Chris, Here is a quick general idea that is very boring.

[Cody]

Chris, the idea of inducing eddy current into a dime that is 12 inches from the coil and then the dime induces current into the receiver is takes a giant leap of faith to say the least. I think there is a much better way to consider what actually takes place. Here is something I threw together while taking a shower for a person like yourself that understands electronics.

Consider a transformer with a single primary and secondary and the primary being pulsed by a square wave. We can monitor the primary and secondary with an oscilloscope and air as the core and depending of design see a fairly nice square wave in both. I think of an infinite variable core with metal particles embedded. It is pretty easy to visualize that we can see and analyze signal in the secondary of the transformer compared to the signal in the primary. The various electrical parameters of the transformer and of the metal particles are going to paint a pretty clear picture of what the core is doing. What are fairly constant are the soil or base core material and what changes abruptly are the metal particles. We then know to reject the constant parameter and only look at the abrupt changes. What we have done is reject the constant parameter of the core.

The data is in serial form as it enters the input and is amplified and fed to six demodulators. I think we have to keep in mine we are dealing with one signal at a time which is demodulated into 3 frequency ranges for high, mid, and low, filtered and then processed. It would be nice to overlook the 28 frequencies but we really cannot in my opinion since metals respond differently to different frequencies.

Now we sample the secondary so many times, a sequence, to see what the signal looks like. Use precise time measurement we get a very good idea of what the metal is in the core that causes the abnormality in the signal. We can store constant core data as instantaneous and historical in that we look at the constant core data at an instant in time, the core data that just went from instantaneous to historical and predict the future core data. We need some algorithms to tell the processor what to do with the data, store some lookup tables to compare the data to, generate an audio and visual presentation and then I think we have the basics of the detector.

The other function are mostly to help us to get a good look at the core data iron particles and deal with the motion of the coil to feed the data into the receiver. Audio, Fast, Deep, and the like are for that purpose. It is picking those good targets from the core mixed in with trash metals so we need to be able to adjust a few things to do that. As a last thought consider a piece of iron in the core that is fairly large and several particles of other metals that vary the core. That large piece of iron is going to be a problem. It is not that we don

 

[Captain Kirk]

Cody,

It is my opinion that the eddy current in conductive target does in fact produce a signal in the receiver coil. I would be inclined to treat the result in the voltage domain of the receiver coil rather in the current domain. The voltage is what gets induced by the magnetic field. The current in the receiver coil simply depends upon how you load the receiver coil. By the way, if you load the secondary coil with a short circuit, then you have a current transformer (But, in this case you better not connect a voltage source to the primary or you will essentially short out the voltage source.)

Refer to my earlier post. There are really two components of the signal in the receiver coil. There is the one produced by the coupling of the primary to the secondary without any effects of the target. Then there is the signal produced in the receiver coil as the result of the magnetic field disturbance of the target. The detector electronics are constantly measuring and storing the information about the first signal. Then we can essentially subtract that signal from subsequent signals. What is left is the target induced signal.

What do you think.

HH,
Glenn

 

[AK in KY]

I love to read Codys and Glenns post as they have been very helpful to me. I hope I'm not alone but this time I have no clue what so ever. Keep on posting, I may get the next one.

 

[Cody]

I don

 

[Cody]

I should have said I agree with both of your posts. I am trying to monitor my dogs, parrots, and grandkids all at the same time. As I said I think these post generate interest and I know speeking for myself I certainly enjoy you sharing your expertise. I broke down and printed the patents and will force myself to look at them a little closer. I swore I would never get into technical reading again after all those years but those habits are hard to break.

 

[Captain Kirk]

n/t

 

[Captain Kirk]

Cody,

At the risk of being ridiculously technical, pontificating and blowing smoke, I submit the following for your consideration:

In the final analysis what is picked up in the receiver coil is, for all practical purposes, the result of the magnetic field enclosed by the receiver coil. That is, the receiver coil does not see any direct effects of eddy current in a target under the coil.


#1 The transmitter coil generates a magnetic field. A portion of that magnetic field is captured by the receiver coil.


#2 The transmitted magnetic field causes an electric field in the plane perpendicular to the magnetic field. The magnitude of the electric field is proportional to the time rate of change of the magnetic field. You can NOT properly talk about a phase relationship between these two fields for any form of excitation except sinusoidal. The Explorer is a time domain machine and "phase shift" does not apply.

#2.1 The electric field generated by the magnetic field, then causes eddy currents in the target.


