Current starved VCO cadence design

[destro98]

Hi,

I am trying to design a current starved VCO in cadence, i tried some circuit and it is oscillating, but i don´t know how to vary the input control voltage to get the oscillation frequency from maximum to minimum. i am attaching the png of my design. Any kind of help is appreciated regarding this concept.

Thank you.

 

[dick_freebird]

You want to feed "input" with a DC current source
if you want anything linear-looking. You can drive
it with a voltage source but should expect a response
between exponential and square-law in voltage-freq
transfer characteristic.

In a practical circuit you would probably take a control
voltage, run it through a linear-ish voltage to current
converter and push that current down "input".

 

[destro98]

You want to feed "input" with a DC current source
if you want anything linear-looking. You can drive
it with a voltage source but should expect a response
between exponential and square-law in voltage-freq
transfer characteristic.

In a practical circuit you would probably take a control
voltage, run it through a linear-ish voltage to current
converter and push that current down "input".

thanks for the help, but i can't understand what do you mean by "running control voltage through in practical circuit" are you saying we should use a DC voltage source as a control voltage?

 

[dick_freebird]

I'm saying that when you make a VCO as a "product"
(piece-part, IP chunk, whatever) the expectation is
going to be that it is (linearly) voltage controlled. Not
current, unless you advertise it as a CCO.

So then you would be tasked to take an input voltage
and do what you must, to get a reasonably linear
df/dV across the band of interest. This could be as
simple as a minimal op amp and current mirrors, one
of which drives a I*R feedback leg and another of which
drives the CCO core.

 

[BigBoss]

I think the input reference current changes the oscillator frequency.

 

[destro98]

I think the input reference current changes the oscillator frequency.

Yes, the input current source changes the oscillator frequency, when i change the current source value the frequency changes, but as it is a voltage controlled oscillators how can i know frequency variations with respect to the voltage variations. How can i know the voltage variation applied for the certain frequency? that was the main issue i am facing with this model.

 

[destro98]

I want to control the input current, but as it should be voltage controlled, i want to convert voltage to current and give the output current as the input, so that i can vary the voltage and can get the required frequency. Can you help me with how can i design a Vto I converter without using opamp, using opamp brings complications in cadence, can we design a converter circuit using MOSFET?

Thank You.

 

[dick_freebird]

Op amp brings complications?

Even a little 5-transistor buffer? Or maybe 7, since you
might like common mode range to include ground?

If you expect circuit design to be free of complications,
expect much disappointment.

 

[destro98]

Op amp brings complications?

Even a little 5-transistor buffer? Or maybe 7, since you
might like common mode range to include ground?

If you expect circuit design to be free of complications,
expect much disappointment.

Yes OP-amp is the easy one, but we have to design the conversion without the op amp as per our project.