Extech readings

[Magnut]

Looking at the inductance one could predict how the pickup may sound. Most pickups (single coils) are above 2 Henries. The higher you go the warmer it will sound but eventually the highs will drop off.

The way Q was described to me is that it is a factor of signal quality in that a higher Q is better. The signal is like a pendulum swinging. Release it and with a high Q it takes a while for it to come to rest. A low Q is like the pendulum swinging in a liquid, it quickly disipates. So a Q of 1 or lower would be a signal that drops of fairly quickly. Most single coils that I have tested are 2 and slightly above.

This is how I understand it so far. Someone else may have a better understanding than I do.

[Possum]

....

With single coils yes generally lower inductance lower Q it will be bright, higher it will be darker, but as I posted before when you get into a pickup that has alot of steel in it the inductance reading will throw you off. As I mentioned before take a single coil, take a reading, put a fat steel bar on the magnet poles and read again, inductance will DROP, and AC resistance will go UP. Its counterintuitive and will trip you up if you are designing pickups rather than just using existing designs....

[Mike Sulzer]

Mike,

Could you summarize the kind of things one can learn about a pickup coil from using the full features of the Extech LCR meter at both the 1K Hz and 120 Hz test frequencies.

The Extech measures AC resistance (AC-R), L, Q and D. I found that when I multiply the AC-R times the Q, I get a value very close to the coil's XL. Do you find this to be useful?

If the Q of a coil is below 1, how would you interpet that?

Joseph Rogowski

Joseph,

Q = 2*pi*f*L/R, that is, XL/R, where R is the AC resistance. So if you multiply Q by R you do indeed get XL.

If you make a resonant circuit at a frequency where the Q is less than 1, there is not a well-defined peak. For example, a guitar with the tone on zero.

The Extech gives you L and R at two frequencies. I do not think that this is enough to really determine the response of the pickup. L and R at the resonant frequency of the complete circuit containing the pickup would be more useful.

This is why I think it is better to measure the complete curve of impedance versus frequency, both for the pickup alone, and loaded by realistic resistance and capacitance as in the actual guitar circuit.

[Mike Sulzer]

With single coils yes generally lower inductance lower Q it will be bright, higher it will be darker, but as I posted before when you get into a pickup that has alot of steel in it the inductance reading will throw you off. As I mentioned before take a single coil, take a reading, put a fat steel bar on the magnet poles and read again, inductance will DROP, and AC resistance will go UP. Its counterintuitive and will trip you up if you are designing pickups rather than just using existing designs....

There might be a difference between "the inductance reading went down" and "the inductance really decreased". The third term in the error equation for Extech inductance measrurements (in percent) is 5d, where d is the dissipation factor. At 120 Hz, the Q of a pickup is low at 120 Hz, and so the d is high, leading to a significant error. I think you need to check this error before claiming that the inductance really decreased.

[Magnut]

I would agree that you must be careful about different situations. The variables are vast and each situation is different. After you get to "know" a coil you may be able to surmise how each of the readings may affect the outcome. However all that being said there is no substitute for the end result, damn the meter readings.

[Possum]

...

You need to try this and you'll see it in action, Joe Gwinn has commented before that this is a real thing. You do it, try it for yourself. I really don't care about the theory I know this is what happens with the meter so I know when to watch for it and to be aware that only going by the inductance reading is a mistake. you're really only going to see this if you are designing pickups and try different levels of steel loading. Its only useful if you are changing something out, otherwise if you are just measuring a pickup on its own you won't see it and it only happens when steel is added or increased in a pickup. It never goes UP when you add more steel, the AC resistance always goes up when adding more steel, so I pay attention to all 3 readings.

[Mike Sulzer]

You need to try this and you'll see it in action, Joe Gwinn has commented before that this is a real thing. You do it, try it for yourself. I really don't care about the theory I know this is what happens with the meter so I know when to watch for it and to be aware that only going by the inductance reading is a mistake. you're really only going to see this if you are designing pickups and try different levels of steel loading. Its only useful if you are changing something out, otherwise if you are just measuring a pickup on its own you won't see it and it only happens when steel is added or increased in a pickup. It never goes UP when you add more steel, the AC resistance always goes up when adding more steel, so I pay attention to all 3 readings.

Yes, there is no doubt that adding a conductor in the magnetic field of the pickup system increases the losses and alters the inductance. One would generally expect to see more effect at 1KHz than 120Hz. And one would trust the accuracy more at 1KHz since the Q of the coil is higher.

[David Schwab]

its not an error and its easy to do, take a strat pickup for instance and put a steel bar across the pole, sticking it to it, the inductance drops, the AC resistance goes UP....

Eddy currents perhaps?

[Mike Sulzer]

Eddy currents perhaps?

Yes; the problem is to measure their effects accurately in a high loss (low Q) system.

[salvarsan]

Yes, there is no doubt that adding a conductor in the magnetic field of the pickup system increases the losses and alters the inductance.

The thing I find odd is that the eddy currents in the steel overcome
any permeability increase that you'd expect to see as increased inductance.

This is probably why DiMarzio & Blucher did so much experimentation with ferrites
(soft magnetic and non-conductive) to fine tune the magnetic circuit, or so I infer
from their patents.

-drh

[Mike Sulzer]

It is surprising, but it really does decrease the inductance. I have seen the effect more at the high frequencies, and so I am surprised that it is significant at 120 Hz

The thing I find odd is that the eddy currents in the steel overcome
any permeability increase that you'd expect to see as increased inductance.

This is probably why DiMarzio & Blucher did so much experimentation with ferrites
(soft magnetic and non-conductive) to fine tune the magnetic circuit, or so I infer
from their patents.

-drh

[soundmasterg]

Joseph,

Q = 2*pi*f*L/R, that is, XL/R, where R is the AC resistance. So if you multiply Q by R you do indeed get XL.

If you make a resonant circuit at a frequency where the Q is less than 1, there is not a well-defined peak. For example, a guitar with the tone on zero.

The Extech gives you L and R at two frequencies. I do not think that this is enough to really determine the response of the pickup. L and R at the resonant frequency of the complete circuit containing the pickup would be more useful.

This is why I think it is better to measure the complete curve of impedance versus frequency, both for the pickup alone, and loaded by realistic resistance and capacitance as in the actual guitar circuit.

Yah, that would be nice...but how to do it accurately is the question without spending tons of money on test equipment?

Greg

[salvarsan]

... I think it is better to measure the complete curve of impedance versus frequency ... loaded by realistic resistance and capacitance as in the actual guitar circuit.

Maybe sorta like this?


http://www.salvarsan.org/pickups/dummy.html

It's an old discussion and lip service to it infrequently resurfaces.

-drh

[soundmasterg]

Maybe sorta like this?


http://www.salvarsan.org/pickups/dummy.html

It's an old discussion and lip service to it infrequently resurfaces.

-drh

That looks pretty cool and useable, but what other devices are necessary to test? An oscilliscope? Signal Generator?

Greg

[Mike Sulzer]

Yah, that would be nice...but how to do it accurately is the question without spending tons of money on test equipment?

Greg

It is not so expensive if you have a computer with a sound card or equivalent capability. See Joe's thread. The solution can use just one chip, but the operation is more complicated than an Extech.