selection of RF inductors

I am designing an RF transciever board. I have quite a few inductors in the RF section. But, I am a novice in RF circuits and hence am not so aware of the parameters of importance. My usage of inductors is for switching power circuits and I usually just check the DCR and the saturation current. Then based on the formula for inductance I select it.

But, when it comes to RF inductors, how do I select them ? I have read that there are 2 more parameters viz - Q factor and SFR. The Q factor , I understand is Q = Xl/Rl i.e - Impedance of the Inductance/Resistive component of Inductance.

Now, what is the SFR ? I find confusing definitions online. I would appreciate if anyone could point me to a tutorial on this please.

Also, how does the Q factor and SFR affect the performance of the circuit ?

• No time for a complete answer, but go google "interwinding capacitance". – The Photon Sep 16 '15 at 16:22

No it's SRF (not SFR) i.e. self-resonant frequency and it's the frequency that the inductor naturally resonates at due to self capacitance of the windings.

For RF stuff Q factor is important to note especially when they tell you at what frequency the Q factor was measured - if it is at a high frequency and the value is OK (20 upwards is a rule of thumb I guess) then you can be sure that at that frequency the eddy current losses in the core are reasonably low.

Ideally, for critical uses (such as oscillators) having a graph of Q factor versus frequency is quite useful because, for a given value of inductance you'd expect Q to rise proportional to frequency - at some high frequency this graph will begin to flatten and this usually tells you what the max frequency of use is.

• I am looking at an inductance (part number MPCI 10 000 010). The parameters are - Inductance - 0.01uH,Qmin - 60,Test Freq - 150MHz,SFR - 2000MHz. Now, say I use it in an RF front-end circuitry of 2.4GHz. From photons comment below, I can safely deduce that since the SFR is 2000MHz, it would be unwise to use it in a 2.4GHz system,right? But, how can I analyse the aptfulness(is that a word!) of the Q factor in this case ? Qfactor = Xl/R. How will it affect the performance of any system ? – Board-Man Sep 17 '15 at 9:19
• So, Qfactor = (2*pifL)/R, where R- internal parasitic resistance of the inductance or DCR. The rating of DCR = 0.025Ohms. The Qfactor of 60 is 60 at 150MHz. So from the above values I can find Q. Now , Q = 0.9419. So, how come they get it as 60? The data sheet I refer is -dalitech.com/Resources/MicroSpire_Catalog.pdf page 18 please. – Board-Man Sep 17 '15 at 9:37
• 10 nH at 150 MHz has a reactance of 9.42 ohms therefore Q = 9.42/0.025 = 377 BUT there are frequency-proportional-core-resistive losses that in effect increase the perceived series resistance - this reduces Q. There are also "core effects" that shunt the inductance as frequency rises and of course this also reduces Q - this also partly contributes to a non-infinite SRF. SRF is also contributed by inter-capacitance between windings. Skin and proximity effects increases the AC series resistance of the winding and this reduces Q too. – Andy aka Sep 17 '15 at 10:13
• What you want is a Q factor that rises linearly from zero (or low) frequency and remains fairly linear past your maximum frequency operating point. If it is still a straight line at your operating frequency than you can be sure that its performance is just as good at low frequencies as at your maximum frequency BUT this is usually a tall order so, if I'm designing (say) a colpitts oscillator I look at the Q graph and estimate the reduction in L (due to SRF parasitic capacitance) and I also estimate the increase in resistance and work out what problems this might cause me. Easily done in spice. – Andy aka Sep 17 '15 at 10:25
• – Andy aka Sep 17 '15 at 10:42

SRF is self resonant frequency. It's the frequency where the inductance resonates with the parasitic capacitance between one coil and the next (called interwinding capacitance). Including the main parasitics, a real inductor acts something like this circuit:

simulate this circuit – Schematic created using CircuitLab

At some frequency, the admittance due to the capacitance cancels the admittance due to the indcutance ($j\omega{}C=-1/(j\omega{}L)$, which gives a net zero admittance (slightly modified by the resistive parasitic).

Above this frequency, the component behavior is dominated by the capacitance rather than the inductance, which means you need to choose an inductor with SRF above your operating frequency if you want it to act like an inductor instead of a capacitor.

• so, what u mean is that at the SRF the admittance of the 2 components viz - the inductance and its parasitic capacitance is the same. And, if the frequency of the signals is increased, the capacitor is more dominant. What do you mean that the capacitor dominates? What behaviour does it show?Does it mean that at greater frequencies the admittance of the capacitance is more than that of the inductor ? So, basically, the whole purpose of using an inductor is killed. Is my understanding right please ? – Board-Man Sep 17 '15 at 9:44
• Below resonance the reactance is positive and increasing with frequency (as expected for an inductor), above resonance the reactance is negative and decreasing in magnitude with frequency (like a capacitor). It behaves like a capacitor and not like an inductor, so if your circuit calls for an inductor it will likely not work correctly. – The Photon Sep 17 '15 at 16:00