What does "great linearity" means for a ring oscillator?

[littlerock]

Hello all. I just want to ask the significance or meaning of the phrase "good linearity" or "great linearity" as an advantage for the design of a ring oscillator. I am reading a paper about a bootstrapped ring oscillator and it says that "the proposed ring oscillator provides great linearity while the ring oscillator is controlled by power supply regulation". What does it mean?

 

[barry]

If you're using a voltage to control the oscillating frequency then the linearity is the relationship of voltage to frequency. In other words, if you plotted voltage on the x-axis and frequency on the y-axis, you'd get a straight line for "great linearity".

 

[LvW]

I rather think that te term "linearity" concerns the QUALITY of the oscillation signal (ideal: Sinusoidal). Each amplifier with a "good linearity" provides an output with a "good THD".
When the voltage-to-frequency control mechanism is specifies, we explicitely would read something like "Linear control function"

 

[barry]

The deviation of a sinusoidal signal from the ideal has nothing to do with linearity. A sine wave IS nonlinear! Maybe this is a language problem.

 

[vivekroy]

Is this the paper you are talking about?
https://ieeexplore.ieee.org/document/6123581
The abstract itself defines what is meant by linearity.

 

[LvW]

The deviation of a sinusoidal signal from the ideal has nothing to do with linearity. A sine wave IS nonlinear! Maybe this is a language problem.

"A sine wave is non-linear" ??? Do you know the definition of linearity ? Sorry, but this is nonsense.
The degree of non-linearity of an amplifier is given as "total harmonic distortion (THD)" .
An ideal linear amplifier would provide an ideal sinus function if the input is a sine function. Each deviation from the sine function at the output is caused by non-linearity.
As far as an oscillator is concerned - EACH "linear" oscillator must contain a certain degree of non-linearity. Otherwise it would not work (due to the oscillation condition).
In some simple cases, this non-linearity is provided by hard-limting (power rail). This is severe non-linearity that should be avoided. For this purpose, each "good" oscillator contains intentonally a non-linear part or subcircuit (amplitude regulation).

 

[barry]

I

"A sine wave is non-linear" ??? Do you know the definition of linearity ? Sorry, but this is nonsense.
The degree of non-linearity of an amplifier is given as "total harmonic distortion (THD)" .
An ideal linear amplifier would provide an ideal sinus function if the input is a sine function. Each deviation from the sine function at the output is caused by non-linearity.
As far as an oscillator is concerned - EACH "linear" oscillator must contain a certain degree of non-linearity. Otherwise it would not work (due to the oscillation condition).
In some simple cases, this non-linearity is provided by hard-limting (power rail). This is severe non-linearity that should be avoided. For this purpose, each "good" oscillator contains intentonally a non-linear part or subcircuit (amplitude regulation).

Sorry to disagree, but a sinusoid is a transcendental function, NOT a linear function. The linear response of an amplifier means if you plot the output versus the input you’d have a straight line, regardless of whether your input is a sine or anything else. The paper referred to in post 5 talks of linearity in the CONTROL VOLTAGE of the oscillator, as I previously mentioned.

 

[littlerock]

Thanks for the response everyone. Now I understand that the linearity mentioned here is about the linear relationship of the control voltage and resultant frequency of the ring oscillator. Thanks for the help.

 

[LvW]

I
Sorry to disagree, but a sinusoid is a transcendental function, NOT a linear function. The linear response of an amplifier means if you plot the output versus the input you’d have a straight line, regardless of whether your input is a sine or anything else. The paper referred to in post 5 talks of linearity in the CONTROL VOLTAGE of the oscillator, as I previously mentioned.

OK - I see that there wasa misunderstanding between us.
Of course, the sine function is - from the mathematically point of view - not a linear function. However, speaking of transfer characteristics or control functions in electronic circuits, I think it is clear that an (ideal) linear function would not distort a sinusoidal input.
On the other hand, I see that I have misunderstood the core of the question as, indeed, the control mechanism was meant (and NOT the quality of the signal).
This was not clear to me from the beginning (because - as I have mentioned - each oscillator needs a ceratin amount of non-linearity). This has caused the misunderstandiung.