Understanding the bottlehead for silly experiments

Audiotechyo · 2393

0 Members and 1 Guest are viewing this topic.

Offline Audiotechyo

  • Jr. Member
  • **
    • Posts: 20
on: February 14, 2018, 01:06:27 PM
Greetings, I have had my Crack kit for a while and really, really enjoy it. Being a young electrical engineer I just can't help but want to experiment with it, some things I am interested in trying are as follows.

1) add a pre-out from my crack at the pre-amp stage, this pre-out would then be passed through a unity gain buffer stage in order to ensure that as little energy from the actual signal is used. I am trying to keep my crack from changing its sound in any way if I can.

2) potentially also take this new signal and add an active filter to cut off the low end so I can have a sub out since I have a nice powered sub right below my desk.

3) while I dont think adding this will give me literally any real accurate information I would like to add a VU-Meter I salvaged from a busted TEAC Tape Deck years ago. I have tested this VU-Meter and it does indeed still function when I hook it up to my powersupply. (Needles move up and down)
With this I also have the intention of hooking it up via a unit gain buffer stage.

So my question is, from my understanding, the pre-amplifier stage of the amp ends at the bottom of the 1watt resistor on the given schematic that I dont think I am allowed to post here. The schematic says this point will be 90Volts DC, which I assume means I will have to use a cap to remove the DC part of the signal and then I should get what I want.

I am still new to EE so please let me know if I have solid understanding of whats happening or if I am missing some key points. I really have worked very little with amplifiers, the two stage small signal transistor amplifier I designed in college wasn't without a lot of help and although I think I grasp how it works, I dont have a scope at my disposal to check things.



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19772
Reply #1 on: February 14, 2018, 01:53:41 PM
1) add a pre-out from my crack at the pre-amp stage, this pre-out would then be passed through a unity gain buffer stage in order to ensure that as
little energy from the actual signal is used. I am trying to keep my crack from changing its sound in any way if I can.
This is totally doable with a pair of mosfets setup as a source follower running a few mA of current.

2) potentially also take this new signal and add an active filter to cut off the low end so I can have a sub out since I have a nice powered sub right below my desk.
Most subs have active crossovers built in, so adding this complexity in the Crack may not be the best idea.

3) while I dont think adding this will give me literally any real accurate information I would like to add a VU-Meter I salvaged from a busted TEAC Tape Deck years ago. I have tested this VU-Meter and it does indeed still function when I hook it up to my powersupply. (Needles move up and down)
With this I also have the intention of hooking it up via a unit gain buffer stage.
The mosfets should have no problems driving the meter.  If you use one mosfet per channel and cap couple them to the meter (one cap from each fet), then you'll get the desired channel separation and no loading on the circuit itself.


So my question is, from my understanding, the pre-amplifier stage of the amp ends at the bottom of the 1watt resistor on the given schematic that I dont think I am allowed to post here. The schematic says this point will be 90Volts DC, which I assume means I will have to use a cap to remove the DC part of the signal and then I should get what I want.
That's *if* you want to remove the DC.  If you put a cap there, then you need a resistor from the other end of that cap to ground, and that will load the driver stage of the Crack (not recommended).

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Audiotechyo

  • Jr. Member
  • **
    • Posts: 20
Reply #2 on: February 14, 2018, 04:28:42 PM
1) How interesting, I never learned about source followers in school.
2) Thats a good point, this probably isnt a value added thing, I just wanted to practice my filters I guess.
3) Ill have to look into how best make this work as well.
4) So this is one of the reasons I wanted to use the OPAMPs, I didnt want to load the crack at all if possible. If I have this huge DC part of the signal I would assume that I'll then need to use a voltage divider in order to get the signal to a line level output (2VRMS I believe is tpyical?) With this source follower circuit do I really achieve the isolation that I would expect to get from using a Unity Gain Buffer circuit? Am I wrong to assume that I will achieve a well isolated circuit with my buffer stage? I want to use the more efficient option as its just good engineering but I want to understand the trade offs.

Thanks you so much for your reply! I am really excited to learn about source follower circuits and to work on this project.

Edit: upon some quick research it seems a source follower is also commonly referred to as a common drain amp which, haha, I DID learn about but I had forgotten them. Im not sure if my professors ever pointed out that it could also be used as a voltage buffer, but then again maybe I just forgot it.
« Last Edit: February 14, 2018, 05:19:04 PM by Audiotechyo »



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19772
Reply #3 on: February 15, 2018, 04:12:27 AM
4) So this is one of the reasons I wanted to use the OPAMPs, I didnt want to load the crack at all if possible.
The plates of the 12AU7 are about 240V at startup and they drop down to 60-100V during operation.  If you don't want to load this stage, you need to directly couple a device to those plates with very high input impedance.  I'm not so sure an opamp is going to handle those voltages well.

If I have this huge DC part of the signal I would assume that I'll then need to use a voltage divider in order to get the signal to a line level output (2VRMS I believe is tpyical?)
There is no DC part of the signal; the signal is all AC.  The amount of AC present depends on your source output and the position of the potentiometer.


With this source follower circuit do I really achieve the isolation that I would expect to get from using a Unity Gain Buffer circuit?
I still don't see how you're going to implement a unity gain buffer with an op-amp without loading the circuit.  The source follower will not load the driver stage.

Am I wrong to assume that I will achieve a well isolated circuit with my buffer stage?
Draw one up and post it.

I want to use the more efficient option as its just good engineering
Well, the source follower requires one fet, one loading resistor, one coupling cap, and one resistor to keep the output side of the coupling cap at ground potential.  What does your opamp circuit require? 


Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Audiotechyo

  • Jr. Member
  • **
    • Posts: 20
Reply #4 on: February 16, 2018, 12:02:12 PM
The plates of the 12AU7 are about 240V at startup and they drop down to 60-100V during operation.  If you don't want to load this stage, you need to directly couple a device to those plates with very high input impedance.  I'm not so sure an opamp is going to handle those voltages well.

I was thinking about this the night I wrote it, one of the reasons I was choosing an OP AMP in my thoughts was because they should be able to have very, very high input resistance. However maybe I am just too inexperienced and hung up on theoretical OP AMPs. Once I thought more about the voltages I could be dealing with I started rethinking the OP AMP route because getting something that can handle those higher voltages is probably very expensive, and probably even harder to find one with a good slew rate and also being unit gain stable.

There is no DC part of the signal; the signal is all AC.  The amount of AC present depends on your source output and the position of the potentiometer.
Hm, I guess I was confused with what I was looking at on the Crack schematic that I was looking at.
I have attached a picture to try to clarify what I got the idea from, maybe I just have a fundamental misunderstanding? I assumed I would need a capacitor to get rid of this DC component in my signal.

I still don't see how you're going to implement a unity gain buffer with an op-amp without loading the circuit.  The source follower will not load the driver stage.
Draw one up and post it.
I will work on this however I still think that there must be some misunderstanding on my end about the schematic.

Well, the source follower requires one fet, one loading resistor, one coupling cap, and one resistor to keep the output side of the coupling cap at ground potential.  What does your opamp circuit require?

I had assumed that the source follower would require less components to achieve however I was unfamiliar with how high the input resistance would be, I am also pretty inexperienced in what constitutes high output resistance, or what would be high enough. As said before from my limited experiencing of op amps I would assume they have very, very high output impedance, granted maybe what gives them this feature from the beginning could be a source follower at the beginning stage, I will have to look into that.

I want to thank you for dealing with me, to be honest right out of college I have been mostly doing systems engineering work for the past 2 years and I am rusty on my real electrical knowledge, plus even in college I somehow wondered how I was even passing. Kind of sad to admit that I guess...



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19772
Reply #5 on: February 16, 2018, 01:06:59 PM
Yes, signal can be taken from the plate of the 12AU7 as you've drawn, but the DC needs to be blocked (with a cap) at some point.

The input impedance of whatever buffer you use doesn't have to be that high.  200K should be high enough.

You know, the output of the Crack is already cap coupled and you could just feed a pair of opamps from there to drive your meter.

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Doc B.

  • Administrator
  • Hero Member
  • *****
    • Posts: 9664
    • Bottlehead
Reply #6 on: February 16, 2018, 01:16:32 PM
Yes, I was wondering why a voltage amp with a tube buffer needs the addition of a buffer. As far as a meter is concerned, they tend to affect the sonics so much that I would leave one out whenever given the choice. We spent a few weeks designing and building the Neothoriator and then spent months on end getting the neon meter tubes buffered heavily enough to stop creating distortion. Ditto connecting VU meters to the Tube Repro. It usually creates more problems than are worth it in the case where you don't really need a meter.

Dan "Doc B." Schmalle
President For Life
Bottlehead Corp.


Offline Audiotechyo

  • Jr. Member
  • **
    • Posts: 20
Reply #7 on: February 16, 2018, 01:39:58 PM
Yes, signal can be taken from the plate of the 12AU7 as you've drawn, but the DC needs to be blocked (with a cap) at some point.

The input impedance of whatever buffer you use doesn't have to be that high.  200K should be high enough.

You know, the output of the Crack is already cap coupled and you could just feed a pair of opamps from there to drive your meter.

