I would like to try a choke-loaded follower (6AS7G) to drive the grid of my zero-bias, 211, Class A2 amplifier.
The behavior of chokes in a plate circuit is widely discussed...due to inductive collapse, a choke is capable of voltage swings nearly 2x that of a resistor in the same position. But what if we use a very large (50 H) choke in a low-impedance environment such as the load for a cathode follower?
A resistive-loaded cathode follower generally has a gain of ~0.9 or so under normal conditions. What if a choke is present as the load instead of a resistor? Will the inductive collapse of the choke effectively raise the gain of the follower to ~1.8 instead of the usual ~0.9 - or does the inductive collapse not enter the picture in this situation?
Oh and BTW, the bottom of the choke will be returned not to ground, but to a negative supply so the follower can swing equally + and - about the zero-bias point. The DC bias current flowing through the choke is selected so that the follower output is at zero volts in the resting, no-signal state.
Thanks for any insights!
The behavior of chokes in a plate circuit is widely discussed...due to inductive collapse, a choke is capable of voltage swings nearly 2x that of a resistor in the same position. But what if we use a very large (50 H) choke in a low-impedance environment such as the load for a cathode follower?
A resistive-loaded cathode follower generally has a gain of ~0.9 or so under normal conditions. What if a choke is present as the load instead of a resistor? Will the inductive collapse of the choke effectively raise the gain of the follower to ~1.8 instead of the usual ~0.9 - or does the inductive collapse not enter the picture in this situation?
Oh and BTW, the bottom of the choke will be returned not to ground, but to a negative supply so the follower can swing equally + and - about the zero-bias point. The DC bias current flowing through the choke is selected so that the follower output is at zero volts in the resting, no-signal state.
Thanks for any insights!
You won't pick up appreciable gain from choke loading the cathode follower. Maybe instead of .9 you'll have .95.
Depending on what you're up to, you may not need the negative supply at all. If you are going to go through the trouble of using a negative supply in the first place, then a resistor or a CCS would be OK to use and would remove a reactive component where you may end up needing a feedback loop.
Depending on what you're up to, you may not need the negative supply at all. If you are going to go through the trouble of using a negative supply in the first place, then a resistor or a CCS would be OK to use and would remove a reactive component where you may end up needing a feedback loop.
Thanks, audiowize. Can you please explain why the 2x swing of normal inductive collapse will not occur? Or hopefully there a text reference that I can read somewhere - Valley-Wallman or Langford-Smith perhaps?
I'm not sure why I'd need a feedback loop when I'm driving the grid of the output tube with a very low output impedance cathode follower? I do not believe that choke loading will raise the output Z of the follower - or am I mistaken?
I've built this driver before using a simple resistor as the CF load, with a negative supply as the return. The purpose of the negative supply in an A2 driver is so that the follower can swing equally as far negative as it can swing positive when direct-coupled to the grid of the final tube. It's pretty common procedure amongst the A2 crowd, whether they use MOSFETS or tubes as the follower.
I'm not sure why I'd need a feedback loop when I'm driving the grid of the output tube with a very low output impedance cathode follower? I do not believe that choke loading will raise the output Z of the follower - or am I mistaken?
I've built this driver before using a simple resistor as the CF load, with a negative supply as the return. The purpose of the negative supply in an A2 driver is so that the follower can swing equally as far negative as it can swing positive when direct-coupled to the grid of the final tube. It's pretty common procedure amongst the A2 crowd, whether they use MOSFETS or tubes as the follower.
You get more swing, but that just means you can drive the grid harder! You still get complete degenerative feedback, as you still have a cathode follower. As an example, when I use a choke in the cathode, I am often biasing an amp at 20-30V DC through the choke, which lets me swing up pretty much as far as I want, but it also lets me swing down more than the 20-30V of bias that I have.
That's more a comment about transmitting triode amps in general. If you're loading the 211 with a 10K+ output transformer, then you can get away without feedback. You could go lower and get more power, but you have to deal with damping and distortion if you do this.
Yes, I tend to use a negative rail when the bias voltage magnitude is really low. If your modeled bias voltage is 0V, but you may want it to be adjustable between -10V and 10V, you're better off with a negative rail and a resistive or active load for the cathode follower.
So now I'm confused! In your first post you wrote that gain with choke loading will just be slightly higher than the typical 0.9. But here you mention there will be more swing. Does that not imply more effective gain in the circuit? Or am I confusing the two terms?
Let's say your cathode is at +20V and resistively loaded, then you can swing +/-20V from the cathode follower before you ram into ground.
If your grid is at +20V but you have a 100H choke as the load. You can now sing well below 0V, and maximum voltage swing is greatly increased. (If this bugs you, remember that a choke loaded plate can swing above AC ground (B+), similarly the cathode with a choke load can swing below AC ground as well).
To obtain this larger max voltage swing, you must apply more voltage to the grid. This is still a cathode follower with degenerative feedback and it won't have more than unity gain.
If your grid is at +20V but you have a 100H choke as the load. You can now sing well below 0V, and maximum voltage swing is greatly increased. (If this bugs you, remember that a choke loaded plate can swing above AC ground (B+), similarly the cathode with a choke load can swing below AC ground as well).
To obtain this larger max voltage swing, you must apply more voltage to the grid. This is still a cathode follower with degenerative feedback and it won't have more than unity gain.
Last edited:
BINGO! The light just lit up in my head when you described the circuit conditions.
So...in essence, inductor collapse/increased swing DOES in fact occur, but we only realize it in the negative direction! Going positive of course it doesn't matter what the load is...resistor or choke.
Thanks again!
You read and repeat the words, but do not understand the concept.
Plus "inductive collapse" is a poor description.
The actual physical fact or phenomenon is that *when you vary current through a choke, you get a voltage proportional to inductance and said variation*
"Inductive collapse" is a poorly chosen narrow description of a particular case, when currently abruptly stops, and is more useful to explain car ignition than Audio preamps.
Analyze it better.
Again, do not simply "repeat the phrase"without understanding it.
Where do you think the 0.9X "number" comes from?
A Cathode Follower "would" have 1X gain, period, BUT its internal impedance is in series with the load, including the cathode to ground resistor.
So there you have the two legs of an attenuator.
IF the choke has a higher impedance **at the frequencies of interest** than the resistor you would have used there, you will have slightly higher *effective* gain; better said: you will have same gain as before (1X) but slightly less attenuated.
Not *real* higher gain.
NO.
"Inductive collapse"(it makes my head hurt) not only does not enter the picture, your entire theory is wrong.
So the cathode will also sit at a negative voltage. Voltage drop across a choke is negligible.
Now you will swing almost nothing downwards and almost twice as much upwards.
Doubt you want that.
Wrong.
DC causes negligible voltage drop acros an inductor.
To satisfy your conditions of inductor connected to a negative supply and follower output sitting at or near 0V means you will need to pass a horrendous current through said inductor.
Your entire tone of ridicule and condescention regarding an honest question I posted makes MY head hurt.
Next time I want to ask a question, I'll perhaps run it by you first so that you can approve of my terminology.
And you get the award by going to my "ignore list" with your first post to me. That really takes talent.
Last edited:
Member
Joined 2009
Paid Member