Amplifier bootstrapping and negative feedback

[boylesg]

I understand what these are supposed to do but I don't yet fully comprehend how they do it.

I have been reading about them but have not yet come across a site that I have been able to understand fully.

Is there some one in electrical engineering teaching who might be able to help me out and try explaining it to me as simply as possible, perhaps with some appropriate analogies if there are any?

I really want to try and understand a bit better how these things work......just for personal satisfaction if nothing else.

 

[ninju]

Imagine a bowl. You have a little ball in the bowl. If you nudge the ball up the walls of the bowl and leave it, gravity pulls the ball back to the center. This is an example of a negative feedback system. So, negative feedback stabilizes a system.

With respect to electronics, you can find 4 basic amplifiers, VCVC, VCCS, CCVS, CCCS. Application of appropriate kind of feedback to these sources makes the amplifier

1. A better source of the signal.
2. A better load for the signal.

eg, considering the case of a VCCS (Voltage Controlled Current Source), it supplies current and senses voltage. Hence, its input impedance increases and the output impedance increases.

Bootstrapping is some kind of a kind of positive feedback. (according to my analysis, nobody told me its a positive feedback)

It is used to increase the input impedance of the circuit and also in startup in a few cases.

 

[boylesg]

Imagine a bowl. You have a little ball in the bowl. If you nudge the ball up the walls of the bowl and leave it, gravity pulls the ball back to the center. This is an example of a negative feedback system. So, negative feedback stabilizes a system.

With respect to electronics, you can find 4 basic amplifiers, VCVC, VCCS, CCVS, CCCS. Application of appropriate kind of feedback to these sources makes the amplifier

1. A better source of the signal.
2. A better load for the signal.

eg, considering the case of a VCCS (Voltage Controlled Current Source), it supplies current and senses voltage. Hence, its input impedance increases and the output impedance increases.

Bootstrapping is some kind of a kind of positive feedback. (according to my analysis, nobody told me its a positive feedback)

It is used to increase the input impedance of the circuit and also in startup in a few cases.

Thanks. I understand what positive and negative feedback is in general, from my medical science days, but not how it works in this amplifier.

Perhaps what I need to do is implement this amplifier in the javascript circuit simulator thing and get a visual representation of what is happening to the current in the various legs of the circuit.

I added a volume control to my amplifier - it was rather easy and exactly as I assumed it would be implemented.

I also added a passive Baxandall tone control circuit and it sort of works except that I think there is something wrong with one of the 100K pots I salvaged from a stero unit. The signal coming out of the circuit is quite weak but when I twiddle one pot in particular the signal comes through much stronger briefly, and when it does I can detect a change in tone. Perhaps the wiper is damaged some how.

I need to try this with some brand new double gang pots.

Volume control is logarithimic pots, but what is recommended for bass and treble?

 

[ninju]

Could you post the circuit diagram?

I think you might have an impedance matching problem. Most of the speakers are 8 ohm speakers.

 

[Syncopator]

I think you might have an impedance matching problem.

Doubtful. The output stages of audio amplifiers are voltage sources. So long as the power supply can supply enough current, and the output stage can withstand it, the load impedance is not critical.

 

[boylesg]

Here is my circuit:

Stage 1 / A has a voltage gain of 20, input impedance of 1k and output impedance of 100r.

It all works beautifully with really good sound quality and I have to turn the PC speaker volume right down to avoid distortion. Adding amplifier volume control and passive tone control changes that some what but can be alleviated by adding an opamp to the output of the tone control. But that is beside the point for the purposes of this discussion.

First things first.

I don't undestand how the connection between C1 + and R2 22k works. With the stage 1 / A on its own, R2 is connected to Vcc where it forms a voltage divider at the base of Q3 and biases it to what ever volts so that Q3 is in the middle of its conducting range.........

But with that connection severed and R2 connected to the output of stage 2 / AB, what is Q3 now biased to? Is the bias of Q3 now controlled by the voltage level of the imput signal?

 

[boylesg]

Have I got my head correctly around how a complementary pair works?

1) You bias the 2 bases apart by 1.2V, or what ever, so the base of the npn is at +0.6V and the base of the pnp is at -0.6V relative to the centre of two diodes or the transistor in a 'rubber diode' or what ever.

2) Under these conditions the npn is just on the verge of conducting and the pnp is just on the verge of not conducting.

