# Linearity in amplifier stages.

#### Rohit156

Joined May 18, 2018
7
How linearity is maintained to i/p to o/p in amplifier with considerable gain at cost of efficiency and stability?

#### dl324

Joined Mar 30, 2015
10,501
Welcome to AAC!

Is this school work?

#### Rohit156

Joined May 18, 2018
7
Welcome to AAC!

Is this school work?
Not at all, just curious to know about it.

#### ronsimpson

Joined Oct 7, 2019
607
You just bias the amplifier so it operates in its linear region.
OR you give the amplifier 1000x more gain than it needs and use feedback. Here the gain is set by the resistors and not the internal errors of the amplifier.

#### Rohit156

Joined May 18, 2018
7
So far I learned that for keep it input linear it must be coupled to amplifier such that it will be less than Vt. (Considerable bjt device)
Or consider for cmos with condition of I/p coupling far less than Vgs-vth .
Now question is
How linearity is maintained from one stage to another in process of archiving higher gain?

For eg : if suitable input is applied to first stage of amplifier unit let's assume it have considerable gain such that it will surpass our approximation that I/p must be far less than Vt, so how further stage keep in track so that it will more will have more gain than previous stage with less compromisation with linearity.

Last edited:

#### dl324

Joined Mar 30, 2015
10,501
How linearity is maintained from one stage to another in process of archiving higher gain?
Most bipolar opamps have 3 stages. Each stage is biased to operate in it's linear region. In most opamps, the second stage is compensated for stability.

#### Rohit156

Joined May 18, 2018
7
OR you give the amplifier 1000x more gain than it needs and use feedback. Here the gain is set by the resistors and not the internal errors of the amplifier.
View attachment 203563
[/QU
Most bipolar opamps have 3 stages. Each stage is biased to operate in it's linear region. In most opamps, the second stage is compensated for stability.
I would like to know how further stage would be kept in biased in linear region so that previous amplifier o/p swing will be tolerable to a limit so that it doesn't steer into another region? Are there any other special amplifier configuration to handle this?

#### Rohit156

Joined May 18, 2018
7
So far I learned that for keep it input linear it must be coupled to amplifier such that it will be less than Vt. (Considerable bjt device)
Or consider for cmos with condition of I/p coupling far less than Vgs-vth .
Now question is
How linearity is maintained from one stage to another in process of archiving higher gain?

For eg : if suitable input is applied to first stage of amplifier unit let's assume it have considerable gain such that it will surpass our approximation that I/p must be far less than Vt, so how further stage keep in track so that it will more will have more gain than previous stage with less compromisation with linearity.
Assuming it has already properly biased to be in linear region with AC coupled from I/p , curious to know about behavior amplifier o/p swing tolerable to further stage to operating properly with less compromisation with linearity.

#### crutschow

Joined Mar 14, 2008
24,710
If you want HIFi level linearity (<1% harmonic distortion) in a transistor stage, then the state usually has some negative feedback.
This also stabilizes the gain and makes it less dependent upon the individual transistor current gain.
An easy way to do this is to have an emitter (BJT or source (MOSFET) resistor that is unbypassed.
To allow for independent setting of stage bias and the amount of feedback, two resistors are often used, with only one of them AC bypassed with a capacitor.

Below is the LTspice simulation of a single-stage BJT amp illustrating the difference in distortion and gain.
It can be seen that the V(out) yellow trace with an unbypassed 100 ohm emitter resistor has noticeably lower distortion than when it is shorted out by S1 (blue trace) where the top of the trace is noticeable flatter than the bottom of the trace.
The tradeoff is that the gain is about 4 times less with the feedback (input signal is 4 times larger).

This distortion is mostly caused by the logarithmic relation between the base input current and voltage as shown in the bottom base [ib(Q1)] current traces.

#### Rohit156

Joined May 18, 2018
7
Thank you for you contribution for making discussion more meaningful so far.
Still one question I would like to ask,
How tradeoff between linearity and stability can affect amplifier performance?

#### crutschow

Joined Mar 14, 2008
24,710
How tradeoff between linearity and stability can affect amplifier performance?
Define "performance" which is a very general term.