply off the 6.3V heater - which has a grounded center tap already in place? i.e. ;
I'd be running a few op-amps off these voltages, input signal ground referenced, output signal floating (no current through ground). FWIW, I'm pretty sure my amp's heater winding can do better than 3A...
Last edited: