Hello, problem with my circuit Audio amplifier in class AB

[beleanmarian]

Hi, I have a project to do in college, AB class audio amplifier. I did a high pass filter, low pass filter, signal amplification and with the help of a potentiometer to get between AO min and AO max.
The problem is in class AB where I do not know how to calculate the values for resistors or what types of transistors to use.

And the voltage on the load resistance

It's not going well

There are 3 signals because I set the potentiometer to 0 0.5 and 1.

Please help me.

 

[Nigel Goodwin]

What is your exact assignment?, there's little there of use in an actual useable audio amplifier.

 

[beleanmarian]

The theme of the project is: Audio amplifier. As data we received the value of the input signal, the minimum and maximum value of AO between 3V and 12V and two frequencies, one minimum, one maximum and the RL = 4 ohm. I have to set the signal to oscillate between 3 and 12V, which I did, but the current amplification(class AB) part doesn't work. Nigel Goodwin
We're just simulating, so I probably have to show through this circuit what the audio amplifier does.
I have to send the report with the schematic and the dimensioned components.
Basically we had to modify an input signal and put it on the power amplifier in class AB.
Sorry for my bad english

 

[Nigel Goodwin]

Have you ever looked at the schematic of an audio amplifier? - the circuit posted doesn't really look like one.

 

[beleanmarian]

So, I talked to the teacher, we have to choose a power, we have the voltage and then I disconnect the current and from there to resize the resistors.
The "speaker" is the load resistance.

 

[audioguru]

Your second opamp is driving the dead short of a 1 ohm resistor R7. So you and your simulator do not know the output current limits of an opamp.
Since the 4th opamp feeds between the diodes then the opamp turns OFF the transistors and R8 and R9 turn them on but not enough since when the output swing becomes higher and more base current is needed then the base currents from R8 and R9 become less.

Your lowpass filter cuts ALL medium and high audio frequencies, why?

 

[beleanmarian]

I am a beginner in the world of electronics, I received this topic .. if you can help me with a circuit diagram based on filters with the respective frequencies and the signal to be between AO min and AO max I would be happy And the current amplification

 

[audioguru]

What is the value of R7? It cannot be only 1 ohm.
The output swing is limited to only +4.2V and -4.0V because obviously R8 and R9 cannot provide enough base current for the transistors.
Why doesn't your teacher help you?

 

[beleanmarian]

the teacher does not help anyone unfortunately, he claims to know

 

[beleanmarian]

After the filter part, I have to get with the signal between 3sinwt volts and 12sinwt. I calculate the amplification as vout / vin, ie for k = 0 at the potentiometer the minimum amplification is 3 / 200uv which is equal to 1+ R4 + kr5 / r3 where I take the values of the resistors and for k = 1 maximum amplification, 12 / 200uv where do I get the value of the potentiometer,right?
R7 has the value 1 and the other resistance has 1k because I wanted to get with the amplification in the field of volts

 

[audioguru]

Why doesn't your teacher teach you this stuff??

Please do not send me "conversations" that are not attached to this thread.
3 (Blank) might be 3V? Then 3V/200uV= 3V/0.2mV= a gain of 15 thousand times. 12 (Blank)/200uV= a gain of 60 thousand times.
What produces an input signal that is only 200uV? The output from a low level microphone is 50 times higher at 10mV. The opamps will amplify their own noise when the gain is more than about 2 thousand times. The TL081 opamp is for "general purpose". With such a high gain you need "audio" opamps that produce much less noise. An LM833 dual audio opamp is good.

The datasheet for your TL081 opamp, an LM833 and most other opamps shows a minimum load of 2k ohms. The 1 ohm input resistor of your third opamp is the load of the second opamp then R7 must not be less than 2k ohms, so use 2k ohms. If the feedback resistor is 300k ohms then the gain is 150 times and the gain of the 4th opamp must be from 15k/150= 100 times to 60k/150= 400 times. Can you calculate the values for R3, R4 and R5 so that the gain of the 4th opamp is from 100 to 400?

Your opamps are all DC-coupled together and therefore amplify DC voltages. All opamps have an input offset voltage that is shown on their datasheet. It is 5mV max for an LM833. If your 1st, 2nd or 3rd opamp has an input offset voltage of 5mV then the output of the 4th opamp could be trying to go to 5mV x 60k= 300V but can't then it will be saturated and will not amplify audio. Therefore the input of the 3rd and 4th opamps must have input coupling capacitors. The output of the 4th opamp must have a coupling capacitor to feed AC to the diodes in the output stage.

The output stage has a low output voltage swing because R8 and R9 do not provide enough base current. Reduce their values so that the output swing is higher but not too low that the 4th opamp cannot drive them.

 

[unclejed613]

https://www.analog.com/en/technical-articles/paralleling-amplifiers-increases-output-drive.html follow the example there. and use that summing point to drive the output stage. if you need voltage gain, the circuit using all four op amps in a quad op amp is what you are looking for, set the gain of all the op amps at the same amount using the formula Av=( R2 / R1)+1

 

[beleanmarian]

I finally solved the problem, this is the final circuit. I only have one more problem, why is the first sinus lower? Do I have to change the values from the filters? I noticed that after the filter goes up the first sine goes down, from where I know what values I should put, to like the amplifier, not to change the signal.
I want to mention that I changed the FREQ from 1k to 374, this representing the central frequency. At 1k it behaves ok, the first sine is perfect, but at 374 (center frequency) it changes its shape

Only from 5ms does it stabilize

 

[schmitt trigger]

Your circuit does not have global feedback. There will be significant distortion.
The output uses a traditional Darlington on top, and a Sziklai pair on the bottom. Their DC bias would be uneven.

 

[audioguru]

U2B, U3B and U4B are doing absolutely nothing and can be replaced by a piece of wire.

Let your 'scope show the DC offset voltage at the transistors side of C3 which causes the output to have some DC offset when the power is first turned on, then C3 charges in 5ms. It is caused by the DC gain being much too high.

Most audio amplifiers use one opamp as an AC preamp. Then the power amp uses only one opamp (yours has 6 opamps) with AC and DC negative feedback
from the output to the input for extremely low distortion. Yours has no negative feedback for the output transistors.

Your filters are extremely simple with a gradual slope and a very broad corner. The gradual slope causes their effect to be minimal and the broad corner causes them to affect frequencies 1/5th and 5 times their calculated -3dB frequency.

 

[beleanmarian]

I saw that after the filter goes down it does this thing...
Is there no solution? To add something, to return to normal

 

[audioguru]

I did not notice that this is an unreal simulation, you did not build it.
I and the simulation also did not notice that the Sziklai pair behaves as a single PNP transistor, then your circuit needs only 3 diodes. With 4 diodes there will be massive output transistor current but the simulation software ignores it.

 

[beleanmarian]

At 1,5kHz (VSIN set)

At 374(center frequency)

Filter low pass

high pass

there is the problem..

If I change the resistance from the filter high pass to 10k it no longer starts from 0, as you can see in the image. If I leave it at 1k, it is centered at 0.

 

[audioguru]

The higher DC offset voltage in R4 caused by increasing its value is obviously caused by the input bias current of U3A. It would be cancelled if the combined value of R5 parallel with R6 equals the new value for R4.

But the tiny offset voltage would be covered up by the real input offset voltage of an opamp being amplified by the very high DC gain of your circuit.
Your simulation ignores the real input offset voltage of an opamp.

 

[beleanmarian]

So, I understand that there is no way for the sinus to look good