Advanced PI coils

[Aziz]

Yes, the Geonics coil is basically a noise cancelling arrangement, but the configuration can be described as "tophat" albeit a square one. A closer arrangement is that of Corbyn (GB patent 2071327A) filed in 1979. Coils 1 and 2 are TX connected in series and reverse phase. Coils 3 and 4 are noise cancelling RX. The reverse winding arrangement of the TX is to reduce the ground signal, according to the patent, although it could no doubt be adjusted to give induction balance.

Eric.

Hi Eric,

yes, I know this patent already. The Corbyn coil differs from our tophat coil. Our tophat coil here is a true new invention and achieves superior performance. No counteracting of TX by BX (bucking coil) - the opposite happens. Better pin-pointing feature. Higher sensitivity to shallow & small targets. Fully anti-interference feature. Easy to build and balance. Can be made large. And goes deeeeeeeeeeeeeeeeeeeeeeeep.
There is sweet (but candyless ) magic in this coil.

The german coil patent differs too. A total different coil.

Cheers,
Aziz,
the madman, magician, anti-patent-troll and inventor *LOL*

[Ferric Toes]

Another coil patent filed in Germany (DE 3615652 A1) in 1986. I can't read German but Fig 5 suggests sine and square wave drive (Impulserregung in text). 32 and 34 are Integratoren.

Eric.

[Ferric Toes]

Ok guys,

I have found an interesting paper from Geonics and their used coil configuration:

http://www.geonics.com/pdfs/technicalnotes/tn32.pdf

(see figure 1 on page 6)

The EM63-MK2 is of the same type to EM61 but the TX coil is larger. A typical PI machine. Rx1 and Rx2 are connected in anti-interference configuration. But it isn't an induction balanced coil. It's a different coil.

No wonder, the Tophat(c)(r)(tm)(blabla) coil is quite complex to be invented easily. It hasn't even been invented by the famous guys from ML.

Cheers,
Aziz

Yes, the Geonics coil is basically a noise cancelling arrangement, but the configuration can be described as "tophat" albeit a square one. A closer arrangement is that of Corbyn (GB patent 2071327A) filed in 1979. Coils 1 and 2 are TX connected in series and reverse phase. Coils 3 and 4 are noise cancelling RX. The reverse winding arrangement of the TX is to reduce the ground signal, according to the patent, although it could no doubt be adjusted to give induction balance.

Eric.

[Davor]

Because I have great expectations for the differential coil. Having a coil that takes care of the most of the interferences and also that cancels the ground signal ... well, I want such a coil. Standard concentric just doesn't cut it.

I'm up to designing a 4-quadrant Rx that will have no problems using a DD differential coil, as the one the animation starts with. It is worth it.

[Tinkerer]

Could you see about this configuration anyway? In case there is not much extra to compensate, perhaps there are some non-inductive ways to do that in order to keep the anti-interference goodies.
Say, if Tx coil is at 1.1*r ?

Why not use a standard, concentric, co-planar IB coil configuration with the RX coil center tapped? R, RX=0.5R TX

[Davor]

Could you see about this configuration anyway? In case there is not much extra to compensate, perhaps there are some non-inductive ways to do that in order to keep the anti-interference goodies.
Say, if Tx coil is at 1.1*r ?

[Aziz]

..
My doubts are in the coupling between Tx coil and the outer Rx coil. I'm afraid it could screw the balance due to proximity. Aziz, please check the balance of this contraption with your fabulous software.

You are right with your doubts.
Nearby the TX coil, there is a high magnetic field strength, which will be induced by the outer RX coil. In the center region, the magnetic field strength is weak and you would need much more turns to compensate the outer RX coil.

Balancing is in principle possible but with different turns count of the RX coils (or radius). This would mean, that you screw out the anti-interference feature.

Aziz

[Davor]

I might have something very nice. It is based on Sergey_P's work on the concentric differential coil. I already had some discussions with Sergey on this one and for the time being neither of us did it a go in real life. To shorten the discussion I made an animated gif picture of it. Lets christen it Davor-Sergey coil.



The animation begins with a well-known differential "tanga string" coil. there are two D's, blue and green, one against each other, fed in anti-series. Around them is a TX coil, red. Phases of the Rx coils are in reverse, and the whole shebang is balanced. Also it cancels far field electromagnetic noise, and also a ground signal. And it also has a drawback - one of the Rx coils is in counterphase and that creates problems in Rx.

