I am working on a pattern reconfigurable antenna and for this i need to bias the PIN diode from a battery through bias line. For this I need a RF choke inductance to isolate the ac from the bias line. I have seen different value of inductor is being used by different author. Is there any specific formula to select the RF choke to select for frequnecy band of 1.8 to 3 GHZ?
It depends a lot on the self resonating frequency of the inductor and the current flowing through it. Pick one with a self resonating frequency above 3GHz. Above 3 GHz it will be capacitive, so will pass AC current. What bias current do you need for the PIN diodes? You will also need to find one with a low series resistance, since this can introduce a loss which will increase the insertion loss.
You may install two or more chokes in serial.
That eliminates a number of issues concerned with the choke shunt capacitance.
Coilcraft have a good selection front end on their website and provide free samples. They also provide graphs so that you can see where the resonances occur. Don't forget that the pin diode impedance will change between states and you need the impedance of the bias line to be accounted for in both states.
http://www.coilcraft.com/apps/finder/rffinder.cfm
I trust you use PIN diodes to control the impedance seen by the parasitic element of the antenna and you use the coil to do low-pass filtering to supply the driving DC current for the PIN diode, and so would like to have the DC current to pass through.
Thus, the inductor/choke must be only a very small reactance and minimum possible resistance (together, they will mean a small enough impedance). At the same time, you want to block the RF from going leaking into the power supply (this would mean two consequences: a) this will change the impedance seen by the antenna, and b) normally, you do not want RF into your power supply.
Thus, at and near DC, i.e. at low frequencies, the reactance of the coil, X, should be low enough to permit the current. The X is expressed by X=j*2*pi*f*L [Ohm], where f is frequency in Hz and L is inductance in Henry (http://www.electronics-tutorials.ws/inductor/ac-inductors.html). As you normally cannot choose f, your only variable is L. For example, you could select X to be smaller than 1/10th of the smallest equivalent resistance of the PIN diode among the two states.
At higher frequencies, you normally want X to be large enough to ensure that the RF currents see the coil as a high enough impedance.
Unfortunately, any real coil has certain parasitic capacitance between its turns, after a certain high frequency, this capacitance starts to dominate. Together with the inductance, this starts to be a resonance circuit. You can see a calculator to estimate the frequency etc. at
http://hamwaves.com/antennas/inductance.html
As mentioned by Gavin, it is critical to select this self-resonance frequency to be sufficiently above the maximum frequency you need to operate at.
As mentioned by Alexey, connecting two or more chokes in series helps to reduce this parasitic effect. This is because these chokes are further away from each other than the distance between the turns of a single coil.
As mentioned by Dominic, you could get coils from for example Coilcraft. I trust they still supply free samples and even sample kits.
One possibly important note:
Having a large value of L means that the current through the choke may only change slowly. You can calculate the characteristic time as L/R, where R is the total equivalent resistance in series with the inductor. You can see the details on models and calculations at
https://en.wikipedia.org/wiki/RL_circuit or http://www.electronics-tutorials.ws/inductor/lr-circuits.html
If you need to control the PIN diode really fast, you may have to sacrifice the "blocking effect” and permit the current to move faster by having a lower value of the inductor. Alternatively, you need to also increase the R. However, increasing R will mean that in order to drive the same current through the PIN diode, you will need to have higher yet fast-controlled voltage from the power supply. This would be another design challenge.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators