First step in amplifier

[julian403]

I'm developing an amplifier with transistors. I don't know if It is recommended to put as first step an differential amplifier. What I have to do? Put the signal input in one input an the other input to ground. Like this



With bigger differential gain mode and with commun mode equal to zero.

 

[Audioguru]

The transistors are not biased so they do absolutely nothing.
The base of each transistor must be at a voltage more positive than their emitters and more positive than the current sink.
Do it like one of these two ways:

 

[kam1787]

First stage rather depends on the requirements: budget, impedance matching, gain, linearity, etc.

 

[LvW]

And - as a first step - I would not start with a current source symbol in the common emitter path.
Instead, use dual supply voltages (to satisfy Audioguru´s comment) and a common emitter resistor.

 

[julian403]

Thanks. What are RD+ and RD- in the transistor's inputs? RD- is the ground of signal? but the ground to signal must be ground power suplies which has Vcc and Vee.

 

[Audioguru]

RD+ and RD- are the differential input signals. Since it is a differential amplifier then the output is the difference between the inputs. If you ground one input then it is not a differential amplifier anymore.
If an input is connected to Vcc or to Vee then that transistor will be saturated or cutoff and will do absolutely nothing, and since the emitters of both transistors are connected together then the other transistor also will do absolutely nothing. The inputs should be connected to a DC voltage near half the power supply voltage which is 0V (ground) in a dual-polarity supply. A dual-polarity power supply has its 0V (ground) as the negative wire of the positive supply connected to the positive wire of the negative supply.

 

[julian403]

But I have only one signal audio and it goes to one input. What goes to the other?

 

[Audioguru]

A "balanced" signal has two wires, each with a signal but they are out-of-phase. When they connect to both inputs of a differential amplifier then common-mode interference is cancelled.
In a differential amplifier, one input can be for the signal and the other input can be for negative feedback.

 

[LvW]

But I have only one signal audio and it goes to one input. What goes to the other?

Julian, why do you intent - as a first step - to start with a differential amplfier?
If you want to amplify one singlke audio signal, why not using a simple common emitter stage?

 

[julian403]

If you want to amplify one singlke audio signal, why not using a simple common emitter stage?

Because I want to do it right and i saw always amplifier which use as first step a differential amplifier. I want to make an stereo amplifier but i just analyzing one.

The problem I have is that the books do not teach this staff and teacher too. They just teach you how to calculate the commun gain and the diferential gain but not where to use this staff. But I will try.

 

[Audioguru]

Because I want to do it right.

Then why don't you first learn about the details of a transistor, like how to properly bias one?
You posted a circuit showing a differential pair of transistors that have no bias so they do absolutely nothing. A transistor must be properly biased so that it conducts some current then its output can swing up and down or positive and negative like the input signal but with amplification.

EDIT: Do you understand why I added a negative supply and/or added resistors to the transistor the bases of your circuit?

 

[julian403]

You added resistor on base to polarize the transistor and put it in active zone. So in colector can flow current Ic=βIb . And a negative tension to use the current source, because it can be a current mirror (for example).

Now I know I must to polarize the base to have a current base. But, In one input there is the signal audio and the other? the ground signal?

 

[Audioguru]

You added resistor on base to polarize the transistor and put it in active zone. So current can flow in collector: Ic=βIb.

No.
In the example where I used a single positive supply voltage (no negative supply), I added two resistors as a voltage divider to feed a "half-the-supply voltage" to the base. Then the range of Beta and the range of base current do not affect the biasing.

And a negative voltage for the current source, because it can be a current mirror (for example).

In the example where I added a negative supply for the current source the bases can be biased with simply a resistor to 0V.

Now I know I must polarize the base to have a base current. But, In one input there is the signal audio and the other? the ground signal?

The other input should be the inverting input and can have a voltage divider from the output of the amplifier for negative feedback.

Please learn about negative feedback. An audio amplifier or an opamp have an extremely high voltage gain of 200 thousand times or more. The negative feedback reduces the gain to what you want and reduces the distortion. Negative feedback from the output also sets the output DC voltage so that it can swing symmetrically up and down the maximum amount possible.

- - - Updated - - -

How much power do you want from this audio power amplifier? What supply voltage will it use?
Have you learned about the "classes" of amplifier output types: class-A, class-AB or class-D? Which class will your amplifier use?

 

[julian403]

I have a +50 0 -50 AC transformer with 6 Ampere. I'm going to use (as power) a class AB. Where it's the circuit.



As Tip142 and Tip147 has a gain of 5000 to have a Ic=5 [A] it needs Ib=1 mA. So, as a before step I'm going to use a commun emitter but I wanto to use a differential amplifier too, to delete noise.

 

[Audioguru]

The TIP142 and TIP147 have a maximum allowed collector to emitter voltage of only 100V so they will be instantly destroyed when you turn up the volume because then they will have almost 140V.
If your transformer is only 35V-0V-35V then it will produce a supply that is plus and minus 48.5VDC. Then the output darlingtons will not be destroyed by too much voltage.

Assuming that your speaker is 8 ohms and that maximum output has a peak voltage of 45V then the peak current is 45V/8 ohms= 5.6A. The maximum continuous undistorted output power will be 127W and the darlingtons will each heat with about 60W. Then you will need a HUGE heatsink and probably a fan to help cool them.

The current gain spec of 5000 is for "typical" darlingtons that you cannot buy, you get whatever they have which might be with the minimum rated current gain of 1000. Then the driver transistors must supply at least 5.6mA.

I do not know why you show 2M in series with two 100k resistors in series with another 2M resistor because their current will be almost nothing. The 100k resistors are shorted resulting in severe crossover distortion. If the 100k resistors are not shorted then their current is much too low.

Your darlingtons have nothing to prevent" thermal runaway". The base-emitter voltage of transistors drops when they get warmer so your darlingtons will conduct more current which causes more heat which causes more current which causes more heat which causes more current etc until they melt. Audio amplifiers use a few series diodes or a small transistor bolted to the heatsink to automatically adjust the bias voltage preventing thermal runaway.

You have extremely high value emitter resistors of 10,000 ohms. Audio amplifiers use only 0.22 ohms to avoid too much voltage loss. With a peak driver voltage of 46V then the peak emitter voltage will be about 44.6V then the 8 ohm speaker will have a maximum peak voltage of only 0.0367V and its maximum undistorted continuous output power will be only 0.00008W which is almost nothing.

I think you must learn the basics of transistors and look in Google Images at some audio amplifier schematics.

EDIT: with the amplifier producing 127W and heating with 120W then the total power from the transformer will be 247VA. The transformer must be rated for 3.5A for one amplifier or 7A for a stereo amplifier.

- - - Updated - - -

A differential amplifier cancels interference noise only if the signal is balanced. Most signals are not balanced.
You are using a differential amplifier so that the second input can be used for negative feedback.