Minelab 8000

[Aziz]

Hi all,

I have finally added an 18 inch DD-IB coil to the coil comparison.
It is not good performing. The TX/RX coils are relative small compared to other coils. This is the reason for.
The zipped Excel-file of the comparison data is attached.

[Aziz]

Hi all,

the Super-D coil comparison (GPZ 7000 coil) would cause me much more work. I have to sweep the target over a large surface area instead of sweeping it at it's maximum response axis line.
At each target height, I have to take the maximum value target response. So I will have a large two dimensional target response table.
Designing the coil won't be an issue anymore however. But putting the target response is a big deal.
Will it be really worth all the work?

[Aziz]

Regarding the Super-D coil:

I don't know any internals how the two RX coils being processed, connected, nor their inductances.
Is this a two-channel processing or are the RX coils connected as anti-interference coil.
As long as I don't know the real answers to the questions, I can not make the coil analysis.

[Carl-NC]

Is this a two-channel processing or are the RX coils connected as anti-interference coil.
As long as I don't know the real answers to the questions, I can not make the coil analysis.

They are combined as anti-interference coils. That's why you can't simply bob the coil to GB, you need that yellow toroid ferrite they include.

[Aziz]

They are combined as anti-interference coils. That's why you can't simply bob the coil to GB, you need that yellow toroid ferrite they include.

Hi Carl,

thank you for the info.

As I have assumed, the coil analysis will cause a lot of work. The Super-D coil configuration wouldn't do well too. So there must be gain compensation in the front-end amplifiers to get the required sensitivity.
Cheers

[Aziz]

Hi all,

I am implementing the 3d cubic spline interpolation. I hope, Google AI didn't make any bugs.
This feature will help reduce the number of scanning points and make a smooth coil geometry and more accurate coil designs possible.

So the evolution and development goes on...

[Aziz]

The 3d cubic spline interpolation core functions do work now.
This is a check for debugging. Source shape is on the left. The right part is the 3d cubic spline interpolation of the single section of a D-shaped wire loop.

I have to continue the coding. It is not finished yet.

[Hristo]

Azis check this innovative coil design is not built by any manufacturer yet:

it striped from russian engineer Eugeny R&D circuit::

His Last shared Circuit for DIY-ers:

[Aziz]

Hi Hristo,

according to the Geometric-Mean-Rule for coils (what a nice name), the coil configuration will not give any benefit.
GM = sqrt(a*b*c*..*..)
The variables a.. could be total flux area of the coils, dimensions (width/height/radius/diameter), etc.

It is very obvious, that the RX coil has quite less flux area. The geometric mean would be quite small compared to the whole flux area of the equivalent round coil.

Coils follow strictly to Geometric-Mean-Rule. One can estimate its comparable performance before making any complex simulations.

Aziz

[Aziz]

Hi all,

the 3d cubic spline interpolation is also capable to reduce the number of wires (not smoothing, but making it more rough).
This makes the wire/path management functions a bit complex. I have to invest more time & effort to find an elegant solution.

It's also a nice feature to speed up the numerical calculations.
Or convenient for making same length wire pieces.

Aziz

[EL NINO]

Aziz ..what about the OO coil...? won't it have better performance parameters than the DD coil?

[ivconic]

Aziz, do you draw those graphics on Commodore C64?
You are a good man, if you need help; I can send you an Amstrad CPC to improve all that a little!
(Sorry, don't take this the wrong way, I'm kind of in the mood for a joke today)


P.S.
Joking aside, if you give me a clearly and concisely written explanation and algorithm of your software;
I can try to write a program that will (maybe) do it better from the graphics side.
I'm getting bored these days, I'm a bit rusty, a little brainstorming wouldn't hurt.

[Aziz]

Aziz ..what about the OO coil...? won't it have better performance parameters than the DD coil?

Hi El Nino,

a search coil has a final total dimension (diameter, size or area). To have an equivalent size comparison, we would shrink the OO coil such that, that it would fit into a final comparable round coil dimension (in this case 18 inch diameter coil housing, appr. realistic 5 mm space left at the edges due to coil housing).
There we have it again. TX and RX round coils would have smaller diameter finally. They might be even worser than DD coil shape.

All coil configurations I have presented here have the same total size (18 inch diameter), same TX inductance (300 µH) and same RX inductance (300 µH). The inductances are normalized to 300 µH. Cheating is not allowed for a fair coil comparison.
Aziz

[Aziz]

Aziz, do you draw those graphics on Commodore C64?
You are a good man, if you need help; I can send you an Amstrad CPC to improve all that a little!
(Sorry, don't take this the wrong way, I'm kind of in the mood for a joke today)


P.S.
Joking aside, if you give me a clearly and concisely written explanation and algorithm of your software;
I can try to write a program that will (maybe) do it better from the graphics side.
I'm getting bored these days, I'm a bit rusty, a little brainstorming wouldn't hurt.

Hi ivconic,

there is a minor bug in the graphics engine yet. But this is not so relevant at the moment. I don't have the time to fix it.
New features are more important.
Aziz

[ivconic]

Of course, if the program gives a true picture of the state of the magnetic field and the interaction between the coils; the graphic display is totally of secondary importance.




The El Nino's question about "00" coils makes a lot of sense to me.