Ok then, as per a request, here is a bunch of info on the subject:
No opinions, just facts. and I will attempt to answer this as scientifically and as well as my old brain can manage.
Metal detectors of the VLF variety use electromagnetic fields to measure the difference in the normal ground matrix when the "disturbed" field occurrs (finds a stranger in the area) . This means in simple terms, when there are normal, regular, already magnetically aligned molecules laying in a row, positive to negative one after the other all over and in the ground we are searching in - and thus they all become part of the "mat", the "matrix", the magnetically induced (OR) potentially charged substances we are studying or sampling at the moment..
The VLF detector emits an electromagnetic field much like the magnetic field in a piece of lodestone or magnetically charged iron of some type. An rf signal is also generated. I will send those a simple yet workable project sample to show how this is all accomplished using an AM radio and a handheld calculator, if you ask me. Radar, sonar, and metal detector radiation all work in similar fashions, just as does algebra, Set Theory, and calculus - in their parameters of operand.
If we put a piece of paper over a hunk of magnetized iron or steel and pour iron filings on it we wlll be able to see it's magnetic field displayed. It will look a lot like a bunch of circular or somewhat circular or even at times random or oval patterns encompassing the piece of magnetically charged iron or steel. There is no constant and there is also no given or set determinant either. In fact, that is why there are spikes and incidentals (oddities) in all these types of fields emitted. We would see on a scope little places where there are little strange anomolies here and there and in fact there would actually be no perfect patern of emission at all. There would be in fact, a rather odd-looking pattern in nearly every case. It would rarely appear to be a perfect or near perfectly shaped pattern of emission. This is exactly why we see places in the case of the iron filings over a magnet bunching real tightly and closer together where the magnetic field is the strongest, here and there.
This field can be quite large in size, but more intense the closer we get to the coil. There is no such thing as a modern detector that shuts off it's pattern of electrically and magnetically radiated field at 8" or 10" or 16". In fact, it is likely that the field from and average detector using an 8" coil radiates out to as much as 8 feet - or even more at times. Interference though, can dramatically interupt this pattern and reduce the greatness and width of the radiated field.
In simpler terms, when we locate an object, the detector sees ALL of the entire area it radiates to, and for the ENTIRE time it is turned on. We do have ways to reduce it's magnetic field intensity though, such as limiting the gain or power that sends or creates the electronic magnetic field. We can do this too with discrimination (ground balance is just another form of discrimination, and vice versa), they both do the same thing .
The farther the target is away from the coil the greater time is elapsed and the greater the delay electronically with it's electrical/magnetic signal forming eddy currents produced when hitting the target within the electromagnetic field, but that is a whole 'Nother Story and it takes more than scrolls to describe the process here. Eddy currents take longer to wrap around a large, deep object than they do a small shallow one.
This time delay is what is refered to as "phase shift". Phase shift is a way of saying that there is a timing effect involved, and that the timing has been interupted and changed in duration, intensity, and strength, and if we (didn't) have the phase shift in effect plus at least some form of self adjusting threshold we would be getting nothing but a whole lot of noise from every piece of metal within the normal 8 feet, and that would just about drive us all crazy.
Now here is the most important part of why we can (only) find a coin at 8 inches but a garbage can lid at 8 feet; And this is quite involved physically, not standard fare for the average person to really need to know..:
Phase shift involves the sine wave pattern configuration of positive and negative voltage, going up, and then down, and then up, and then down, in a rythmic snaking pattern on a line, half the time up to a certain point (positive), and then half the time down down to a certain point (negative), and always the same cycle rate. There is a time period (or time delay) involved here. It is called a "cycle", an amount of time it takes for it to occur again and again, over and over. This is also refered to as a frequency, but it's not the same frequency as the "operating frequency".
When the small coin-sized object is detected, the sine wave cycle is very briefly interupted and the time to reinstate the normal sine wave back to normal is very quick, but when the very large metal object is found, the sine wave is distrubed for a much longer time period. This info is transmitted into a circuitry that measures it all and gives us a reading of the entire width and volume and surface area of the object in question, and quite electronically too, not magnetically or electrically. The eddy current produced has to be analized by our detector before it can give a signal identifying it and it's time duration in a cycle (time period and strength in terms of positive and negative voltage).
Some metal detectors do a better job of dealing with the intruder (the target) in the normal ground matrix, and some do not. Many high-gain detectors like Tejons, 1270's, Minelab Explorers, Garrett 1200 through 2500's, most Aces, etc are already overdriven and should produce better results in mild soil than high iron soil. Multi-frequencies such as Minelab Explorers and similar have to chose which frequency it thinks is best for the soil as it changes here and there, and sometimes a hundred times per second, and that is why they are so slow to operate. That's also why they often miss targets or discriminate poorly if scanning too quickly.
But other detectors such as most Fishers, some White's and most Tesoros have circuitry that compensates for the matrix change due to the interference of a new target in the existing status quo of the ground matrix - much better than most other detectors, especially when used in high-iron soil. Fishers have ALWAYS been noted for their propensity to search deeply in bad soils. So have Compasses and Erik Fosters various detectors.
I hope this helps, and pardon me for having to write so much, but sometimes it is necessary in order to get the full meal deal
