The article it came with is scant on detail, just referring to it as a "totem-pole follower" with unity gain. It's part of a mic preamp, basically a buffer to drive the output transformer. I want to modify it to add a master volume control on the output and I'm not sure if I should replace the 2.2M resistor with a pot, replace it with a T-pad, or put an H-pad after the output transformer. It all depends on which part of this circuit is least sensitive to varying impedance, and I can't find any information about this particular configuration of triodes.
thanks!
thanks!
I cropped it out by accident and can't edit the original post, but the secondary of the OT is connected to pins 2 and 3 of an XLR jack, with pin 1 to ground (i.e. balanced output). The OT is actually a Hammond 145O audio driver transformer with a primary of 15k and a secondary of 200R (as close as I could find to 150R without blowing my budget). I'm not posting the whole schematic because it came from an issue of TapeOp magazine that I paid for.
The input to the totem-pole is just two gain stages using a 12AX7 with a A100K pot between. I want this pot to act as a gain control so I can overdrive the second stage for some nice tube crunch. I'm using a Hammond 1140-Mn-A (150:15k 1:10) input transformer to drive the first gain stage.
It's also worth mentioning that I need the output to be mic level (-60 dBV/0.001 volt to -40 dBV/0.010 volt) and impedance (150R) because I want to use this as a vocal distortion pedal and feed the output into something that expects mic levels and impedance (a VoiceTone Harmony-G XT). I realize I could dispense with the totem-pole output and drive an unbalanced signal directly off the previous stage, but I want to drive a balanced signal.
I found this article about building a variable H-pad, and by following that rabbit hole discovered T-Pads (and L-Pads, which is just a voltage divider). I understand how they work - H is balanced, T is unbalanced, and the resistors in series with the signal present a constant impedance to the input and output.
What I don't know is where would be most effective (and least disruptive to operation) to place the pad. I understand that there's no easy way to make a perfectly constant-impedance attenuator and make it variable as well. I'm fine with that, I just want to minimize disruption to the operation of the circuit.
So my options seem to be:
The input to the totem-pole is just two gain stages using a 12AX7 with a A100K pot between. I want this pot to act as a gain control so I can overdrive the second stage for some nice tube crunch. I'm using a Hammond 1140-Mn-A (150:15k 1:10) input transformer to drive the first gain stage.
It's also worth mentioning that I need the output to be mic level (-60 dBV/0.001 volt to -40 dBV/0.010 volt) and impedance (150R) because I want to use this as a vocal distortion pedal and feed the output into something that expects mic levels and impedance (a VoiceTone Harmony-G XT). I realize I could dispense with the totem-pole output and drive an unbalanced signal directly off the previous stage, but I want to drive a balanced signal.
I found this article about building a variable H-pad, and by following that rabbit hole discovered T-Pads (and L-Pads, which is just a voltage divider). I understand how they work - H is balanced, T is unbalanced, and the resistors in series with the signal present a constant impedance to the input and output.
What I don't know is where would be most effective (and least disruptive to operation) to place the pad. I understand that there's no easy way to make a perfectly constant-impedance attenuator and make it variable as well. I'm fine with that, I just want to minimize disruption to the operation of the circuit.
So my options seem to be:
- variable H-pad after the OT
- variable H-pad after the OT, but omit the input resistors so the impedance the secondary of the OT "sees" is not exactly 150R
- variable T-pad before the OT (replacing the 2.2M resistor on the primary)
- L-pad before the OT, so the impedance the primary of the OT "sees" is variable (
Last edited:
White cathode follower, which has a fairly low output impedance.
https://www.tubecad.com/2016/05/blog0346.htm
https://www.tubecad.com/2016/05/blog0346.htm
Last edited:
Thanks! That's exactly what I was looking for.
Edit: I'm so excited I'm literally bouncing in my chair! The "Aikido" variation of the white follower from that article looks like it will solve my problem as it doesn't use a biased input, so I can put my master volume control between the second gain stage and the WCF and not have to mess with the output impedance!
Edit: I'm so excited I'm literally bouncing in my chair! The "Aikido" variation of the white follower from that article looks like it will solve my problem as it doesn't use a biased input, so I can put my master volume control between the second gain stage and the WCF and not have to mess with the output impedance!
Last edited:
That's a white cathode follower. Look it up. Since it's direct coupled from the second stage to the follower, one way to add a volume control like below. This way the volume control carries DC so you have to be careful of wiring and with insulation.
Thanks! I briefly thought of doing this, but initially discarded the idea because putting 150-ish volts through a pot didn't seem safe. 🤔 I also read that DC through a pot is pretty noisy?
Yes, that is fine. Of course, there must be coupling capacitors located both before and after the attenuator.
Only if the 12AU7 grid draws current, which can happen with a bad tube or biased improperly. The above circuit is done expecting no grid current. You can hear a scratchy noise when that happens. The 2.2M resistor (typically 10X the value or higher of the pot) is to ensure the grid has ground reference when the pot wiper loses contact intermittently. (I suggest doing that for ALL pots.) Bear in mind the other way to do it is to have TWO coupling caps in series with the pot sandwiched in between, as Rayma pointed out. The above is the more "purist" approach but has to be done with caution.
I would never do that, and I am very much a purist. And the pot housing must be grounded for shielding,
which would not be safe since there is no HVDC voltage isolation spec on the pot.
Also that circuit can considerably boost subsonic frequencies (by the amount of the attenuation),
which is also never a good idea.
Last edited:
That would work, but I think the output noise level may still be too high.
I would try it and see.
The other option is keep the direct coupling of the two circuits,
and use a 200R balanced H or U pad at the output.
The attenuator will reduce all of the output noise, along with the signal, which is highly desirable.
This addition of the output pad is easier to try first, so that's what I would do.
I would try it and see.
The other option is keep the direct coupling of the two circuits,
and use a 200R balanced H or U pad at the output.
The attenuator will reduce all of the output noise, along with the signal, which is highly desirable.
This addition of the output pad is easier to try first, so that's what I would do.
Last edited:
The schematic in Post # 6, and Post # 11, shows the first 2 triodes are 6072A (12AY7); and next two triodes are 12AU7A.
No 12AX7 there.
Just Saying
No 12AX7 there.
Just Saying