Very large DD coil

[Krzychu_]

Thank you for every suggestion. The coil will be relatively narrow and long (30 ... 40cm X 2m), so the field of detection is wide, and the depth of penetration is more or less the same as its width (~ 30cm). It will only be used on flat plowed fields and meadows. I need to ensure its stiffness especially along the length, so I think over the supporting structure in the form of a 2T beam with an additional system of damping vibrations and transferring mechanical stresses.

[kt315]

you can not to balance huge coil in your room just you have much armature inside walls, electric nets/cables, internet cable,
modems etc.
you have to be ready to tune your coil in the field far away out your home.

[Krzychu_]

I know it. I don't know only, what the end result will be
but I will try

[kt315]

a coil from CAT5 cable. will be good in your case. http://www.metdet.ru/korsina2.htm

[Krzychu_]

I'm not sure. What about screening and selection of quantities and unused wires? I prefer winding with enamelled wire, which I have already ordered. I may use monolithic shielding, but I have to do tests for this when tuning the coils.

[Tim]

Another suggestion would be a stiff coil on/in plywood then mounted on a piece on conveyor belting, like the ML drag coils used here in Australia.

[Davor]

Every coil can be scaled, but mechanical vibrations will create increasingly more of false signals with size. Tx coils of free running oscillators are critical for their Q factor, but with a thicker braided wire it should work fine.

[Krzychu_]

What about RX coil?
What determines the inductance and resistance of the receiving coil in multi-frequency detectors? In DFX the same non-resonant coil operates on 2 frequencies (3kHz and 15 kHz). Is the induced voltage the most important (or for example the Q-factor of the RX coil)?

[Davor]

There are 2 conflicting constraints with Rx coils of "normal", and 3 in multifrequency machines.
An Rx coil has its inductance and resistance. When you increase the number of turns by 2-fold, you get twice the resistance, and 4 times inductance (N^2), but also twice the voltage. Effectively you have voltage transformation by the square root of Rx to Tx coils inductances' ratio. Since the Rx input circuitry has a certain input noise equivalent resistance, and only the best of the best rigs have it down to 50ohm or 1nV/sqrt(Hz), it is opportune to raise the voltage and resistance up to the value close to that noise related resistance. However, as you go up with inductance, your coil's self resonance goes down, and there's your constraint.
The other constraint is a coil's weight. With a certain wire thickness your voltage to resistance ratio is constant, and it increases with wire thickness. However, pushing it to extremes would lead to a heavy coil. Reducing the coil's resistance below, say, 30 ohms is largely pointless because there are no front end circuits capable of exploiting lower input resistances noise-wise. If there is an imaginary Rx front end with input noise equivalent resistance at, say, 10 ohm, the best coil would have resistance at about 10 ohms.
The third constraint relevant for multifrequency is related to the need to maintain amplitude and phase across the used frequency range, against the voltage applied to the Tx coil. This has a lot to do with the Tx coil's Q factor on the lower bound and Rx coil's self resonance on the upper bound. Both have to be quite good.

[Krzychu_]

The weight of the coil is not a limitation. It can be heavy because it will be on wheels. I intend to duplicate the transmit coil (TX) to preserve the "factory" parameters L = ~ 540uH and R = ~ 2,7Ohm. I hope that the 4 ... 5 section of such coils will work in parallel. I will use only a higher frequency - 15kHz. My RX DeTech 13 "DD coil has, for example, about RL = 45Ohm and L = 13.65mH, and Eclipse 6X10" DD RL = 60Ohm and L = 13.9mH, which means that the Q-factor should be larger for a larger coil, that is, the diameter of the wire should also be larger. The calculations show that such RX rectangular coil with dimensions of 20X200cm would have about 60 turns of wire. With a diameter of 0.5mm, its resistance would be around 24ohms, and at 0.3mm around 63ohms, so theoretically the variant with a thicker wire seems to be better ... (?)

[Davor]

...or not. Basically you gain nothing with Rx coil resistance below about 50 ohm. It is because the input noise of 1nV/sqrt(Hz) is just as noisy as a 50 ohm resistor.

[Krzychu_]

Is it therefore possible to wind the receiving coil with a wire with a diameter of, for example, 0.22mm and obtain an inductance of around 100mH at a resistance of approximately 315Ohm may be the right solution? The input circuits look as follows:

How to choose the right inductance and RX coil resistance? What criteria and assumptions are important?

[Krzychu_]

I've seen a few photos of several different coils and I'm "a little bit surprised" with their shielding, because I've often seen windings without screening EMI (???) - Is it common practice?

[Davor]

Your Rx preamp in a previous post is not a very good choice.
Regarding shielding, If you build a centre-tapped balanced coil for Rx, you'll be just fine without shielding, but your Rx front end must be able to accommodate such symmetrical arrangement, say, by configuring a preamp as an instrumentation amplifier. Tx coil is always shielded in commercial builds because those must go through EMI certification, and shielding is the way to pass without breaking any sweat. In your personal build you don't have to shield a Tx coil at all

[Krzychu_]

The presented preamplifier is the input stage of the DFX detector designed by Whites and, if possible, I intended to use it without any changes. I am more interested in why it is not a good choice and how to redesign it to work together? The shielding I saw was done carelessly as a graphite layer painted on the inside of the case, only for the relative electrostatic charge compensation, not as an EMI screen, and as you can see in most solutions it probably works (?).