For Mikebg

[6666]

Hi Mikebg

I have been trying to link together some of your comments that have been spread over several threads over several years, in which you talk about using the LM386 and this comment "search head contains IB (induction balanced) TX and RX windings."

from this post
http://www.geotech1.com/forums/showp...67&postcount=2

1. Does this LM386 schematic match this coil ?
2. if so is there any tuning required of the coil ?
3. Do you have any information on the wire and turns of each coil


http://www.geotech1.com/forums/showp...2&postcount=18
The text for the head (coil)
1 is the first RX winding,
2 is oval nonconductive sheet (plastic, hardboard or plywood),
3 are consoles made of plastic,
4 is the plastic box containing bad ground,
5 is TX winding,
6 is the second RX winding and
7 are plastic nuts and bolts.

thanks
6666

[mikebg]

Hi 6666
The short answers are:
1. Does this LM386 schematic match this coil ?
A. Yes, both RX coils of this search head operate as a tank circuit with center tap. However LM386 is not the best IC for a wideband gold detector because the maximal frequency is below 20 kHz. To find small nuggets, you need other ICs with input symmetry and more high maximal frequency. Unfortunately they make more battery drain.
2. if so is there any tuning required of the coil ?
A. Yes, damping with resistors of all three coils is required to make almost flat frequency response of the system in the frequency band 2.5 kHz ... 18 kHz.
3. Do you have any information on the wire and turns of each coil.
A. Yes, but my coils are designed for meteorite detector and for measurement purpose at audio frequencies. We (you and I) should design and test other search head and other preamp circuit for a prospecting detector.

[Davor]

I thought of some kind of double balanced scheme where Tx coil would also be split and put into some kind of "lucky clover" design. So I would end up in two pairs of mutually orthogonal coils. In such a way Tx emission would be localised to a search area (less interference to other MDs in vicinity), each of the Rx coils would be balanced against a pair of Tx coils, hence twice the balance.

Ofcourse, I didn't test it yet. Still lab-less

[mikebg]

I thought of some kind of double balanced scheme where Tx coil would also be split and put into some kind of "lucky clover" design. So I would end up in two pairs of mutually orthogonal coils. In such a way Tx emission would be localised to a search area (less interference to other MDs in vicinity), each of the Rx coils would be balanced against a pair of Tx coils, hence twice the balance.

Ofcourse, I didn't test it yet. Still lab-less

Hi Davor. There are lot participants in this forum which are lab-able. Can you sketch out your idea with a drawig to discuss it in the forum?
My idea mentioned in the post #2 is to use as first stage of preamp a video amplifier
as uA733 or NE592 to reach full swing amplification to 200 kHz.

[Davor]

I think this drawing will show you what I have on my mind. There is also some room for pinpoint sensor in the middle.
Why Tx in anti-parallel AND orthogonal configuration? This configuration does NOT work as a VLF antenna, but more like a humbucker electric guitar pickup, and it may prove beneficial for self-oscillating Tx and otherwise (that work in fact as a kind of Tx/BFO/PWM mutant). Besides, this configuration should produce a very localised field, and perhaps more sharp detection ... I guess.

[6666]

Hi Mikebg
thanks for the reply, there are some interesting ideas in your posts about 386
will have a closer look at them
thanks
cheers
6666

[mikebg]

Davor, please show in the drawing where is the target. To generate a strong signal, the target should be illuminated by a strong TX field. Search the WEB for "humanitarian demining". You will find similar search head solutions in mine detectors.

[Davor]

I figure that much - I'm new to this technology, and it has been around for some time. Someone must have done it already.

Anyway, I expect to have a very strong field beneath a centre of the whole thing, but with a sharp notch at the very centre in a shape of "+". There should not be any doubt whether you passed over a target or not. So far I didn't play with simulations or the real thing, however it works perfectly well in my head

There is an easy way to decomposition of the working principle of double coils - they are a dissection of a torus. I expect the field to behave in great detail as a leaky torus as well. In a far field the contributions of coils are cancelling each other, and this arrangement should work fine in environments with lots of electric noise.

The arrangement as per my drawing should be perfectly orthogonal. Points and x-es indicate field orientation.

[mikebg]

Davor, I can not understand where the target is illuminated by strongest TX field.

[Davor]

Just about where you put it on your drawing A bit closer to the middle would be a notch.

You may observe this arrangement as a leaky torus. Due to the symmetry, which is beneficial for establishing balance, at the very axes of symmetry between coils you would observe cancellation, hence notch.

I ruined your drawing with some odd circles that would indicate H field. If I become friendly with FEMM any time soon, I'll be able to put it in nice colours. However, I have some other things to do as well.