Ah see, thats what I meant when I said:

So my question is, from my understanding, the pre-amplifier stage of the amp ends at the bottom of the 1watt resistor on the given schematic that I dont think I am allowed to post here. The schematic says this point will be 90Volts DC, which I assume means I will have to use a cap to remove the DC part of the signal and then I should get what I want.

Then when you replied with:
That's *if* you want to remove the DC.  If you put a cap there, then you need a resistor from the other end of that cap to ground, and that will load the driver stage of the Crack (not recommended).

So then I started thinking about NOT using a cap to remove the DC (offset I think is the correct term?) and I replied with:

There is no DC part of the signal; the signal is all AC.  The amount of AC present depends on your source output and the position of the potentiometer.

I started getting confused, but for now I want to ask some other questions;

Yes, I was wondering why a voltage amp with a tube buffer needs the addition of a buffer. As far as a meter is concerned, they tend to affect the sonics so much that I would leave one out whenever given the choice. We spent a few weeks designing and building the Neothoriator and then spent months on end getting the neon meter tubes buffered heavily enough to stop creating distortion. Ditto connecting VU meters to the Tube Repro. It usually creates more problems than are worth it in the case where you don't really need a meter.

Is the tube 12AU7 being used as a buffer? I was not really thinking about it in that sense, I was assuming that it was acting as a pre-amp? Am I over thinking the difference between a buffer and pre-amp? I am unfortunately not experienced in typical audio amplifier design so maybe this obvious fact flew over my head.

If meters effect the sonics of a tube amp then I'm interested in why and how to get past that. You mention neon meter tubes, I can take out the bulbs and replace them with LED's to attempt to mitigate that issue, my meter is not a true vintage piece that I would feel bad about ruining. Its a cheap salvage meter from a very cheaply made cassette tape deck, just in case you thought I was using something of actual value from real tape players. I assumed that if I got the signal low enough that an OPAMP Unity Gain Buffer would provide great isolation to nab the signal and use it for whatever I wanted without loading the rest of the amp, it sounds like this was a naive assumption on my part, which I was always worried I had over simplified it in my head.

As always I really, really appreciate the time you take to answer my questions and all the help and advice you provide. I wish I wasn't all the way in Florida so I could participate in the meets and the like.



Offline Doc B.

  • Administrator
  • Hero Member
  • *****
    • Posts: 9664
    • Bottlehead
Reply #8 on: February 16, 2018, 02:03:17 PM
The 6080 is the buffer that follows the 12AU7.

Dan "Doc B." Schmalle
President For Life
Bottlehead Corp.


Offline Audiotechyo

  • Jr. Member
  • **
    • Posts: 20
Reply #9 on: February 16, 2018, 03:35:29 PM
I see, well I guess I thought that a typical audio amplifier would have a pre-amp stage and a power amp stage, Im aware that there are dedicated pre-amps and dedicated power amps but in this case I figured that wasn't the case. I don't really understand why there is a buffer after the 12AU7 then, does it keep the load from the headphones isolated from the amplifier stage then?



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19772
Reply #10 on: February 17, 2018, 07:06:36 AM
The 6080 is setup as a cathode follower (think current amp).

The 12AU7 is setup as a voltage amplifier.

The 12AU7 is unable to cope with headphones as a load (it can't swing enough current to drive headphones, and its internal impedance is quite high). 

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Audiotechyo

  • Jr. Member
  • **
    • Posts: 20
Reply #11 on: February 17, 2018, 11:20:00 AM
Ah, I see, well that makes more sense then why in other threads about using this as a pre-amp it was suggested to just take the output from the headphone jack into RCA jacks... very interesting.
Well then, in my case I guess I want to take from the buffer tube as well. Is there a more detailed spec of what the output is? I guess at this point I am interested in what audio measurements I should be interested in? Should I be thinking about RMS voltage? For testing should I just run a sign wave from my computer out of my DAC?



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19772
Reply #12 on: February 17, 2018, 05:30:37 PM
For testing should I just run a sign wave from my computer out of my DAC?
Sure, give that a go.

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Audiotechyo

  • Jr. Member
  • **
    • Posts: 20
Reply #13 on: February 17, 2018, 08:02:02 PM
I wish I had a scope so I could see what kind of amplitudes I should expect, I will work on borrowing one, in the mean time is there any way you could give me an idea of how large they might get? I suppose this in part has something to do with what size of a signal my DAC puts out to begin with... which I dont currently know.



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19772
Reply #14 on: February 18, 2018, 04:56:21 AM
I suppose this in part has something to do with what size of a signal my DAC puts out to begin with... which I dont currently know.
You'll get roughly 12-14x what you put in out of the jack with the volume pot all the way up.

You'll never, ever listen to the amp with the volume pot all the way up.

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man