Measuring the capacitance of an inductor or Inductors

[Possum]

...

Well permeability of screws varies wildly. Vintage screws are usually way more permeable than modern screws.

Can someone do an example? Lets say a typical humbucker 4 henries 4K resonant peak? What is the capacitance? Or does there need to be other measurements involved?

[belwar]

Im going to try some tests tommorow (or at the latest this weekend).. I had to buy another oscilloscope probe to do the test so I picked it up on the way home, and the pF caps came in from mouser today...

So far what I have planned is:

1 - Im going to see if the capacitance changes with different core materials.. I think what im going to do is use a slug coil and make slugs from different metals with different permeabilities.
2 - See how the capacitance is affected by different keepers
3 - Examine the capacitance difference between Polysol and vintage style stinky PE.. i'd do formvar but I dont have any.. I user a machine winder with a tension gauge so I could get the coils pretty damn similar in build / tension.

Plus the pickup testing for my own fun starts!

I'm going to test 3 different Wal Bass P/U's, about 8 Lane Poors, as well as my own P-90's and buckers. Plus a score of vintage pickups .. including a P-13.

[Possum]

....

I think my point was missed, capacitance shouldn't have anything to do with core materials other than there is either iron or steel there or not. The fault lies in using inductance as part of the formula because inductance is highly affected by what the core material IS. Then you're only measuring at a resonant frequency, well capacitance changes with whatever frequency you're at, so using the inidividual's resonant peak freq seems like a bad idea if you're testing several pickups, seems to me like you should use the same test frequency for every pickup, not the peak. You dont get inductance readings by reading at peak frequency so why would you do this with capacitance? It seems like these results based on this method doesn't give any useable information. Thats what I'm trying to get at, a method for using one standard to compare different pickups of the same type. So if you have a humbucker with a peak frequency of 3KH and one at 4khz and thats where you're measuring for capacitance you are using two different test frequencies, so you're not really comparing with identical test standards.

[David Schwab]

AC current flowing in the coil causes a fluctuating magnetic field which induces a current in the core. This reduces the effective permeability, leading to a reduction in the inductance, more at higher frequencies. My claim is that this effect is very small when the cores are open (unlike, for example, a toroid). Joe is not convinced. We will eventually get the matter settled.

Right, but if you factor in the permanent magnet, the eddies are more pronounced.

All you have to do if put a cover on a pickup to hear the effect. Since it's the highs that are affected, it would stand to reason that it is more pronounced as the frequency increases.

[David Schwab]

1 - Im going to see if the capacitance changes with different core materials..

Capacitance is the effect of the insulated conductors (turns of magnet wire) being in close proximity to each other. If you separate them, the capacitance should go down. Heavy build insulation, as well as scatter winding do this.

The core material effects inductance. Now of course it all starts to get tricky when you start talking about mutual inductance, which is from the magnetic field created by the flowing electrons in the coil itself... etc., etc.

As Mike pointed out, when you are trying to measure this stuff, what are you seeing? A real change, or an anomaly in the measurement?

[belwar]

well capacitance changes with whatever frequency you're at, so using the inidividual's resonant peak freq seems like a bad idea if you're testing several pickups.

Does capacitance change with frequency? I might be wrong, but I dont think it does... The point of this thread is to determine that actual self capacitance of an inductor. If you look at Joe's Post #83 on page 3 of the Lane Poor Thread (http://music-electronics-forum.com/t14915-3/#post119005) you'll see how you cant use an LCR meter to measure the Capacitance of an inductor because it provideds a non-real answer and not the self capacitance of the inductor.

When you buy a capacitor off the shelf it doesnt vary with the frequency.. its a capacitor rated for a specific farad rating up to a certain voltage.

I don't believe Inductance and Capacitance change in with frequency, but Inductive Reactance and Capacitive reactance do.

[Joe Gwinn]

Funny you should mention that. Does it change the capacitance, or just change the measurement?

Over the weekend I measured a humbucker coil with and without the cores in place. I get 78 pdf with he cores in place and 88 pf with them out.

The metal core will increase the capacitance all by itself. Try a piece of thinwall brass tube with a lengthwise slit (to prevent eddy currents). The brass tubing sold in hardware stores and hobby shops should work. One can also buy aluminum and stainless steel tubing from the same sources.

I measured the resonant frequency by measuring the zero of the phase using my I-V circuit and Mac with Electroacoustics toolbox. The measurement was done with no external capacitor and with 94.3 pf in parallel. Then I solved the equations for the square of the resonant frequency to get both L and C.

Try multiple capacitors so you can see if the points fall on a straight line or not. Inductance variation will cause the line to curve. Other effects may also. Seeing the effect of various cores should help to tease the effects apart.

Joe might say the difference is due to eddy currents. I might say that it is due to the leakage flux from short open cores. But we really do not know. I doubt that it represents a real change in the C, but that is certainly not proven.

The presence of a metallic core will most certainly change the self-capacitance, but eddy currents have nothing to do with it. As mentioned above, the effect can be explored with a tubular core, with and without a lengthwise slit.

