4 quadrant VLF discrimination

[Sergey_P]

The slope of the notch filter for phase Veratore, determined slope GEB ( cos(x) ). A method of reducing the window of phase I suggested in the topic http://md4u.ru/viewtopic.php?f=77&t=5497

[golfnut]

Doesnt the Gold response live in the thundering great Null though!

S

[Davor]

No, but quite near it. A quality notch at sea water vector would lower the gold amplitude by a little bit, but will also increase contrast for small gold. So it is lose a little, win a great deal.
In VLF you can apply such a sharp notch with little effort, and it can be very precise. Such notch is always present with PI rigs, and not that sharp at all. To make it sharper you have to sample very early, but there are limits. So far PIs were sold by pretence they work well on the beaches, and the nasty bit about inability to sniff small gold is usually not voiced as loud. I think this application of notch in VLF will work better than PI on the beaches.

The slope of the notch filter for phase Veratore, determined slope GEB ( cos(x) ). A method of reducing the window of phase I suggested in the topic http://md4u.ru/viewtopic.php?f=77&t=5497

You may improve slope by applying some compression too. OK, it is level determined, but as accurate as any weighting function. I applied time-accurate compression in IGSL of mine and it's phases are immaculate from closest proximity to the reception limit. IMHO with the wild variation of signal levels, a precision weighting is a bit of an overkill.

[Sergey_P]

Making a clipping window phase - you cut a small gold.
Need to focus on conductivity, and not on the phase

[Ferric Toes]

Hi Dave,

Not related to this thread, but relevant to your last couple of sentences. The lightest weight speaker I ever saw was a helium filled party balloon made of piezofilm. The "string" contained two fine wires to the drive amplifier on a desk, and the sound was true omnidirectional and point source. Must be something about it that didn't make practical sense.

Ferric.

[Ferric Toes]

I'm not sure how a VLF machine as described would cope with sea water. Experiments that I did in the past with PI showed that both amplitude and time constant changed considerably depending whether you were just on a wet beach, up to your waist, or diving at 50ft. Water in the Mediterranian Sea is also saltier, with a higher conductivity, and hence longer time constant. Near a river estuary the opposite situation applies. It also varies with different search coil diameters. Changes in time constant and amplitude would surely indicate that phase angle would also be changing in the frequency domain. One good thing about iron mineralised ground is that the decay is always the same (give or take a small percentage) and it is only the amplitude that changes.


Ferric.

[Davor]

Sure, there are some differences in phase angle related to the salinity, but sea water is still a very bad conductor, worse than the worst metal, with meter thick skin depth, diamagnetic, and hence the resulting water response vectors are tightly grouped in a very narrow span of angles. I guess when excited by some higher frequency you could observe some little heeling of the vector with salinity, yet still quite apart from, say, gold flakes.

Making a clipping window phase - you cut a small gold.

The idea is not to make it a window but a real sharp notch, and that's precisely why I need a nice analogue amplifier. But sure, if I overdo the signal disappearing trick it would affect small gold in dry areas too, but on the beaches such small gold signal is overwhelmed by sea water response anyway. Only the detecting technologies with sea water disappearing tricks do the business there. PIs are for example deaf to such short taus.

[Sergey_P]

Multiply: X*p ("magnetic susceptibility" * "conductivity") ~ τ (tau target): sea water, gold - very little. PI such targets - cuts off (small times)

Small target - a small phase shift ...
The difference of sea water and the target - the conductivity! The response is proportional to ~ p*w^2 ("conductivity"*"square of the frequency") - increase the frequency of.

[Ferric Toes]

PIs are for example deaf to such short taus.

Yes, existing commercial PIs for beach and water hunting are deaf to short taus because of the overwhelming response from seawater due to it's huge effective volume. However, PI as a principle can be almost as sensitive as a high frequency induction balance detector. I built a test unit for beach work that sampled at 5uS after the initiation of the TX switch off and it was almost unuseable on wet sand due to the water conductivity, unless you were very careful about holding the coil parallel at a set height to the sand. 10uS delay is about as short as you can go for beach hunting and with the coil in about a foot of water. For diving work 30uS would be the limit for a 10in coil and as coil size increased you would need to back off the delay still further. I've considered ways of reducing the water signal by using a differential coil arrangement similar to that used for cancelling low frequency em noise, but never tried it. Maybe a notch system could be designed such as is used in PI for ground cancelling. Still much to do.

