EDDY CURRENTS

[Tinkerer]

Eddy currents

The TX pulse of a PI-MD induces eddy currents in the target.
There are the surface eddy currents or skin effect eddy currents.
There are the core eddy currents.
Some of the eddy currents have a resistive relationship.
Others have a reactive relationship.

I would like to learn more about all these eddy currents, understand them so as to devise a method to measure them.
What for?
The signal that we capture with the RX coil, is composed of all these eddy currents. As long as these eddy currents are all lumped together, we get little information about the target.
If we can device methods to separate these eddy currents so that we can measure the partial signal amplitude of each kind of eddy current, we will be able to read much more information about a target.
Lets start with the skin effect eddy currents:
At the first instant in time, when the TX magnetic field cuts accross the target, theses eddy currents develop along the surface of the target. Questions:
1.What is the duration in time of the development of these eddy currents? Micro seconds? Nano seconds?
2.How deep do these eddy currents penetrate into the target surface? Micro meters? Nano meters?
3.At what point in time do these eddy currents subside? Do these currents expand along the surface of the target at a certain speed?
4.What happens when the eddy currents reach the far corners of the target? Are they reflected back to their point of origin? Do they oscillate between their point of origin and the far corners untill they have dissipated their energy?
5.How long does it take for these surface eddy currents or skin effect currents to subside? Does this time have a relationship with the surface area of the target, as presented to the expanding or collapsing magnetic field of the TX pulse?

I have a piece of Alu foil of 22 micron (micro meter) thickness.
I also have a piece of gold foil of 100 nano meter thickness.

I will cut both to the same surface size and detect them with my PI-MD.
What will the response signal tell me?

I expect to get only a skin effect response from the gold foil.
But, what about the Alu foil? Is it thick enough to support core eddy currents?

Both targets are of the resistive kind, the conductivity varies somewhat. Will this make a difference?

I would much appreciate the help of the forum in my effort of learning to understand what is happening when the TX magnetic field cuts acccross the target.

Please help.

Next we will look at the core eddy currents.

[Monolith]

Eddy currents

Tinkerer,

I wish I knew how to help. It certainly would be interesting to know if it is possible to look at the signal generated by skin effect.

Do you mind telling me how you get a piece of gold foil of 100 nanometer thickness?

Monolith

[Tinkerer]

Measuring Skin Effect

Tinkerer,

I wish I knew how to help. It certainly would be interesting to know if it is possible to look at the signal generated by skin effect.

Do you mind telling me how you get a piece of gold foil of 100 nanometer thickness?

Monolith

Hi Monolith,

To answer your question:
The alu foil is 3.5 Inches X 3.5 Inches square. In millimeters this is about 6,806 mm square. A watchmaker was so kind as to measure the thickness with his Micrometer, 22 micrometer.
The gold foil is of the same surface area. There is no way I can measure the thickness. It is much, much thinner, in fact, when I wanted to look at it, it became airborne and flew away, to auto-destroy itself upon landing.
I had to get another one.
I have to accept the thickness as indicated by the seller. I did check for information about gold leaf on the Internet. 100 Nanometer is an accepted measurement for gold leaf.
I have made my first measurements of the Skin Effect.

The alu foil gave 350mV at 20 cm from the coil.
The gold foil gave 2mV at 20 cm from the coil.

There is some difference in conductivity between gold and aluminum, but the results above show clearly that the thickness makes most of the difference.

Interestingly, a 1gram gold bar gives also about 2mV at 20cm from the coil.
The surface area of the gold bar is 8.5mm x 15mm = 127.5mm square

So the gold leaf has a surface area about 53 times greater than the gold bar, but gives the same signal amplitude.

For this test, the coil was an composed of TX= 280mm diameter, RX= 140mm.
Both coils about 300uH.
Preamplifier and S&H, total gain 1000.

Now, my question to the forum:

Does this test qualify for a Skin Effect measurement?