The Rx coils are then transformed, Packman style, and the left D consumes the right one. Area ratio is maintained, so that in effect it is still a differential coil with anti-interference and anti-ground-signal features. A step before the final one is a configuration that has the previously right coil in the middle and consuming half the area, and the left coil encircling it, but in a shape of a belt. The clamps connecting the inner part of the left coil and the outer part have currents in opposite, and can freely be discarded.

Now the final stage... Both Rx coils started with N turns, thus the outermost coil still has N turns. It's inner part has also N turns. And the other coil has also N turns. And all of them are connected in anti-series. So in effect we have outer Rx coil with N turns, and the inner coil with 2N turns. Radius ratio is 1:sqrt(2).

A bonus feature... if this is used as a monocoil (e.g. just discard the Tx coil and use the differential coils as a monocoil), there surely is no influence between Tx and Rx. It will function as a differential coil both ways, Tx and Rx, but it will also cancel far fields and to some lesser extent the ground. Why? because to cancel ground it would require a non-differential Tx coil.

My doubts are in the coupling between Tx coil and the outer Rx coil. I'm afraid it could screw the balance due to proximity. Aziz, please check the balance of this contraption with your fabulous software.

[sido]

Ok, thanks for the detailed analysis.
I understand the principle of operation and does make a very quiet coil, seeing TX and (RX) are both balanced.

Sid

[Aziz]

Hi Aziz, How is it easiar than a concentric coil when you still need to null the Transmit with bucking coil?

How do you determine the distance apart between the 2 recieve coils and whats the ratio of turns between the two RX's?

Sid

Ok, I know which coil do you referring to now.

The difficulty in making a standard concentric co-planar coil is it's final fine balancing. This is usually done by extra small loop in the BX/RX+ coil region. This isn't required with the offsetted RX- coil. And you have a very quiet anti-interference coil too.

You first place the RX- coil to the desired planar distance (minimum distance should be the diameter of the TX coil). Then you adjust the co-axial distance between the BX and RX+ by putting more or less distance spacers until you get a rough residual minimal total RX voltage. Of course, you have to try some BX coils (with varying turns count) to achieve the minimal residual voltage condition.

If you have found a reasonable BX coil with reasonable distance gap to the RX+ coil and roughly balanced it at the desired distance of RX- coil, then the variation of the distance between the concentric co-planar coil assembly and the planar RX- coil gives you the possibility to fine balance finally. That's the trick to have a fine balance means.

You can't do the fine balance with distance variation of BX and RX+. Its too course and minimal variations lead to big TX voltage induction at the end. And fiddling with extra loop, well, ... (ok, would work but hey!, we are going to have the benefit of anti-interference feature here and fiddling isn't really required by distance variation of RX- coil part ..)

Standard concentric co-planar coils wouldn't work with additional RX- coil as it is already balanced to the RX+ coil alone. We have to make the whole coil assembly by ourselves of course.

Aziz

[sido]

Aziz to the picture above i am referring.

Sid

[Aziz]

Hi Aziz, How is it easiar than a concentric coil when you still need to null the Transmit with bucking coil?

How do you determine the distance apart between the 2 recieve coils and whats the ratio of turns between the two RX's?

Sid

Sid, which coil are you referring to?

[sido]

Hi Aziz, How is it easiar than a concentric coil when you still need to null the Transmit with bucking coil?

How do you determine the distance apart between the 2 recieve coils and whats the ratio of turns between the two RX's?

Sid

[Aziz]

Anyone want more good AI IB coil designs?
But is a prior art (almost 2 years ago published).
Remembering my comments here?

http://www.geotech1.com/forums/showthread.php?17341-EXTRA_EXTRA_DEEP-DISCRIMINATING-DETECTOR&p=122467#post122467

Same principle. But not co-axial arranged. Just planar offsetted RX- coil.
Very easy to build and balance. Easier than the simple concentric co-planar coil.

Just to give you an idea how it is arranged:


A forgotton good idea.

Cheers,
Aziz

[Aziz]

Al-foil Strip

Hi all,

I have found the picture of the folded Al-foil strip (see below):


And I have found the exact specifications:
It's made of 13 µm thick Al-foil, 30 cm width and 30 m length. The whole folded 30 m Al-foil has a resistance of 0.2 Ohm total. And when fully folded, the stripe has a width of 5-6 mm.

The whole 30 m Al-strip cost me about 2 EUR material. (And some millions of nerve cells for folding.. ).
Cheers,
Aziz