|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
76.90.38.219
In Reply to: RE: I don't ...RE: I Think You're Good to Go... posted by danlaudionut on September 09, 2014 at 09:44:05
I would think the electrons flowing out of the grid would want to flow towards the supply that created the +50 volt condition at the grid and that isn't the bias supply.
What if, just for the sake of the conversation, we made the output tube cathode biased and connected the low side of the secondary of the interstage transformer to ground.
Now where are the electrons from the grid of the output tube going while the grid is positive WRT the cathode? (What is satisfying the grid current?)
Doesn't it somehow have to be the driver tube?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Follow Ups:
> > the supply that created the +50 volt condition
Which is the Interstage.
> > Doesn't it somehow have to be the driver tube?
I would think it would be the transformer.
DanL
"I would think it would be the transformer."
Yes and that was my point.
The transformer is getting it from the driver tube.
Isn't the driver tube driving the primary, which drives the secondary and because of the inductance of the secondary isn't the secondary storing energy and that energy is what's supplying the grid current?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Those grid electrons (in the case of cathode resistor bias and a grounded IT secondary) are going to ground, since the come out the other end of the secondary winding. The do not magically tunnel through the iron core to get to the copper wire of the primary!
Current loops are complete circles - in this case, electrons come from the PSU ground, go through the bias resistor/capacitor, to the cathode, from which they jump to the grid when it is positive enough to attract them, then to the IT secondary and back to ground - loop closed.
"Current loops are complete circles"
I understand that but doesn't the source of the grid voltage need to be part of that circle?
I was thinking that the source was the energy within the secondary, imparted there by the primary, imparted there by the driver tube.
I think in terms of the positive side of a power supply as being an electron "vacuum cleaner", needing to suck up electrons and any current (electron) flow caused by the presents of the positive side of a power supply (being connected to a circuit with the other end of that circuit connected to the negative side of the power supply) has to flow back through the positive side of the power supply to get to ground.
So, in this case, the power supply has (indirectly) energized the secondary of the IT and the electrons DO find ground through the ground side of the IT but the energy is "supplied" ultimately by the driver tube not a bias supply that's happens to be between the ground side of the IT and ground.
The grid current flows through the bias supply, but it's not sourced from the bias supply?
Paul, I realize that the above probably reads like a jumbled mess to you.
I only hope you can decipher it well enough to see where my thinking is wrong and teach me something.
Thanks.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
The "source is, as you said, the IT. It is approximately a voltage source, so at any instant it's like a battery. When the signal is at a +200v peak, and the bias is -150v, then the grid is held positive to the cathode by +50v, causing the grid current.
The current in the secondary does not have the same electrons as the current in the primary, even though through the magic of electromagnetism the currents and voltages mirror each other.
That's what I thought.
The grid current is not being sourced by the bias supply but instead is being sinked through it?
If the total current (from and through) the bias supply is more than it's rated for Dan will have problems?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
"The grid current is not being sourced by the bias supply but instead is being sinked through it?"
Not necessarily. Here's another explanation, and I apologize if it's too simplistic. I'm just trying to lay out the whole concept from A to Z...
To set this up, the secondary of the IT swings voltage at its upper terminal referenced to ground. In order to do that, its lower terminal must be held as closely to AC ground as possible. We can't actually ground it, because that would short the negative bias needed by the grid. So, a bypass capacitor is connected from the bottom terminal to ground.
Here's what happens when this arrangement is driven in A2...
When the audio signal at the upper terminal of the transformer swings more positive than the grid bias, electrons are drawn from the grid. They flow down through the transformer secondary winding toward AC ground. The electrons are pulsed at an audio frequency, because in this design, the grid only draws current on positive-going peaks. The current pulses flow out of the secondary at the bottom terminal and through the bypass capacitor to ground. That's the complete path for grid current if the bias supply is isolated with a series rectifying diode, as most are. 10 ma of grid current creates exactly 10mA of current through the capacitor.
The problem with this arrangement is that the pulsed electrons entering the capacitor cause the average DC potential on the capacitor to become more negative. In other words, when grid current is drawn, the capacitor immediately becomes more negative than the bias supply that charged it up in the first place. If there is no DC path to ground at this point in the circuit, the capacitor will continue to charge as long as grid current is drawn. It won't stop until the capacitor and the grid become so negative that grid current no longer flows.
What's needed is a way to prevent the DC voltage on the bypass cap from increasing. That can be done in several ways, including a Zener of exactly the right voltage (as Paul noted), a bias supply that drains off the excess voltage by sinking it to ground, or even a resistor. If the purpose of the amplifier's A2 capability is only to handle musical peaks without clipping, and if those peaks push it into A2 only occasionally, the resistor will work just fine. The capacitor will charge up slightly when a peak draws grid current, but the resistor can be selected to discharge it in time for the next peak. Of course, this requires that the bias supply provide more current, because it will see the resistor as an additional load.
We could look at this from the perspective that all these techniques actually fall under the role of the bias supply. However, I think it's important to realize that DC stabilization of the capacitor when grid current is drawn is really a separate function. If a DIYer has a bias supply capable of providing 1 mA of negative current (current that charges the cap to the desired negative potential), it can be used in A2 as well as A1. We only need to make provision to drain the excess average voltage off the capacitor when grid current is drawn. Adding an additional circuit to do that might be a better alternative than scrapping the entire bias supply.
--------------------------
Buy Chinese. Bury freedom.
"I apologize if it's too simplistic."Simplistic works for me!
So if we had a shunt regulated bias supply, with the shunt component being the last element of the supply and fully capable of shunting that kind of current, then we're good?
And (conceptually) it would only be the shunt component that would need to be rated for the higher current, not the supply itself?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 09/10/14
"And (conceptually) it would only be the shunt component that would need to be rated for the higher current, not the supply itself?"
Exactly right. I'd say you're down with this one! :)
--------------------------
Buy Chinese. Bury freedom.
nt
It's easier using conventional "hole" theory ...
How can you get electron holes for your 50V grid
from the -125V supply which is hole starved?
It would be easier to have -175V across
the secondary than get more holes from B-.
DanL
Yes, I see the problem - we are used to think of power supplies as suppliers of something, that is as sources of unidirectional current. But if it's a real voltage source, it will provide whatever current is needed - positive or negative - to maintain the voltage. In the current case, it seems the bias "supply" is actually a sink at DC.
We certainly expect our power supplies to take negative current at audio frequencies - otherwise SETs would not work! The difference here is the need for negative DC.
Elsewhere in this thread there's a discussion of the circuit with cathode resistor bias - so the negative bias supply is zero volts and the ground is the power supply. This power supply is capable of DC in either direction, no sweat.
OBTW, this reminds me of the time when many were using battery bias in the grid line, and found that sometimes the battery would rupture - from being overcharged.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: