Hi , I was practicing the pushpull amplifier , then I figured out that circuit performs better without diode biasing. I share my simulation with you. If it is not , at least I think it should be not better , can you explain me please ?

 

Edit: I guess it is very inefficent without biasing and It is better to use lower resistances when biasing at R4 and R5

 

How do you define "performs better"?

Diodes are often used clamped to the heatsink on which the output transistors are mounted.
Because of the negative temperature coefficient, the diodes prevent thermal run-away.

 

How do you define "performs better"?

Diodes are often used clamped to the heatsink on which the output transistors are mounted.
Because of the negative temperature coefficient, the diodes prevent thermal run-away.

I said depending on output voltages but yes it is not. I just wondered.

 

I said depending on output voltages but yes it is not. I just wondered.

The diodes bias the transistors so they're always on to minimize crossover distortion. It works best when the diode drops match the base-emitter junction drops. Your simulations don't show any crossover distortion.

Where have you seen circuits drawn as you've shown? They were obviously drawn by someone who doesn't know what they're doing.

 

I thought the purpose of the diodes was to avoid crossover distortion. Have I been misinformed?

 

The diodes are not there to avoid cross-over distortion. You can avoid cross-over distortion using bias resistors.

Cross-over distortion is avoided by biasing the transistors in the linear region.
When the bases are connected, the transistors are class-B amplifiers. There is severe cross-over distortion.
You want to move away from class-B and move towards class-A, hence it is called class-AB.

You want to increase the bias voltage, i.e. separate the two base voltages. You can do this with bias resistors or with diodes.
Using two or more diodes (also with additional resistors) is a convenient way of providing two P-N junction forward voltages which gets you closer to a class-AB amplifier.

 

In the original circuit, without the diodes then both transistors in series are operating in class-A and drawing a very high current all the time.
Each 1k base bias resistor has (5V/2)- 0.7V= 1.8V across it resulting in a base current of 1.8mA. The typical hFE of 300 for the BC547C results in a collector current of 1.8mA x 300= 540mA even without a signal. But the BC547C transistor has a MAXIMUM ALLOWED CURRENT OF ONLY 500MA.

The heating in each transistor is (5V/2) x 540mA= 1.35W all the time. But the BC547C transistor has a MAXIMUM ALLOWED HEATING of only 0.625W.

With the diodes, the average transistor current without a signal is only about 20mA and the heating in each transistor is only about 0.05W.

 

I wanted to use this circuit as output stage. Thank you all for explaining to me.

 

When I set R4 and R5 higher values , as the input signal magnitude decreases , circuit gets away from unit gain

 

With audio amplifiers, including complementary ones, there is always a compromise between efficiency and distortion. Crossover distortion is a challenge indeed. I suggest studying the output sections of good quality audio amplifiers to see how a solution is reached, and then using a simulator to evaluate the effects of changes to the component values. It is often possible to learn a lot from observing how others have succeeded.

 

I will try , I also study differantial pairs now

 

Have a look at this thread:

https://forum.allaboutcircuits.com/threads/designing-a-simple-audio-amplifier.152741/

 

Study the biasing in this circuit.
You can replace Q1 and Q3 with NPN transistor 2N3904 and Q2 with PNP transistor 2N3906.
R6 provides negative feedback.

 

Study the biasing in this circuit.
You can replace Q1 and Q3 with NPN transistor 2N3904 and Q2 with PNP transistor 2N3906.
R6 provides negative feedback.


View attachment 329803

I worked on this , Q3 is preamplifying the signal , and remaining part is same. R3 and R4 changes gain. I guess R1 and R2 is for cables inner resistances.
But with Q3 , circuit gets distorted earlier compared to other setup. I upload both setups with asc file. Can you detail it to me please ?

 

R3 and R4 changes gain.

The gain is set by negative feedback via R6 and R5.

I guess R1 and R2 is for cables inner resistances.

No, they provide negative feedback to the biasing network.

 

The gain is set by negative feedback via R6 and R5.

No, they provide negative feedback to the biasing network.

Actually yes but the output is different from feedback gain and R3,R4 affects the gain

 

Hi demir,
You are over driving the Amp input, also made minor changes in the values.
The amp needs further design changes.

You must take more care when drawing and laying out your simulations, also use Labels at the Nodes. F4.
E

 

Hi demir,
You are over driving the Amp input, also made minor changes in the values.
The amp needs further design changes.

You must take more care when drawing and laying out your simulations, also use Labels at the Nodes. F4.
E
View attachment 329818

Hi Eric , seems like R5 must be small for higher current ability of amplifier. Is that right ?
I will try to make my circuits more readable.
Also it is good to see you again

 

seems like R5 must be small for higher current ability of amplifier.

Hi d,
This is the result for stepping R5. [Blue plot is R5=10R]
Try it for, say R2.
E