Eric, from what I recall from "EE-101"...

[Anonymous]

Gold is the best conductor, but silver gives the best warp of a coil field making the detector see silver as the best conductor. Some other metals are in the same area, but it is not conductor properties which the detector reads, its which metals are most easily saturated by the electromagnetic field of the detector. Silver is more easily saturated so that silver reads higher, even though gold is a better conductor in a circuit with electricity actually flowing through it.
Is there any way to change the detector bias towards conductivity as opposed to metal saturation in metal identification?

 

[Anonymous]

And would not a low frequency penetrate deeper into the target surface causing stronger eddy currents than a higher frequency?

 

[Prospector Al]

Hi Vlad,

Actually, gold is not the best conductor. It's used on connector pins only because it not as easily corroded as silver and copper.

Silver has 73% and copper 77% of the resistivity of gold. (Reference:
The Handbook of Chemistry and Physics)

However, the resistivity of the target is only one half of the equation that determines its response to a coil pulse.

The response is determined by the "Time Constant" of the target.

T = L/R , where T is the time constant, L is the inductance of the eddy current path, and R is the resistance of the path.

(You might use T = "The Transylvanian Equation" as a mnemonic aid, to remember this. It's a little know fact that your namesake, Vlad the Impaler was the inventor of the first PI detector.)

The transmitter coil current pulse generates a voltage pulse that is imposed on the target. The eddy current engendered by the voltage pulse is determined by another equation:

I = Io X e.sup -t/T , where I is the eddy current,T is the above defined time constant, t is the length of the voltage pulse, e is the base of the natural logarithms and Io is the current that would be attained, if the pulse were infinitely long...

The interpretation of these formulas is that for a given size target, high conductivity leads to a longer time constant, and consequently, it takes a longer voltage pulse to "charge" it, than is the case with a target with lower conductivity. The eddy current in the target with a short time constant also decays faster, which induces a higher voltage in the receiver coil:

E = dI/dt , which means that the voltage E is proportional to the change of the current per unit time.

The inductance and resistance of the eddy current path is also affected by the shape of the target. A gold coin generally has a longer time constant than a nugget of the same weight, owing to its irregular shape.

Most gold alloys are solid solution of copper and gold. Thus, the resistivities are intermediate between the two. According to my calculations, the resistivity of 1O carat gold is 1.75 and that of 14 carat gold 1.77 ( Silver = 1.60 )

As far as "warping the field" is concerned, iron does the best job, because its high permeability concentrates the field and therefore
causes high eddy currents to flow, despite that fact that the resistivity of iron is 4.6 times higher than that of gold...

These are the simple facts--the shape of the coil pulse further complicates the picture.

Happy Hunting,
Prospector Al

P.S. Moderators, if I'm out line, putting all this technical stuff here, please let me know...

 

[Anonymous]

When a detector's field hits something one of the by products of the eddy current is heat-is this large enough to be measured in some way? Does this effect also make a hot rock, hotter?

 

[Eric Foster]

Hi Prospector Al,

"Most gold alloys are solid solution of copper and gold. Thus, the resistivities are intermediate between the two. According to my calculations, the resistivity of 1O carat gold is 1.75 and that of 14 carat gold 1.77 ( Silver = 1.60 )"

Unfortunately, it is much worse than this. Alloying gold and copper, or gold and silver gives a much higher resistivity than you might expect. I was fortunate once, to have access to some gold/copper and gold/silver standards. These were square pieces of metal, all the same size and thickness, with the only difference being the alloy makeup. They range from pure gold through to pure copper in 5% stages i.e. 95% gold / 5% copper, through 50/50, all the way to 5% gold / 95% copper, to pure copper. Same with the gold/silver standards. Only 5% addition of the other metal caused a dramatic rise in resistivity (drop in conductivity). I'm not sure, metallurgically, why this is so, but it is a demonstratable fact.

If you look at the decay from a 10 carat gold ring, and compare it with a similar size silver ring, the silver ring will have a much longer decay curve. Same if you compare the gold ring, and an equivalent ring in pure copper.

Eric.

 

[Prospector Al]

Hi Eric,

I stand corrected--an excellent example of the difference between
theory and practice.

I don't understand what the mechanism would be to cause this though...
It's true that some alloys have thermal properties that are unexpected. The "eutectic" ones, e.g., whose melting points exhibits a dip at certain proportion of the component metals. I wonder if the electrical properties might show a similar dip?