Tinkerer

[bbsailor]

Hi Monolith,

To answer your question:
The alu foil is 3.5 Inches X 3.5 Inches square. In millimeters this is about 6,806 mm square. A watchmaker was so kind as to measure the thickness with his Micrometer, 22 micrometer.
The gold foil is of the same surface area. There is no way I can measure the thickness. It is much, much thinner, in fact, when I wanted to look at it, it became airborne and flew away, to auto-destroy itself upon landing.
I had to get another one.
I have to accept the thickness as indicated by the seller. I did check for information about gold leaf on the Internet. 100 Nanometer is an accepted measurement for gold leaf.
I have made my first measurements of the Skin Effect.

The alu foil gave 350mV at 20 cm from the coil.
The gold foil gave 2mV at 20 cm from the coil.

There is some difference in conductivity between gold and aluminum, but the results above show clearly that the thickness makes most of the difference.

Interestingly, a 1gram gold bar gives also about 2mV at 20cm from the coil.
The surface area of the gold bar is 8.5mm x 15mm = 127.5mm square

So the gold leaf has a surface area about 53 times greater than the gold bar, but gives the same signal amplitude.

For this test, the coil was an composed of TX= 280mm diameter, RX= 140mm.
Both coils about 300uH.
Preamplifier and S&H, total gain 1000.

Now, my question to the forum:

Does this test qualify for a Skin Effect measurement?

Tinkerer

Tinkerer

The short answer is yes. Here is why.

Eddy currents can be described in three stages.

Stage 1: Early or surface eddy currents. These currents occur when the material is very thin or the pulse width is so short that the later stages do not occur.

Stage 2: Mid or migrating eddy currents. In this stage the surface currents have formed but there is additional energy that causes the eddy currents to migrate toward fully saturating the full thickness of the metal target.

Stage 3: Late or saturated eddy currents. In this stage the eddy currents saturate the full thickness of the target. Any additional TX pulse power does not cause any additional eddy current stimulation.

The real issue is how long does the eddy currcent discharge time last in relation to the initial pulse width and the size/thickness of the target.

If you can increase the pulse width and obtain longer target Time Constants, then the target is not fully saturated until adding a longer pulse width does not causes a longer decay curve. When you reach the longest decay curve, adding any additional pulse width will not cause any longer target discharge signal.

Here are the discharge times and time constants for some common targets.

US copper penny, 350 us total discharge time, 70 us TC
1 Sq Inch of Alum foil .002" thick, 50 us total discharge time, 10 us TC
US Nickel, 100 us total discharge time, 20 us TC

The TX charge curve is exponential. In one TC the current grows to about 63 percent of max. In another TC the current grows 63 percent of the remainder or to about 85 percent. After the third TC the current grows again 63 percent of the remainder or to about 95 percent. The later TCs (TC 4 and TC 5) add less power to charge the targets than the earlier TCs.

The rule of thumb seems to be that smaller, thinner targets get fully charged faster with less TX pulse width. Another variable is the material that the target is made from. Some metals initially charge higher and faster but also discharge very fast like small gold nuggets.

Since targets are unknown, untill recovered, you want to focus on looking for the most common targets in an area but would like to find the larger, deeper gold nugget also but pumping out a PW to fully energise a large, deep nugget is a waste of power for the smaller more common nuggets. This suggests that you cover the same ground twice; once with PW settings for small nuggets and again for larger, deeper nuggets with a longer PW.

Target TCs just tell you the most efficient PI PW setting for that class of targets only. Nickels and gold jewlery can use a 3X TC or 60 us Pulse Width while a copper penny needs about a 210 us TX PW to be practically detected.

Where you sample on the target discharge curve will produce a signal in the receive window. If that signal is more vertical than horizontal it means that the target is a fast discharge target. If the later signal is more horizontal than vertical, that tells you that the target has a different characteristic or is a large, deep target that only has a surface charge. There in lies the difficulty in trying to deduce the absolute identity of a target with PI pulse analysis. Take narrow samples at a very early time, the mid time and the late time and compare the signal amplitudes to deduce the target identity although not perfectly all the time but better than conventional PI go/no-go target indicators.