In transformer design, it is well known that the presence of a metallic core (or nearby electrostatic shield) increases coil capacitance, and that this is independent of magnetic leakage. The usual remedy is thick insulation between coil and nearest metallic body.

[Joe Gwinn]

Does capacitance change with frequency? I might be wrong, but I don't think it does...

Capacitance does in fact change somewhat with both frequency and voltage in practical capacitors, but I doubt that one can detect the effect in guitar pickups.

The point of this thread is to determine that actual self capacitance of an inductor. If you look at Joe's Post #83 on page 3 of the Lane Poor Thread (http://music-electronics-forum.com/t14915-3/#post119005) you'll see how you can't use an LCR meter to measure the Capacitance of an inductor because it provides a non-real answer and not the self capacitance of the inductor.

This is true, but has nothing to do with the question of parameter variation with frequency. In other words, one would see the same effect even if all components were ideal.

When you buy a capacitor off the shelf it doesn't vary with the frequency.. it's a capacitor rated for a specific farad rating up to a certain voltage.

It depends on the capacitor type. Some vary quite a lot. Temperature also matters. However, film capacitors and the self-capacitance of pickup coils is essentially constant, where "essentially constant" means smaller than the other errors.

I don't believe Inductance and Capacitance change with frequency, but Inductive Reactance and Capacitive reactance do.

Inductance does vary in metal-cored inductors. Mike and I are debating just how big the effect is in pickups. The issue will be settled in the lab.

Capacitance in film capacitors and the like is pretty constant, which is one reason I specified film capacitors.

Cheap ceramic capacitors can make pretty good contact microphones.

Reactance for sure varies with frequency, almost by definition.


To complete the picture, the resistance of resistors varies with voltage and frequency, the details varying by resistor type.


More generally, the parameters of all practical components vary with voltage, current, frequency, sometimes illumination, et al. It's only a matter of degree.

[belwar]

As Mike pointed out, when you are trying to measure this stuff, what are you seeing? A real change, or an anomaly in the measurement?

Good point.

Im going to keep that question in mind as I start with experimentation.

I'm one of those people that learn by doing and making mistakes.

[Possum]

....

Belwar while you're scrambling your brains, look this over, these guys actually did work on all this and more, how's your math?
http://online.physics.uiuc.edu/cours...asurements.pdf

[Mike Sulzer]

Right, but if you factor in the permanent magnet, the eddies are more pronounced.

The permanent magnetic field has nothing directly to do with it. The permeability (ability to become magnetized) does matter, and this can be a function of the permanent magnetic field that is present.

[Mike Sulzer]

The metal core will increase the capacitance all by itself. Try a piece of thinwall brass tube with a lengthwise slit (to prevent eddy currents). The brass tubing sold in hardware stores and hobby shops should work. One can also buy aluminum and stainless steel tubing from the same sources.

Maybe I got it backwards. I will check.

Try multiple capacitors so you can see if the points fall on a straight line or not. Inductance variation will cause the line to curve. Other effects may also. Seeing the effect of various cores should help to tease the effects apart.

Yes, this is what I have in mind, time permitting.

In transformer design, it is well known that the presence of a metallic core (or nearby electrostatic shield) increases coil capacitance, and that this is independent of magnetic leakage. The usual remedy is thick insulation between coil and nearest metallic body.

I would think that increasing the space between coil and shield would be better. Using an insulating material should increase the capacitance if its dielectric constant is greater than t hat of air.

[Mike Sulzer]

On the subject of cores causing a change in capacitance:

A grounded shield is different from ungrounded cores. If the part of the coil closest to the shield has an oscillating voltage that differs from ground, then charge must flow.

And in a flat plate air capacitor, inserting an ungrounded metal plate between the capacitor plates will increase the capacitance since it is now equivalent to two capacitors with smaller spacing in series.

But, how much charge has to flow as a result of ungrounded cores? The potential across the inner winding is a small fraction of the total, and so the fact that the potential on the surface of the core must be the same everywhere has only a small effect on the flow of charge in the coil.

[David Schwab]

The permanent magnetic field has nothing directly to do with it. The permeability (ability to become magnetized) does matter, and this can be a function of the permanent magnetic field that is present.

So how does that explain dropping a permanent magnet down a copper pipe, or a thick sheet of aluminum, and having the eddy currents slow the drop of the magnet? Permanent magnets produce eddy currents in nearby conductors.

So in a pickup, when the magnetic field is disturbed, eddy currents will be produced in any metal parts on the pickup. The field from the permanent magnet is much stronger than the field produced by current moving in the coils, and in fact you wouldn't have that current without the magnet.

It doesn't matter if the metal is magnetizable, just as long as its a good conductor.

[David Schwab]

But, how much charge has to flow as a result of ungrounded cores? The potential across the inner winding is a small fraction of the total, and so the fact that the potential on the surface of the core must be the same everywhere has only a small effect on the flow of charge in the coil.

Your cores in a pickup should be grounded, and in a typical humbucker they are.