#3 The eddy currents in the target distorts the magnetic field picked up by the receiver coil. So what causes the distortion of the magnetic field? A perfect conductor (with zero resistivity) will not allow any changing magnetic field to pass through it. This is because the L/R time constant of the eddy current build up goes to zero and instantaneously diverts (distorts) the magnetic field around the target.


#4 The receiver senses the distortion of the magnetic field. That distortion changes the induced voltage in the receiver coil. That is what is measured.


#5 RADIATED AND RECOVERED ENERGY

#5.1 I do NOT want to get into the details of an antenna system that radiates energy into space that is never recovered by the antenna system (for example a 50kW transmitter). But, the following basic ideas are presented.

#5.2 Radiated energy is transmitted in a combination of electric and magnetic fields in such a manner that the energy stored in these two fields is swapped back an forth in a sinusoidal manner. There may be several of these fields present simultaneously (in a spectrum such as radiation from the sun). Here it is proper to speak in terms of the phase relationship between these two fields for a specific frequency component.

#5.2 I do NOT believe that metal detectors are designed to radiate energy into space. There is a small amount of consequential radiation and that is the reason why metal detectors 50 feet apart can interfere with one another.

#5.3 I do believe that the detector coil is designed send energy into the target matrix and then recover almost all of that energy in the next half cycle of the waveform. The detector is designed to determine how the magnetic field is effected by the ferrous and conductive targets in the target matrix. The real driving force is the magnetic field.


ABOUT A DIME AT 12"

I do believe that a dime at 12" produces a very weak disturbance in the receiver coil voltage. That is why the signal is so faint. On the beaches of Southern California, I consistently dig dimes at about 10" in the wet sand, but I do not remember very many (if any) dimes being dug at 12". There is still a signal at 12" for a dime, but I suspect that it is pretty much in the noise level by then.

HH,
Glenn

 

[Cody]

I would be lost for words if I talked to any engineer that thought this detector worked liked a radio TX and RX. There are a lot of folks that think it does so this exchange provided an opportunity to dispel that idea for readers that find this a topic of interest.

I am not sure how the data is processed as far as the various frequencies are concerned. That area does not interest me all that much but will look at the patent and see what is there. The data in the receiver coil is amplified, fed to 6 demodulators in sets of two, filtered, averaged and processed. I need to look at what the demodulators are up to although I think I already have a pretty good idea.

Have a good evening,

 

[Cody]

I doubt if most user realize they are using a PI detector. It is a sub-class of one that discriminates but still a PI. We don't just pulse the soil with electromagnetic energy then analyze the soil matrix to analyze the matrix, reject the soil, and declare that a target is good or bad. It is not that simple. There is a series of pulses, a cycle, that pulse energy into the soil matrix and the results of those pulses are analyzed. Those series that make up the cycle are repeated over and over as long as the detector is on.

The RX and TX coils are balanced in a way that if the matrix is air then there is no signal induced in the RX coil. The soil matrix will cause a signal to be induced into the RX coil. The first thing the RX coil "sees" is amplification and the next are 6 demodulators.

Minelab tells us this is what makes the whole thing work and the major value of time domain. We can avoid frequency domain design (FD) which is much more costly and difficult to work with time domain rejection of the soil. The trick is to pulse the soil with a series of pulses and sample those pulses by turning the demodulators on and off at very precise times. The energy pulsed into the soil decays more quickly in the soil than a target. As far as I am concerned at this point in time that sums up the idea of a time constant. We are going to look at short, long, and medium time constants. They will tell us what is soil and what type of target is in the matrix. The demodulators are paired so we get that data from a series of pulses, a cycle, then the microprocessor tells us what is soil to reject, and what is a target to accept or reject. This is done with the sampling of the pulses in a cycle to see how the energy decays in the soil and target to generate a time constant. A time constant for a target is compared to stored time constants to ID a target. We then accept or reject the target based on user input to accept or reject a time constant. The time constant is based on the rate of decay of energy in a target and to the user is a pixel on the display. There are more than one time constant for a dime and more than one pixel. The TC for soil is very quick compared to those for targets. The microprocessor is going to do some tricks with long, short, and medium time constants to ID targets. I will stop here and let this get digested, analyzed, and corrected.

We can sum this up by saying there are a series of pulses divided into cycles that are analyzed to generate a time constant. A time constant is very short for soil and longer for targets. There is a lookup table of TCs for the ID of targets. A targets TC is compared to a stored TCs and accepted or rejected based on user input. Mostly what is left is to generate a tone or visual display.


Have a great Sunday,