PI coil static shielding

[F117]

First I want to apologize for my ignorance on coils and coil shielding but lately I have been doing a lot of field tests on salty beaches with my PICKINI detector and found out empirically what is probably already explained and documented somewhere on this nice forum:

1. Unshielded flat spiral coil ID 14cm OD 24cm - 30 turns:



Forget about the alu tape, this version is completely unshielded and potted in 2 component resin.
It performed well on dry sand, but a lot less on the wet beach.
Whenever the head touched dune grass or the sand, it gave false signals.

2. Unshielded round coil diameter 20 cm - 27 turns:
Same environment, but no false signals when touching ground or grass.

I assume electrostatic charging of the coil head is much worse with the flat spiral as it has a much larger surface than the siimple round coil.
If static charging / discharging is what causes the false signals, how can this be resolved by connecting a shield to the signal ground of the detector ? I would assume that the electrostatic charge is with respect to the real ground potential, the beach sand. Should there be a real ground potential lead dangling from the detector over the ground ??

With a shielded coil, none of the previously mentioned problems occurred.

Forgive me if this sounds like a dumb question, but can somebody explain to me how a shield connected to a virtual ground can reduce the static potential on the coil surface with respect to the real ground?

Regards,
- Bernard

[Qiaozhi]

Forgive me if this sounds like a dumb question, but can somebody explain to me how a shield connected to a virtual ground can reduce the static potential on the coil surface with respect to the real ground?

The shield is also know as an "electrostatic" shield (there's a clue there) or a Faraday shield. The shield material acts like the plate of a capacitor, with the other plate being the ground. Without the shield, the coil itself becomes a capacitor plate, and any electrostatic discharge will be picked up by the detector. With a shield, the electrostatic charge conducts away to 0V (through the cable screen), and hence is not detected. The detector should be immune to any noise on the 0V line.

[F117]

This makes sense to me :

The detector should be immune to any noise on the 0V line.

- the detector GND instead of the "real" ground potential.

Thank You, George for this explanation.
Best regards,
- Bernard

[F117]

Electrostatic shielding is also explained on p 246 of "Inside the Metal Detector".
Lessons learnt: read the book first from start to end

Best regards,
- Bernard

[Dixix]

So far I have had bad experiences with electrostatic shielding. The problem is the performance takes rapiede from. (1/2 less) I used aluminum tape (http://www.conrad.biz/ce/de/product/...x-50-mm-Silber). Is the aluminum tape is too thick? Is it enough maybe the coil shield only part way? Or must the coil be wrapped completely?




Dirk

[6666]

Did you leave a gap in the shield ?

[Dixix]

Yes and no. It is difficult to explain. Best on photos to see. Here the shield is consistent but not completely encloses the coil. Is that enough? Has anyone ever tried?



If I completely adhesive to the coil with Alutape , then the coil have no more power.

[mickstv]

I would suggest reading the article at the link below.

http://www.geotech1.com/cgi-bin/page...oils/index.dat


Al foil/tape will slow the coil down too much, especially when it's stuck so close to the coil. You need to get some spiral wrap and wind two layers around the coil, then put your shielding material over the top of the spiral wrap.

[Dixix]

Thank you for the super tip on the article by J. Rogowski. Since a lot of mysteries are revealed. For I have also learned that not all cables from the coil to circuit work well. Is also my experience with microphone cable. Probably the wire diameter is too large (to great Resistor i think). There's really a lot of components to consider for a good coil! Also the tip to keep a distance from the shield to wire .....
J. Rogowski describes in his article only "normal" coil. My question: What about the flat - Coil (top -Bernard's flat-coil).
Or the so-called basket - coils. I probably should try it, as these coils respond to shield. I have taken a basket coil for the "PICKINI". This works very well! But without shielding. I have to test it .....
Best regards,
Dirk

[mickstv]

I think the best method especially for flat-coils would be to use a graphite type shield.


Have a look through these theads, linked below.

http://www.geotech1.com/forums/showt...light=graphite

http://www.geotech1.com/forums/showt...light=graphite

http://www.geotech1.com/forums/showt...light=graphite

[bbsailor]

Thank you for the super tip on the article by J. Rogowski. Since a lot of mysteries are revealed. For I have also learned that not all cables from the coil to circuit work well. Is also my experience with microphone cable. Probably the wire diameter is too large (to great Resistor i think). There's really a lot of components to consider for a good coil! Also the tip to keep a distance from the shield to wire .....
J. Rogowski describes in his article only "normal" coil. My question: What about the flat - Coil (top -Bernard's flat-coil).
Or the so-called basket - coils. I probably should try it, as these coils respond to shield. I have taken a basket coil for the "PICKINI". This works very well! But without shielding. I have to test it .....
Best regards,
Dirk

Dirk, and all interested,

See this web link for an online RLC calculator: http://www.calctool.org/CALC/eng/ele...cs/RLC_circuit. This will allow you to see the effect of coil inductance and coil circuit capacitance on the coil's self resonance and see how all this affects the value of the damping resistor necessary to sample as fast as possible. When making coils, the most handy set of tools are a signal generator that goes up to about the 2 Mhz range and an oscilloscope.


Flat coils have less capacitance as the coil turns only lay next to wire on either side of it compared to a bundle wound coil that has wires all around it. However, a flat wound coil will need more turns of wire compared to a bundle wound coil. Basket weave coils also have less capacitance as the wire passes over each other winding in a small area. Flat type coils and basket weave coils are also harder to shield and are less rigid than a well made bundle wound coil which can be made more stable and make less noise when bumped.


Use the RLC calculator to see the effect of coil capacitance, coax capacitance, MOSFET Capacitance and shield capacitance on the damping resistor value that should be in the damping factor range of about 1.1 to 1.15 or slightly overdamped. A good rule of thumb is that each 100pf of coil circuit capacitance reduction equals about 1 microsecond of performance improvement in reducing the delay.

There is no universally good coil that can detect anything anywhere, but there are coil designs and sizes that can be optimized for searching for targets with particular time constant ranges in particular locations with specific ground conditions. Coil design should start with the nature of the targets you are seeking as the first design round. TX pulse width, frequency, coil current and battery capacity are all interrelated.

Unshielded coils typically work with delays upwards of about 25 microseconds as the ground signal is pretty well gone by that sampling time. However as you reduce the delay to seek smaller gold targets, the ground will also be detected as well as any electrical interference from power lines or RF transmitters nearby. On the beach, salt sea water and saturated sand will be detected at about 10 microseconds. Some beaches with volcanic black sand are more difficult to detect on than other black sand that is more organic.

I hope my article helps.

bbsailor

[Dixix]

Thank you for your articles. Many many important informations over coils! This informations I have to sort out in my head. Sorry, some of it is not easy to understand for me, an Electronics - newbie, which is not the youngest .... But it's always a pleasure when I can learn something new.
regards,
Dirk.