VCO linearization question!

[hawsona]

linearize vco

Now I have to linearize the VCO output frequency! Use digital method or
RF method!
but I don't know how to change the VCO schemtic to make VCO linearizing!
I know one method is use back to back varactor! But it's not very ideal!

Can any one talk with me for some RF design question?
I will very glad to contact with you!

my mail: carmine_rene@hotmail.com

 

[HBK]

vidmar vco

How abt a ring oscillator?

 

[FANT]

transistor reactance modulator

I have had the same problem with a VCO working in the low UHF band and i have solved it usin a linerizing diode like a 1N4148.
The result is good because on a swing of about 10 % of the center frequency the deviation change no more than 2 %.
Otherwise use ( if the frequency is much higher ) the VCO used by Matjaz Vidmar as 1st LO on his spectrum analyzer ( he use 6 varicap ! ).
I have build it and is very linear.
If anybody can tellme how to upload a schematic ( *.jpeg, for example ) i will send you the one i have used and the one of Vidmar.
Bye.

 

[mike_bihan]

reactance modulator with bjt

would you like to share the schematic with me?
or, it will be helpful if you can tell me the name of the paper described.

Thank you!
han.bi@sina.com

 

[flatulent]

linearized varicap

The back to back varicaps are for improving the phase noise by keeping the diodes from becoming forward biased over the rf cycle.

The linearization methods preceed the varicap by a network with a nonlinear in-out characteristic. This can be done with many methods, such as using diodes.

One problem with this is the extra phase shift through the network which must be taken into acount if the vco is being used in a PLL.

Another method if you are modulating is to demodulate the signal and apply feedback to your audio input stage from the demodulator.

 

[g579]

linearizing varicap

About the question of using analog or digital linearization.
If the VCO control signal is a slow one (or a narrow band PLL) the easiest way is using diode linearizer with op.amp.
Digital linearization is simple too and can be used in slow anf fast cases too. With fast A/D converter to the VCO sweep and an EEPROM lookup table you can switch (regulate) the gain of some aplifier circuit, carriing out the linearization.

 

[strabush]

opamp linearizer

From my experience, if you use a hyperabrupt varactor (or two in series) as the ONLY capacitor in the tank LC resonator, you will get a constant frequency / control voltage ratio, for a wide frequency range. In order to implement such a circuit you need a circuit that the feedback is done through the inductive part of the tank, like a Hartley oscillator topology.

 

[edf]

linearizing vco

I have designed a VCO covering 650-1100MHz in Colpits configuration with varactor-inductor on base and second varactor in emitter that had good linearity. The emitter varactor was in series with fixed value and only affected low end of tuning range.

 

[g579]

tube voltage control oscillator

strabush and edf both are right too.
These things may also give you a good enough result.
The freq - voltage characteristics can be modeled by EDA tools (even linear simulation may give acceptable results)

 

[yingyang]

linearization circuit diode

Can VCO be implemented without a diode or varactor????

 

[strabush]

linearization of varactors

Attached is an article about controlling the frequency of an oscillator without varactors:

 

[g579]

vidmar - spectrum analyzer update

For smaller freq. changes sometimes changing the CB capacitance of a BJT transistor (by changing the collector voltage) may result a VCO without varicap.

 

[edf]

varactor linearization

It is possible to design varactoless VCO's that can be tuned over a 15% bandwidth in the VHF range if you use a transistor designed for forward AGC such as the MPSH30 or 31. The B-E capacitance is current controlled. Use colpitts without lumped cap across B-E for max deviation.

 

[zorro]

rf linearizer

Can VCO be implemented without a diode or varactor????

A very effective and old method is the reactance modulator, called also reactance tube or reactance valve because it comes from the epoch of vacuum tubes.
Search for it.

The idea is:
Imagine an ideal bipolar transistor (it have to be polarized, of course) with an admittance Ybc between B and C, and an Ybe between B and E, and Ybe>>Ybc. The impedance presented by a transistor between C and E is Yeff=Gm*Ybc/Ybe, where Gm is the transconductance of the transistor.

The transconductance depends on its collector current, so varying Ic you can control Yeff.
Put a capacitor between B and C and a resistor between B and E (R<<omega*C). You have a variable capacitor.

(You could make a variable inductor as well changing R with C or putting L instead of C, but always Ybe>>Ybc. The combination given above is the most practical).
Regards

Z

 

[yingyang]

oscillatoer linearizer circuit

Thanks pal for the idea.

However, I am looking for techniques in RF and microwave frequency range. In addition, 1 am looking for a bandwidth of about 40%.

Any idea???

 

[zorro]

question and solution vco

Yingyang, You need a huge fractional bandwidth.
You can modulate narrow-band FM and then apply multiply-and shift method: frequency multiplying multiplies the frequency deviation, and down-conversion conserves it, increasing the fractional deviation.
Regards

Z

 

[g579]

linearization opamp diode

yingyang,

If price is not the major concern YIG (Yttrium - Iron - Garnet) oscillator may be a solution. Every handbook on MW has references about that.