But now for example , I set diff pairs feedback gain to 4 , and set input swing 2Vpp , than it does not work.

The input diff pair and its current mirror should manage a gain of about 100,000

 

The input diff pair and its current mirror should manage a gain of about 100,000

I mean its W feedback. The problem is output swing

 

I mean its W feedback. The problem is output swing

A common emitter amplifier on the output of the diff pair will swing rail-to-rail.

 

A common emitter amplifier on the output of the diff pair will swing rail-to-rail.

But for example , when my input swing is around 1V and I set feedback gain at inverting input to 4 , the signal goes distorted at down side.

 

It will do.
You only need ≈0.1V to drive a common emitter amplifier.

 

It will do.
You only need ≈0.1V to drive a common emitter amplifier.

But I have large inputs so what to do ?
In commercial opamps we can do what I mean. What is missing in my circuit ?
@Ian0

 

But I have large inputs so what to do ?
In commercial opamps we can do what I mean. What is missing in my circuit ?
@Ian0

Global feedback will fix it, it works just the same as in an op-amp.
Connect a potential divider between the output and the inverting input of the diff pair. (The inverting input is the one that doesn't drive the next stage, it drives the diode side of the current mirror)
If the inputs are larger than a couple of volts, then make an inverting amplifier, or divide them down.

 

Global feedback will fix it, it works just the same as in an op-amp.
Connect a potential divider between the output and the inverting input of the diff pair. (The inverting input is the one that doesn't drive the next stage, it drives the diode side of the current mirror)
If the inputs are larger than a couple of volts, then make an inverting amplifier, or divide them down.

Can you draw it please sir, so I can check when I get back to computer.

 

Can you draw it please sir, so I can check when I get back to computer.

This is something similar I've been working on.
R7/R9 and R12/R13 help with linearity if there is some imbalance between the two transistors in the pair.
You can use darlingtons in the output, in which case you will need less current in the voltage amplifier stage.
I used a JFET current source for the long tailed pair, you can keep the current mirror.
C1 is the dominant pole capacitor (you'll need that to keep it stable, especially with slow output transistors)
R21/C8 is the Zobel network (you'll need that to keep it stable into an inductive load)

 

This is something similar I've been working on.
R7/R9 and R12/R13 help with linearity if there is some imbalance between the two transistors in the pair.
You can use darlingtons in the output, in which case you will need less current in the voltage amplifier stage.
I used a JFET current source for the long tailed pair, you can keep the current mirror.
C1 is the dominant pole capacitor (you'll need that to keep it stable, especially with slow output transistors)
R21/C8 is the Zobel network (you'll need that to keep it stable into an inductive load)View attachment 331491

Here you get output from the side of noninv input and can you give large input swings without distortion ?
At this point my problem is how do I change my input stage and to understand why is that.

 

It will work until either:
the base of Q1 is higher voltage than the base of Q6
or
The emitter of Q1 is too low for J1 to keep a constant current.
If you had a current mirror as a constant current source for the long tailed pair, it would continue to work until the input was 1.3V above the negative rail.

 

It will work until either:
the base of Q1 is higher voltage than the base of Q6
or
The emitter of Q1 is too low for J1 to keep a constant current.
If you had a current mirror as a constant current source for the long tailed pair, it would continue to work until the input was 1.3V above the negative rail.

thats what my problem exactly is. how can I improve this

 

Below is an example of relatively simple audio amp circuit that uses a differential input stage (Q1 and Q2) for both AC and DC feedback from the output.
The AC feedback reduces distortion and the gain is determined by the relative value of R4 and R5.
The DC feedback through R5 biases the output at 1/2 the supply voltage as generated by R1 and R8.

 

Below is an example of relatively simple audio amp circuit that uses a differential input stage (Q1 and Q2) for both AC and DC feedback from the output.
The AC feedback reduces distortion and the gain is determined by the relative value of R4 and R5.
The DC feedback through R5 biases the output at 1/2 the supply voltage as generated by R1 and R8.

View attachment 331498

so do I need to use a different topology ?
lm741's inner circuit seems similar with mine but I could not solved my issue.
Its reasonable transistor goes off when it exceeds 0.8V downwards but what may should I look for.
My purpose is to experience something and learn , I analysed the topology in mine , and I found it following Razavi's book.
I know I do not have to use it always I first need to know why am I doing that when changing something.

 

The 741 internal circuit isn't a long tailed pair. That makes it rather different from most available op-amps.
It is a version of the "no-tail pair" or "complementary long-tailed pair" like this one.
https://www.4qdtec.com/pwramp.html
It's not a good amplifier.

 

The 741 internal circuit isn't a long tailed pair. That makes it rather different from most available op-amps.
It is a version of the "no-tail pair" or "complementary long-tailed pair" like this one.
https://www.4qdtec.com/pwramp.html
It's not a good amplifier.

Yes you are right. I just want to learn about the subject and move on since probably I will use opa's instead building one in future.
Do you have any suggestions for me today sir ? May be circuit or tech to look for ?

 

The 741 internal circuit isn't a long tailed pair. That makes it rather different from most available op-amps.
It is a version of the "no-tail pair" or "complementary long-tailed pair" like this one.
https://www.4qdtec.com/pwramp.html
It's not a good amplifier.

https://www.youtube.com/shorts/22MgQhAGwaE here is my power amplifier working.
I hope copyright wont hit it

 

Yes you are right. I just want to learn about the subject and move on since probably I will use opa's instead building one in future.
Do you have any suggestions for me today sir ? May be circuit or tech to look for ?

Read Doug Self's book on power amplifiers, or any of John Linsley Hood's.

 

https://www.youtube.com/shorts/22MgQhAGwaE here is my power amplifier working.
I hope copyright wont hit it

You could even try listening to some music where you would actually notice if it was distorted or not!

 

You could even try listening to some music where you would actually notice if it was distorted or not!

I checked output using oscilloscope and listened different sounds , seems working. Sounds not distorted