I must commend you on your willingness to share your knowledge freely.
Most people active in this field seem to guard their secrets jealously and divulge facts only in patent applications, and then as little as possible...

In the Sciences, people publish their work, and that is no doubt, responsible for the rapid advancement in those fields.

In the metal detector art, each person seems to have to find out the facts for himself. The books published on the subject are intended to
promote a certain brand, rather than to disseminate knowledge.

There still is no scientific treatise on the phenomenon of "anomalous rocks". I suspect you have an academic background, so you might be the person to write such a book...

Prospector Al

 

[Eric Foster]

Hi Prospector Al,

When I find the plots, I will post them. Tried to locate them tonight, but no joy. The effect is quite dramatic. I suspect that alloying ties in the "free" electrons tighter, so that they cannot migrate so readily. Many metals show this effect e.g. stainless steel, where a small percentage of chromium greatly increases the resistivity of ordinary steel.

Regarding hotrocks, I'm sure that many signals are due to deficiencies in the detector, rather than a strange property of the rock. I find that with careful electronic design, many so called hotrocks are not sensed.

My background is both academic and practical, although I often do the practical first and then worry about the theory. Nothing like seeing what works in the field and then improving on it. I have seen many academic papers that get bogged down in complexity and nothing much of any practical use results.

Soon, I will start the book again. It will contain a section on iron mineralised ground and hotrocks, but perhaps there is not so much to write about, as one might think, if you have the right detector.

Eric.

 

[Prospector Al]

Hi Eric,

The resistivity of alloys is interesting stuff, but I'm still more concerned with hot rocks. I have spent hours watching the response of
a detector to "ironstone" in an Australian gold mine. The detector I designed for that application has a coil pulse with a flat top, similar to what you describe in the "sticky" post.

The length of the flat top is 200 uS, obtained with a constant current
drive. What I found there is that if the flat top is longer than 4 X
the time constant of the ore, it doesn't help to make it longer.

For gold nuggets with a long time constant, there is a "carry-over" effect: The leading edge of the coil pulse generates eddy currents in
the target. If the flat top is not long enough to allow the "front" currents to decay to zero, the amount remaining is subtracted from the
currents generated at the trailing edge of the pulse.

For an industrial detector, current consumption is not a problem, so I was able to optimize the design, with a long pulse. This of course is not possible for a portable detector, unless you "go Minelab", with a gel cell in a backpack.

If the coil current is still rising at the time of switch-off, there has been no opportunity for the "front" signal to decay, and sensitivity is compromised for all targets.

So far, my observations agree with yours. There is one aspect of your
account that I am curious about: You mention that the time constant of a metallic target is also affected by the length of the coil pulse.

I have not seen that. (Perhaps, because I have not been looking for the effect.) The change in signal amplitude is adequately explained by the fact that the "front" and "end" eddy currents are of different
polarities and add algebraically. This should not affect the time constant, however.

For the time constant to change, the signal must be generated in different parts of the target. You mention the "skin" effect in one post. I'm familiar with that from RF theory, but I thought that it would come into play at high frequencies only.

Let's assume that the target is not fully penetrated with a short coil pulse. The peripheral eddy currents should have longer T, because the inductance, owing to a longer eddy current path, should be longer than that of the interior currents, having a shorter path.
You mention, however, that the time constant get longer, with a longer coil pulse--exactly the opposite from what I would expect.

Furthermore, the composite eddy current should not show a strictly exponential decay, because the sum of exponentials is not an exponential function.

To explore this in detail requires the kind of setup, with a ferrite core, that you mention. Otherwise, the signal amplitudes are too low.

I'm contemplating a further investigation into the matter, but I have to ask myself if this would be a purely academic exercise, or if there is a chance of accomplishing anything of practical value...

(I also have an academic background, but only down-to-earth, practical results matter now...)

Prospector Al

 

[Anonymous]

While they do melt at the melt point of the lowest melting point alloy, this is caused by the higher temp metal being so diluted, like silver solder.
However look at one of the properties of aluminum, which can melt and give way without any color change or other characteristic-a "hot short."
But if aluminum is heated by an acetylene torch, it forms a refractory oxide that burns, or melts at a much higher temperature than the aluminum. I believe aluminum melts at 1100, but al-oxide is at 3000.