Ferric.

[Davor]

Multiply: X*p ("magnetic susceptibility" * "conductivity") ~ τ (tau target): sea water, gold - very little. PI such targets - cuts off (small times)

Small target - a small phase shift ...
The difference of sea water and the target - the conductivity! The response is proportional to ~ p*w^2 ("conductivity"*"square of the frequency") - increase the frequency of.

More or less correct. Small gold taus are related to the flake thickness because it is smaller than the skin thickness for gold. There sure exist a flake with a tau that is so small it overlaps with sea water ... so you'll not be able to detect it by any detector with non-differential coil on the shore. Due to the irregularities on the sea bed and sea surface you can't expect even a differential coil to wipe the sea signal entirely.

Anyway, point taken, I'll make the sea water masking optional. Multiplying a real "all metal" signal with "1" instead of the imaginary does the trick.

[Davor]

I built a test unit for beach work that sampled at 5uS after the initiation of the TX switch off and it was almost unuseable on wet sand due to the water conductivity, unless you were very careful about holding the coil parallel at a set height to the sand.

Precisely! It is very same with regular VLFs all the time
PI has a time smearing of the strong responses so if it is deaf to sea water, it is surely deaf to small gold too. Ground balancing schemes are normally done for larger delays. Guess this sea water balancing would require some very early sample to make it kick?

VLFs tend to have neat phase responses that do not smear. I think that is the ground for pursuing this 4-quadrant Rx thing with differential coils and perhaps a masking system for notching salts and ferrites. It seem to me that it is an approach with better energy efficiency, discrimination, and perhaps even depth once this masking system kicks in.

Given opportunity a VLF Tx can supply fields comparable to any PI, yet in a very energy efficient way. Noise-wise VLFs can also outperform PIs by high margin on the bandwidth grounds. So far only QED has done something about it in PI by means of integration. There is nothing wrong with VLFs except that they don't accept differential coils, and a 4-quadrant Rx will remedy that.

I'll make a sim of this masking system. I also have some interesting multiplier on my mind to make it kick.

[WM6]

Hi friends, would you be so kind and go in Land? I hate salt sea moisture in my shoes.

What of benefits we can expect from 4th quadrant VLF discrimination in Land? Is this feature applicable to VLF IB (coil) technology only?

[Aziz]

...
There is nothing wrong with VLFs except that they don't accept differential coils, and a 4-quadrant Rx will remedy that.

I'll make a sim of this masking system. I also have some interesting multiplier on my mind to make it kick.

Our magnificent(tm) and magic(tm) cross-coupled balanced LC oscillator can do it. That's a bang master to the targets.
L: Center-tapped TX coil, center tap connected to gnd
C: Parallel connected to the whole TX coil or splitted (left+right cap connected to gnd).

But be careful. The differential voltage can go well beyond the transistor specification limits (use high voltage transistors or clip the high voltage at the expense of additional power consumption). And you need two emitter inductors (not coupled, individual toroid cores) to get better efficiency.

RX coil can be any (single ended or differential).

Aziz

[Davor]

You must have switched to something stronger lately

Hi friends, would you be so kind and go in Land? I hate salt sea moisture in my shoes.

What of benefits we can expect from 4th quadrant VLF discrimination in Land? Is this feature applicable to VLF IB (coil) technology only?

4-quadrant processing will be able to accommodate any IB coil you can think of, but the main advantage is in applying the differential coils that do not make much sense in nowadays 2-quadrant rigs. These cancel any far field EMI and all non-differential signals, including ground.
The sea water masking thing is my idea of a cure for the remaining obstacle in providing a nice ground balance for VLFs. Multi frequency rigs use it already by means of subtraction of the sea water response at different frequencies.

[Aziz]

Hi Davor,

but I'm not happy with the npn-transistors power loss. They get warmer than the pnp ones (10 times more power loss). Nevertheless, the center-tapped TX produces a lot more bang to the targets and the differential voltage (tp2-tp1) goes up.
The switching transitions could be better though.
Aziz