3) A positive audio signal comes in and its voltage adds to the 0.6V on the base of the npn which then conducts and reproduces the signal with a current gain.

4) At the same time the positive audio signal makes the base of the pnp even more positive and hence it stops conducting completely.

5) The audio signal goes negative and makes the base of the npn lower or negative and hence it stops conducting completely.

6) At the same time the negative audio signal makes the base of the pnp even more negative and hence it conducts and reproduces the negative audio signal with a similar current gain.

 

[ninju]

Is the schematic correct? Are you using a 4700F capacitor? Also, is R2 22k ohm and R1 8.2 ohm? i can't simulate the circuit for another 6 hours or so.

How are you biasing Q3? R2 is connected to a capacitor C1, which i believe to be a coupling cap.

What does C3 do ? You have the same resistance for both R7 and R8.

Also, whats the input signal amplitude?

Also, please introduce a cap between Q3 and Q2 to make a capacitive coupling between them. Direct coupling might cause problems. You don't have a problem with capacitive coupling as its a discrete circuit and not an integrated circuit.

 

[boylesg]

Is the schematic correct? Are you using a 4700F capacitor? Also, is R2 22k ohm and R1 8.2 ohm? i can't simulate the circuit for another 6 hours or so.

How are you biasing Q3? R2 is connected to a capacitor C1, which i believe to be a coupling cap.

What does C3 do ? You have the same resistance for both R7 and R8.

Also, whats the input signal amplitude?

Also, please introduce a cap between Q3 and Q2 to make a capacitive coupling between them. Direct coupling might cause problems. You don't have a problem with capacitive coupling as its a discrete circuit and not an integrated circuit.

Sorry ninju, I am not having problems with the circuit and I am not asking for help with it - the amplifier works just fine.

What I am asking is for some one to explain to me how the negative feedback that I outlined works because I don't 'get it'.

To re-iterate....this is what I want to understand:

I don't undestand how the connection between C1 + and R2 22k works. With the stage 1 / A on its own, R2 is connected to Vcc where it forms a voltage divider at the base of Q3 and biases it to what ever volts so that Q3 is in the middle of its conducting range.........

But with that connection severed and R2 connected to the output of stage 2 / AB, what is Q3 now biased to? Is the bias of Q3 now controlled by the voltage level of the imput signal?

 

[ninju]

The fact is, the circuit may theoretically work, but its pretty in-efficient. And 4700 F isn't a capacitor thats used in conventional electronics. They are used to power automobiles and trucks. Also, the bias point of Q3 seems to be changing with respect to the input signal. This defeats the meaning of the word quiescent point.

 

[boylesg]

The fact is, the circuit may theoretically work, but its pretty in-efficient. And 4700 F isn't a capacitor thats used in conventional electronics. They are used to power automobiles and trucks. Also, the bias point of Q3 seems to be changing with respect to the input signal. This defeats the meaning of the word quiescent point.

I don't doubt you but I am not trying to create a high quality hifi system here, just a muck around amplifier.

I have been experimenting with different input and output impedances for the stage 1, combined with the stage 2, and 1k in and 1k out works fine with an ipod and pair of PC speakers. And the capacitor values are quite a bit more sensible.

I have not been able to figure out how to determine or vary the impedances of the stage 2 complementary pair however. But it seems to work with a fairly wide range stage A output impedances, whether efficiently is another matter.

Don't suppose you could explain how to determine the input impedance of a complementary pair stage?

- - - Updated - - -

Also, the bias point of Q3 seems to be changing with respect to the input signal. This defeats the meaning of the word quiescent point.

Does that have anything to do with the negative feedback from stage 2 output?

 

[ninju]

Its just two transistors connected in parallel.So, assuming the transistors are matched, it'd be half of the single transistor configuration, which is load dependent.

 

[boylesg]

Input imedance of around 1k seems to be about right for an ipod. I tried input impedances of 40 and 100 but it causes the resistors of the class A first stage to heat up too much.

Output impedance of 1k for the class A first stage is no good as it wont drive the class AB second stage adequately apparently.

I have tried output impedance of 100r and it works with the class AB second stage OK, but it uses a 4700uF capacitor which is extreme as has been pointed out.

I will try class A first stage output impedance of 200r next which uses a 2200uF capacitor and seems a bit more sensible at least.