Eddy current and metal testing with an LCR meter

[antigua]

I noticed that DER DE-5000's are going for as low as $80 on Amazon, and it's an LCR meter that does it all, measures a pickups inductance, capacitance, DC resistance, and Q factor. 

I've known LCR meters report on the Q factor for a long time, but I wasn't sure how to make use of the LCR meter's Q factor indication, and I'm still uncertain about it, to a large extent. I compared a Strat style pickup and a PAF style humbucker and noted that at 1kHz in "equivalent series resistance" mode, testing inductance, that the Strat pickup measured roughly the same Q factor, around ~2.3 as the steel poled PAF, but bode plots show that Strat style pickups have a much higher Q factor at resonance, and that's something that can be heard in situ. So what does the Q factor at 1kHz say about the function of a pickup?

I set the measure to 1kHZ, because 100Hz is too low to cause significant eddy currents, and 10kHz is beyond the resonance of the coils. I also tried comparing Strat pickups with AlNiCo 3 and AlNiCo 5 pole pieces, which were almost identical Lollars, though the AlNiCo 5 pickup has a higher DC resistance of about 6.5k ohms, versus 5.5kohms. The Q factor reading of the AlNiCo 3 pickups was higher than the AlNiCo 5 even though the detailed bode plots consistently show a lower Q factor with AlNiCo 3 pole pieces. Maybe the lower DC resistance is to blame, but if that's true, it means the Q factor is too heavily dependent on  

On the other hand, I tried testing some nickel silver and brass covers on the PAF type pickup, and the Q factor shifted in an expected way, so much so that it seems to be a reasonable means to ID nickel silver versus brass covers:


A Seymour Duncan SH-1N showing a Q of 2.36, and an inductance that is innacurate at 1kHz. 





With a Seymour Duncan nickel silver cover over the pickup, the Q drops slightly to 2.19




Finally, a brass Epiphone cover causes the Q factor to 1.85, a very obvious drop. 




So if nothing else, this might be useful for some rapid QC. I haven't tried this with a Telecaster neck pickup, with and without brass covers, but I would expect very similar results. 

[ms]

Getting a meaningful Q factor for a pickup is not  easy.  1KHz is a bit low: eddy current losses increase with frequency, and also 1 KHz is below where the human hearing is most sensitive and so the tonal effect of Q is most relevant somewhat above 1KHz.

In my impedance measurement system, I specify Q at 3 KHz. (You could argue that that is a bit high, but it is a frequency where there are significant differences.)  To do this accurately requires measuring the impedance as a function of frequency, and then removing the effect of the capacitance from the impedance by a fitting technique.  Then you can use the real and imaginary parts of the  result to compete Q.

[aquin43]

My approach is to treat the corrected, i.e. current source driven, impedance curve as the gain of a bandpass amplifier and compute the Q from the bandwidth and the centre frequency e.g.

which is a Phatcat Bridge pickup with an added 500pF to bring the resonant frequency down.

I'm not quite sure of the meaning of Q of the coil in this context, though. Because of the frequency dependent losses, the measured Q of the Phatcat remains at roughly 2 over more than a decade range in frequency, very far from the Q proportional to frequency expected of a normal inductor. In addition, the Q of the transmission path from the string will appear lower because of the further filtering effect of the eddy current losses and the shielding effect of the cover.

Arthur