Hi , I have designed a circuit as shown below. Someone was suggested me to add feedback like how it is at first figure. That was working , but I was not able to calculate either the K or the gain. At some point It works , but its dependent on R8.
As I remember from Razavi's lessons , I modified this circuit to second figure. I only changed the input node. Now It seems working and I am able to calculate the gain. I be happy if u review it and correct me if I am wrong.
I also be happy I you can analyse how it works if the first one was also correct.
I add figures and asc file below.
Figure 1:


Figure 2:

 

I also know that , the maximium efficiency I can get at output stage around 75% but It requires good dc biasing. I failed at dc calculation however , can you lead me a little bit please ?

 

When employing feedback, you want the gain without feedback to be very high, at least 10 times the gain you want to set with feedback.

Your amp has a gain of about 10. If you use feedback, it will not be effective at gains above about 2.

What gain are you teying to achieve?

 

Another thought. The feedback works much better when a differential pair is used. You simply apply it, through a divider the inverting input.

 

When employing feedback, you want the gain without feedback to be very high, at least 10 times the gain you want to set with feedback.

Your amp has a gain of about 10. If you use feedback, it will not be effective at gains above about 2.

What gain are you teying to achieve?

In figure 2 , feedback response is pretty close but in first one it was depending on the value of R8. Since this is not a product , I do not have a constant gain. I try to learn by changing values and analysing it.

 

Another thought. The feedback works much better when a differential pair is used. You simply apply it, through a dividet the the inverting input

You are right , feedback work perfect in the diff pair. I was wondering about the output stage

 

Like this , it seems working good. But can you review ?
Edit: the input impedace of the output stage is so small now , it affects the first stage(diff pairs)
Edit2: I solved using 1 emitter follower before output stage.

 

Applying simple dc analysis , I solved the dc point.

 

Hi , I have designed a circuit as shown below. Someone was suggested me to add feedback like how it is at first figure. That was working , but I was not able to calculate either the K or the gain. At some point It works , but its dependent on R8.
As I remember from Razavi's lessons , I modified this circuit to second figure. I only changed the input node. Now It seems working and I am able to calculate the gain. I be happy if u review it and correct me if I am wrong.
I also be happy I you can analyse how it works if the first one was also correct.
I add figures and asc file below.
Figure 1:
View attachment 330947

Figure 2:
View attachment 330948

Hi , I've tested this circuit using 1watts 8ohm speaker and 24V supply and it worked great until tip41 burns hahaha.
never forget to add heatsink if you will run it for long time

 

Another thought. The feedback works much better when a differential pair is used. You simply apply it, through a divider the inverting input.

Hi , when I try to set the feedback from ppa's output and diff pairs inverting side , it does not work at all.
My diff pairs output swing also bounded at -0.8V at down side.
Could you help me ?
Part by part feedback works , but I want it to work like a opamp , where does it fail ?

 

Hi , when I try to set the feedback from ppa's output and diff pairs inverting side , it does not work at all.
My diff pairs output swing also bounded at -0.8V at down side.
Could you help me ?
Part by part feedback works , but I want it to work like a opamp , where does it fail ?
View attachment 331467

You would be better off if the long-tailed pair was driving a common-emitter stage, so I suggest removing Q13 and driving the output via a PNP common emitter stage which replaces Q8. It can then be DC coupled throughout.

 

You would be better off if the long-tailed pair was driving a common-emitter stage, so I suggest removing Q13 and driving the output via a PNP common emitter stage which replaces Q8. It can then be DC coupled throughout.

My purpose for Q13 was impedance matching and the problem starts before Q13.

 

I should probably make output stage unit gain and set feedback at diff pair but what about negative half ?

 

My purpose for Q13 was impedance matching and the problem starts before Q13.

But the voltage amplifier stage is transconductance, it is meant to have a high output impedance, and the impedance of the output stage is high because it is darlingtons, the impedance at that point is set by R12; and I have no idea why you originally had a common base stage in there.

 

But the voltage amplifier stage is transconductance, it is meant to have a high output impedance, and the impedance of the output stage is high because it is darlingtons, the impedance at that point is set by R12; and I have no idea why you originally had a common base stage in there.

Hi , after solving my problem at diff pair , I can go changes there. R12 can be replaced by semiconductor you are right , but current point I have questions about diff pair. Shall I go for same technique with ppa for down side of swing ?
I've tried adding cascode but did not work at all.

 

Hi , after solving my problem at diff pair , I can go changes there. R12 can be replaced by semiconductor you are right , but current point I have questions about diff pair. Shall I go for same technique with ppa for down side of swing ?
I've tried adding cascode but did not work at all.

No, with a common emitter stage for the Voltage amplifier, you don't need anything else.

 

No, with a common emitter stage for the Voltage amplifier, you don't need anything else.

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

 

No, with a common emitter stage for the Voltage amplifier, you don't need anything else.

I've added the asc file btw , I be happy If you can show your suggestion

 

I've added the asc file btw , I be happy If you can show your suggestion

No file supplied.

 

Hi , when I try to set the feedback from ppa's output and diff pairs inverting side , it does not work at all.
My diff pairs output swing also bounded at -0.8V at down side.
Could you help me ?
Part by part feedback works , but I want it to work like a opamp , where does it fail ?
View attachment 331467

there is the asc file but I also add here