Topband: Measuring Inductor Q

[Rick Karlquist]

Jim Brown wrote:
> Study the test setup and discussion about it in the two pdfs on my
> website about RFI for Hams and Coaxial Chokes.
>
> http://audiosystemsgroup.com/publish
>
>>
> My pdfs show a good way to do it with simple equipment. In essence, you
> connect the unknown impedance in series between a signal generator and
> an RF voltmeter, with a good 50 ohm or 75 ohm load across both the
> signal generator and the voltmeter. Then you measure the voltage across
> the 50 ohm resistor over a broad range of frequencies that extends from
> well below resonance to well above resonance and use the voltage
> divider equation to get the magnitude of the impedance. Then plot Z vs
> frequency in a spreadsheet.  Now you have the resonance curve of the
> unknown inductor, and you can curve-fit it with the equation for a
> simple parallel resonant RLC circuit to get R, L, C, and Q.

Jim is on the right track here.  I have a slightly
different approach.

What I like to do is make a parallel resonant tank
with the inductor under test and a high Q capacitor.
(Any NP0/C0G ceramic will have Q>1000).  I ground
one of the terminals of the tank and then very
loosely couple into the tank with a signal generator
by connecting a 1 pF capacitor from the signal generator
to the non-grounded terminal of the tank.  I then
connect another 1 pF capacitor from the non-grounded
terminal of the tank to a spectrum analyzer and observe
the 3 dB BW of the tank.  The Q is then given by dividing
this into the center frequency.  You could possibly
use the S meter of a receiver instead of a spectrum analyzer,
switching in 3 dB of attenuation and getting back to the
same place on the S-meter.  This does not require S-meter
calibration.  Operate around S5 where an S unit is typically
1 or 2 dB.  You can also use a 2 port network analyzer to
do this measurement.  I use an HP8754A network analyzer.
The key is to do a TWO PORT measurement some how.  You have
to verify that the 1 pF capacitor isn't affecting the Q.
If the tank capacitor is much bigger than 1 pF, you are
probably OK.  I have also used sub-1 pF "gimmick" capacitors.

I can get a reading this way that compares favorably to my Boonton
160A and HP4342A Q meters.  Above 70 MHz, beyond the range of
those meters, I must resort to the method above.

Rick N6RK