I hope this helps?

bbsailor

[Monolith]

Skin Effect Eddy Currents

Hi Monolith,

To answer your question:
The alu foil is 3.5 Inches X 3.5 Inches square. In millimeters this is about 6,806 mm square. A watchmaker was so kind as to measure the thickness with his Micrometer, 22 micrometer.
The gold foil is of the same surface area. There is no way I can measure the thickness. It is much, much thinner, in fact, when I wanted to look at it, it became airborne and flew away, to auto-destroy itself upon landing.
I had to get another one.
I have to accept the thickness as indicated by the seller. I did check for information about gold leaf on the Internet. 100 Nanometer is an accepted measurement for gold leaf.
I have made my first measurements of the Skin Effect.

The alu foil gave 350mV at 20 cm from the coil.
The gold foil gave 2mV at 20 cm from the coil.

There is some difference in conductivity between gold and aluminum, but the results above show clearly that the thickness makes most of the difference.

Interestingly, a 1gram gold bar gives also about 2mV at 20cm from the coil.
The surface area of the gold bar is 8.5mm x 15mm = 127.5mm square

So the gold leaf has a surface area about 53 times greater than the gold bar, but gives the same signal amplitude.

For this test, the coil was an composed of TX= 280mm diameter, RX= 140mm.
Both coils about 300uH.
Preamplifier and S&H, total gain 1000.

Now, my question to the forum:

Does this test qualify for a Skin Effect measurement?

Tinkerer

According to your results, it seems that the Skin Effect eddy currents are negligible with only 2 mV compared with 350mV for the thicker material.

bbsailor proposes some additional experiments. Are you going to give us the results of these?

Monolith

[Tinkerer]

Core Eddy Currents

bbsailor posted the Time Constants TC of 3 types of target. (thanks)
US 1c, a copper clad coin with a zinc core. TC=70uS
US 5c, or a Nickel with a TC of 20uS
1" square alu foil, with a TC of 10uS

I started the experiment with the Nickel.

With a TX pulse of 50uS, that is 2.5 TC, we can consider that the Nickel is charged or excited to near maximum response.

TX of 50uS with a coil of 28cm diameter and 300uH, 12V, R=5for coil+ Mosfet.
Target distance from the coil = 15cm
Samples taken at 6uS, 18uS, 30.5uS

Sample # 1 at 6uS, = 35mV
Sample # 2 at 18uS, =19mV
Sample # 3 at 30.5uS, = 11mV

I was surprised how closely (less than 1 mV variation) the actual readings coincided with the calculated values for a target with a TC of 20uS.

Extrapolating the exponential decay curve this means that the value a switch OFF would be 48mV.

It is obvious from these results that early sampling is imperative for targets with a short TC.

Next I will try the alu foil and the copper clad penny (US 1c)

Tinkerer

[Tinkerer]

TIME CONSTANT TC OF 1" SQUARE ALU FOIL

TIME CONSTANT TC OF 1" SQUARE ALU FOIL

Here are the results for a square 1" alu foil.
The parameters are the same as above.

Sample # 1 at 6uS = 24mV
Sample # 2 at 18uS = 10mV
Sample #3 at 30.5us =1mV

Then I repeated with an alu foil of 1" wide and 2"long
# 1 = 65mV
# 2 = 30mV
# 3 = 2mV
The TC is still short, but the amplitude increased.

The gain for the tests above is 1000

Maximum distance detected for 1" square alu foil is 30cm.

Then I changed the damping a bit and took a sample at 5uS, gain 1000
Signal amplitude 40mV for the 1" square alu foil at 15 cm from coil.

The decay curve is not exponential anymore, it is "bumpy" I suspect there is some capacitance carry over.

For the US$ 1c or Penny, I got a blind spot at 18uS.

It looks as if I opened a can of worms with this test.

Tinkerer

[ZED]

Wondering

What is the TC cut off point for when a target has reached the end of its decay,1mv,1uv,1nv,1pv. Is there a standard or a wide spread acceptence ?

Zed

[Tinkerer]

TC cut off

Wondering

What is the TC cut off point for when a target has reached the end of its decay,1mv,1uv,1nv,1pv. Is there a standard or a wide spread acceptence ?

Zed

I think the cut off can be considered to be the point when the signal disappears into the noise level.

Tinkerer

[Tinkerer]

3-D Vector Fields applet

Here is a nice Magnetostatic vector field applet to visualizt the coils magnetic field.

http://www.falstad.com/emwave2/index.html

Tinkerer