AlNiCo pole piece comparison

[antigua]

A few years ago I bought some AlNiCo pole pieces from Addiction FX and recorded values of them with an LCR meter and bode plotter. A post about it is on the next somewhere, but I couldn't find it, so I just performed the tests again, this time a little better. Also, this time I've used Ken Willmott's integrator. 

One of the plots is of "degaussed A5", these are pole pieces that have about 3% of their full strength charge. They would be unusable as is. That and "air core" are included FYI, neither is usable as is.


Zoomed out:



Zoomed in:




Air coil:
Flux density: _ _ 0 Gauss
Inductance: _ _ _ 1.480 H @100hz
Q: _ _ _ _ _ _ _ _1.688 @1khz
Resonant Peak: _ _9.03 kHz
Peak Amplitude: _ 21.3 dBV
Test Voltage: _ _ 5.2 Vpp

AlNiCo 2
Flux density: _ _ ~700 Gauss
Inductance: _ _ _ 2.193 H @100hz
Q: _ _ _ _ _ _ _ _2.40 @1khz
Resonant Peak: _ _7.46 kHz
Peak Amplitude: _ 14.4 dBV
Test Voltage: _ _ 4.0 Vpp

AlNiCo 3
Flux density: _ _ ~550 Gauss
Inductance: _ _ _ 2.214 H @100hz
Q: _ _ _ _ _ _ _ _2.37 @1khz
Resonant Peak: _ _7.49 kHz
Peak Amplitude: _ 13.3 dVB
Test Voltage: _ _ 3.8 _Vpp

AlNiCo 4
Flux density: _ _ ~750 Gauss
Inductance: _ _ _ 2.180 H @100hz
Q: _ _ _ _ _ _ _ _2.38 _@1kHz
Resonant Peak: _ _7.46 kHz
Peak Amplitude: _ 14.3 dBV
Test Voltage: _ _ 3.7 Vpp

AlNiCo 5
Flux density: _ _ 1000 Gauss
Inductance: _ _ _ 2.016H H @100hz
Q: _ _ _ _ _ _ _ _2.23 @1khz
Resonant Peak: _ _7.80 kHz
Peak Amplitude: _ 15.9 dBV
Test Voltage: _ _ 4.05 Vpp

Degaussed AlNiCo 5
Flux density: _ _ ~30 Gauss
Inductance: _ _ _ 1.874 H @100hz
Q: _ _ _ _ _ _ _ _2.07 @1khz
Resonant Peak: _ _8.07 kHz
Peak Amplitude: _ 16.9 dBV
Test Voltage: _ _ 4.2 Vpp



The output voltage varied by pole piece set, but I wanted them normalized in order to compare apples to apples, so I recorded the "test voltage", which indicates how much drive voltage was required in the exciter coil in order to get them all to be even. A lower test voltage indicates that there was a stronger coupling coefficient, and vice versa. 

The main take away from this testing is that AlNiCo 2, 3 and 4 are virtually the same, with just slight differences between them, while AlNiCo 5 stands farther apart. Another take away is that the degaussed AlNiCo 5 varied significantly from the charged up AlNiCo 5, demonstrating how the magnetized state of the AlNiCo alters the inductance and the Q factor of the coil.

A member on another forum claims that these AlNiCo 4 sample might be bogus, and that AlNiCo 4 should more closely resemble AlNiCo 5. Unfortunately, there is not much data available about AlNiCo 4 relative to the other commonly used grades, so I don't have enough information on hand to confirm or deny that this is the case. In this case, the electrical values are very nearly identical to AlNiCo 2, but the residual flux density (Br) was noticeably higher. 

[stratotarts]

Note that the degaussed A5 did not change in the direction that would make them similar to the other alloys. So, if anyone suggests that degaussed A5 is an alternative to A2 or something, they would be wrong.

[antigua]

Jun 29, 2020 9:53:50 GMT -5 stratotarts said:

Note that the degaussed A5 did not change in the direction that would make them similar to the other alloys. So, if anyone suggests that degaussed A5 is an alternative to A2 or something, they would be wrong.

Those degaussed A5's were received uncharged, but received some charge by being in the presence of other charged magnets, so the flux strength was around 30 Gauss, or 3% of full charge. I should have made that more clear. They're not really usable in that state. Had they been charged to, say, 500 Gauss, they would probably split the difference. That being said, they are still farther from A2/A3 rather than closer.

It's true that the Q and inductance was less like A2 or A3, but I think if you just take the effects of magnetic strength into consideration, which I think is not really any difference from the distance between the pickup and the strings, the half charged A5 would be similar to A2 or A3 in that respect. When I did testing of the effect of pickup height on the recorded sounds, little adjustments have a visible impact over time, for example, inter-harmonic beating increased in frequency as the pickup came closer to the strings. Since that's a result of magnetic pull, that sort of difference would characterize a stronger magnet over weaker one, apart from the inductance and Q.

[antigua]

Jun 29, 2020 11:04:15 GMT -5 aquin43 said:

Were the inductances measured at a low frequency?

Arthur

100Hz, I'll note that also. 

I'll try to source more pole pieces from other places, and get a broader sampling. I see A2, A3 and A5 offered regularly, but I've never seen AlNiCo 4 offered in a Fender style pickup. 

[aquin43]

100 Hz inductances measured, so, from the air coil it is possible to work out the capacitance, which comes to 209p9.

Assuming that the resonant Q is high enough that the peak frequency is equal to ω0.

the inductances at resonance and their deviations from the low frequency value can be worked out.


           LF ind      Res         HF ind      Change %
AlNiCo 2    2.193 H     7.46 kHz    2.169H      -1.1%  

AlNiCo 3    2.214 H     7.49 kHz    2.151H      -2.83%

AlNiCo 4    2.180 H     7.46 kHz    2.169H      -0.53%

AlNiCo 5    2.016H H    7.80 kHz    1.984H      -1.6%


Arthur

[antigua]

Jun 29, 2020 11:34:59 GMT -5 aquin43 said:

100 Hz inductances measured, so, from the air coil it is possible to work out the capacitance, which comes to 209p9.

Assuming that the resonant Q is high enough that the peak frequency is equal to ω0*L,

the inductances at resonance and their deviations from the low frequency value can be worked out.


           LF ind      Res         HF ind      Change %
AlNiCo 2    2.193 H     7.46 kHz    2.169H      -1.1%  

AlNiCo 3    2.214 H     7.49 kHz    2.151H      -2.83%

AlNiCo 4    2.180 H     7.46 kHz    2.169H      -0.53%

AlNiCo 5    2.016H H    7.80 kHz    1.984H      -1.6%


Arthur

Thanks for calculating that out. 

I'm a little alarmed, because the DE-5000 in capacitance / parallel mode at 100kHz is showing 110pF, which is obviously wrong, so I did a 1MHz sweep, which takes over an hour to run, and low and behold there's a strong resonance right around 91kHz that confuses the DE-5000. I'm waiting patiently for someone to make an LCR meter that has multiple, or even arbitrary test frequencies, between 10KHz and 1MHz.