Inductor Q at 100 kHz and 1.2 MHz

I measured some inductor Q's at 1.2 MHz - results: for 33 uH Q is 25-30, for 3.3 uH and 10 uH, Q is between 15 and 20. So, the 33 uH inductor has the highest Q, but it is no longer as drastically a difference as at 100 kHz where 100 uH has a Q of 7 to 8, and the 3.3 and 10 uH inductors have a Q of about 2 to 3. Measurement technique: place the inductor L parallel to a capacitor C so that resonance is close to 1.2 MHz, add a series resistor Rs and apply a signal at the resonance frequency. The Inductor has parallel resistor Rp which at resonance forms a voltage divider with Rs - knowing Rs and measuring the input voltage and the across across Rp, the value of Rp can be determined. Example, for 33 uH, a 560 pF capacitor resonates close to 1.2 MHz. series resistance of 6.8k results in a voltage divider with gain of about 1/2, thus Rp is equal to Rs which is 6.8 k. Q is given by Rp /(omega x L) thus Q is 27. Since this inductor was measured at 100 kHz as having a Q of 7.9, it turns out the ratio of Q at 1.2 MHz to Q at 100 kHz is exactly the square root of 1.2M/100k. The smaller inductors seem to have a higher ratio - probably related to them being further from their maximum Q. At frequencies above the maximum Q frequency, Q will drop and go to zero at the self-resonance frequency. Note at low frequency it would be expected that series resistance would be constant and in such a case, Q would be proportional to frequency.

Measured Inductor Q at 1.2 MHz, 100 kHz
Inductance Capacitance Resonance Series R Voltage Parallel R Q(1.2 MHz) Q(100 kHz) Ratio
33 uH 560 pF 1.17 MHz 6.8 k 0.5 6.8k 27 7.9 3.418
10 uH 1.8 nF 1.26 MHz 1.364 k 0.48 1.26 k 15.9 2.1 7.57
3.3 uH 5.6 nF 1.17 MHz 470 0.5 470 18.3 2.9 6.31

Testing Inductor, circuit, and equivalent circuit at resonance.



Effect of filter (with center at f0, bandwidth B) on two sidebands, fL and fU where carrier is tuned until fL is at the center of the filter.