Driving Varactor Diodes

[andre_pblima]

Hello guys,

Well I'm developing a RF phase shifter controlled by voltage for 500 MHz and the main component is a varactor diode. I need take care to isolate the bias voltage of varactor from the RF circuit so that the RF performance is not impaired.
My doubt is: To isolate the bias circuitry of the RF signal from the RF to prevent any degradation of the performance circuit. I should use a resistor, inductor or both to do this.
And what could be the resistor's value or/and inductor's value?

Thank you

 

[tony_lth]

500M RF has an impedance about 50R in main stub, so a resistor value greater than 500R is OK, or an inductor which has the same inductance.

 

[jiripolivka]

Hello guys,

Well I'm developing a RF phase shifter controlled by voltage for 500 MHz and the main component is a varactor diode. I need take care to isolate the bias voltage of varactor from the RF circuit so that the RF performance is not impaired.
My doubt is: To isolate the bias circuitry of the RF signal from the RF to prevent any degradation of the performance circuit. I should use a resistor, inductor or both to do this.
And what could be the resistor's value or/and inductor's value?

Thank you

The real circuit choice depends on that if one or both varactor ends are "live". The better solution is when the varactor is connected to ground by one end. Then the DC bias is best connected by "bias tee" which utilizes a choke to insulate the RF circuit from the DC circuit.
The choke can be made as a "quarter-wave" type, it means cut a piece of thin wire (0.05" or 0.1 mm) approximately 1/4 wavelength long. Then wind it on 0.1" or 2.5 mm drill shaft o a similar round stock, and then expand it a bit before soldering into place. If ~15 cm is too long for 500 MHz , you can find any small SMD coil with an inductance larger than 0.1 uH or 100 nH.
The other choke end should be connected to a blocking capacitor (possibly a flat ceramic type or a feedthru) to bring the DC voltage to the varactor. If the drive voltage should change fast like 1 MHz, select the capacitor value for the best response but not smaller than 100-300 pF for 500 MHz.

If both varactor ends are "live", you will have to treat both ends as above.

 

[biff44]

If you use a resistor, then any currents rectified in the varactor will not be stiffly controlled by your analog driver, and you may end up the phase shift that varyies by input rf power

 

[chuckey]

" connected to a blocking capacitor " (jiripolivka), should read DECOUPLING capacitor.
Biff44 meant once you have isolated the RF on the varactor by your choke and decoupled it to earth, the DC should be fed into it from a low impedance source. If you are using any form of AM, then variations of its amplitude will result in the operating point of the varactor changing which will lead to all sorts of unwanted effects such as PM, change of mod depth, IPs. . .
Frank