Facebook
Google
GitHub
Linkedin
So , why is that gain is different from the feedback ? Or do I calculate it wrong ?
Push pull amplifier biasing
- Thread starter demir-ali
- Start date
Below is my favorite, relatively simple, audio amplifier:
It uses DC and AC negative feedback from the output to a differential input stage (Q1 and Q1) to stabilize the output DC bias at 1/2 the supply voltage, set the amp gain, and reduce distortion.
The feedback value is determined by R4, R5, and C2.
It uses DC and AC negative feedback from the output to a differential input stage (Q1 and Q1) to stabilize the output DC bias at 1/2 the supply voltage, set the amp gain, and reduce distortion.
The feedback value is determined by R4, R5, and C2.
Last edited:
Seems well but it is complex for me to analyse for now. I want to practice previous example, then study differantial pairs and amplifiers. I have no experience with differantial pairs.
I wonder why it is a better setup with Q3 and how to calculate the gain and the affect of feedback ? Excuse me if I am missing something.
Basically, a differential pair is biased and operating normally when the two input voltages are very near equal, thus the output is determined by any small difference (differential) between the two input voltages.
Q3 is a Vbe multiplier circuit which allows setting of the output transistors bias current by adjusting the values of R6 and R7..
You want the output bias current high enough to eliminate crossover distortion, but low enough so it doesn't generate significant bias power in the output transistors.
The AC feedback is determined by the attenuation ratio of R4 and R5 that gives an AC feedback voltage at Q2's base equal to the input voltage at Q1's base.
For the R4 and R5 values shown, the feedback attenuation is about 0.28, giving a gain of about 3.6 (i.e. the output AC voltage has to be 3.6 times the input for the two differential voltages to be equal).
Due to C2, the DC feedback gain is 1, thus the output DC bias point equals the bias voltage at Q1's base of 1/2 the supply voltage.
Hi , I slightly got your circuit by seperated parts but it is still hard to look at the big picture for me.
I also understood the feedback in your circuit but I am very new with these topologies.
Also what does Q3 make better the circuit below ? I know a active load is better than a resistor but here why is that ?
I also understood the feedback in your circuit but I am very new with these topologies.
Also what does Q3 make better the circuit below ? I know a active load is better than a resistor but here why is that ?
Output transistors Q1 and Q2 are emitter followers. They do not provide voltage gain. They provide current gain and very low output impedance.
Q3 is the driver stage. It is a common emitter amplifier. It provides both current and voltage gain. It also presents a high input impedance.
Q3 is the driver stage. It is a common emitter amplifier. It provides both current and voltage gain. It also presents a high input impedance.
How can I calculate the gain in this one ? I can change it by changing values but not by calculating.
When I apply V/I = R over diodes and take it as Rc, RC/RE corrects the gain. But , how should I calculate ?
Audioguru again
- Joined Oct 21, 2019
- 6,826
I made an error for your little low power output transistors. The maximum allowed curent for the BC547C anf BC557C transistors is only 100mA. Your simulations are trying to get 2V peak into 8 ohms= 250mA.
The simulation program does not care if you overload the transistors to cause destruction.
2N3904 and 2N3906 transistors have a maximum allowed output current of 200mA but they have the American EBC pinout.
Your BC547 and BC557 have the European CBE pinout.
The simulation program does not care if you overload the transistors to cause destruction.
2N3904 and 2N3906 transistors have a maximum allowed output current of 200mA but they have the American EBC pinout.
Your BC547 and BC557 have the European CBE pinout.
I knew about it but since it is simulation did not care. My problem is about calculating the gain. I also be happy If you can suggest me some circuit models or example to study.
Hi demir,
This simulation compares the Va/2 Nodes, with and without the connection to the Vj Node.
I have reduced the Vin to 10mVppk in order to not limit the output voltage swing.
Try running the asc file, show your calculations for the expected gain of the Q3 stage.
What are your conclusions on the effect of the Node vj feedback?
E
This simulation compares the Va/2 Nodes, with and without the connection to the Vj Node.
I have reduced the Vin to 10mVppk in order to not limit the output voltage swing.
Try running the asc file, show your calculations for the expected gain of the Q3 stage.
What are your conclusions on the effect of the Node vj feedback?
E
Seems like feedback decreasing dc point and increasing ac swing a little bit. But in this case , as it is hard to calculate Rout at Q3 , how could I calculate or how should I approximate the gain ?
I did same thing but in DC analysis.
So the way of calculating gain is changing values in the simulation ?
Sometimes calculating the gain or frequency response and phase shift of a circuit is very complex.
It is easier to let the simulation determine these things for you.
I mean , if I would know approximate value of Rout , I can set Re. As It is not constant I am confused.
What is Rout?
What is Re?
Show a circuit schematic and indicate what you mean by Rout and Re.
Yes I know , I mean is this the True method of doing this ?
I doubt that there is any "True" method of doing this. All semiconductor circuit calculations require modeling and approximations.
For example, even modeling of a simple PN junction in a diode requires knowledge of the reverse saturation current and its temperature dependence.
