Coil-block communication interface

[JoyJo]

Greetings everyone!
When analyzing the methodology for constructing metal detectors, one cannot ignore the fact that the receiving circuit is placed directly in the metal detector coil. For example, Minelab Equinox. Attached is a circuit diagram of a board with a receiving stage made from an Equinox coil. Yes, this diagram contains an error, but it is enough to understand the general principle.
I would like to clarify:
1. What are the advantages of placing the receiving circuit directly in the coil?
2. What are the advantages of using a disserial data interface?

As far as I understand, placing the receiving stage as close as possible to the RX coil will reduce the level of input noise, and the use of a differential communication interface with the unit will avoid noise pickup from nearby wires for the TX coil.
Are there any other advantages to these solutions?

In addition, an interesting question regarding the use of ADC type in metal detectors: which would be more preferable with a pre-differential input or with a fully differential input?

Thank you for your attention!

[ivconic]

1) S/N
2) Copyright protection
3) Coil ID

[Skippy]

Cables can suffer from 'microphony' , where flexing the cable generates a voltage. If the coil signal is boosted up 15 times, for example, the effect of the microphony is similarly reduced.
The Eqx does have differential transmission up the cable, but unfortunately, the pair of wires are not twisted together, not are they screened from the other wires and the outside world interference.
Other detectors have more than just the pre-amp on the PCB. The Manticore has ( it is assumed ) an acellerometer or similar inertial sensor on the board. Other possibilities when you have a serial interface ( and a microcontroller ) would include selectable pre-amp gain, possible electronic nulling of the induction-balance.

[JoyJo]

Yes. I also noticed that the data transmission wires are not twisted or shielded in the Equinox. Country, of course. I think they just selected the right cable from the existing ones, and did not order the development of their own. I am considering precisely the twisted 2 veins in one screen.
Removing the induction balance. Can I learn more about this feature? Are there any articles or patents?

[bbsailor]

Greetings everyone!
When analyzing the methodology for constructing metal detectors, one cannot ignore the fact that the receiving circuit is placed directly in the metal detector coil. For example, Minelab Equinox. Attached is a circuit diagram of a board with a receiving stage made from an Equinox coil. Yes, this diagram contains an error, but it is enough to understand the general principle.
I would like to clarify:
1. What are the advantages of placing the receiving circuit directly in the coil?
2. What are the advantages of using a disserial data interface?

As far as I understand, placing the receiving stage as close as possible to the RX coil will reduce the level of input noise, and the use of a differential communication interface with the unit will avoid noise pickup from nearby wires for the TX coil.
Are there any other advantages to these solutions?

In addition, an interesting question regarding the use of ADC type in metal detectors: which would be more preferable with a pre-differential input or with a fully differential input?

Thank you for your attention!

Your point #1.

The advantages of placing the receiving circuit directly in the coil are the following.
1. Ability to lower the coil seen capacitance by lowering the impedance going up the wire to the control box.
2. Lower coil seen capacitance allows a higher damping resistor value.
3. A higher damping resistor value allows faster RX signal sampling of smaller or lower target time constants.
4. First amplifier/ buffer gain can be controlled based on the operating environment soil condition or noise pickup.

Most coil coax cable has about 100 pf of capacitance that can be mostly eliminated by engineering the first active stage in or near the coil.

Joseph J. Rogowski

[Carl-NC]

With the preamp in the coil it now drives the RX cable with a low impedance so the cable selection is less critical. The old X-Terra series used ordinary CAT5 cable for the coil.

Placing a micro in the coil let's you ID the coil so you can alter detector settings accordingly. For example, depth estimates depend on coil size. Once you have a micro, an accelerometer becomes easy and offers a lot of opportunity to automate DSP. For example, pumping the coil for GB, or short-sweep averaging. See US9207315 for some more ideas.

I've done a fair amount of work on electronic nulling, it isn't easy especially for a wideband coil. You can either do it as a one-off nulling (say, just in the factory, or by some special enable method) or as a continuous nulling, so that even ground gets continuously nulled at the coil. The latter produces a motion-type nulling which may create problems. I've built some e-null coil prototypes but never one I liked or thought was manufacturable.

[JoyJo]

Thank you for the answers, they are very informative. So I will continue to work towards placing the receiving circuit in the coil. Does it make sense to consider transmitting data over a differential cable using specialized chips? Or can we limit ourselves to simple converters of a unipolar signal to a differential signal and vice versa built on operational amplifiers?

[Carl-NC]

The advantages of placing the receiving circuit directly in the coil are the following.
1. Ability to lower the coil seen capacitance by lowering the impedance going up the wire to the control box.

In a PI with a mono coil, you would need to place both the TX driver and the preamp in the coil since they share the same cable. Placing just the preamp in the coil doesn't help much. And placing the TX driver in the coil has its own problems with metal and clean power. If the PI design only supports split coils then putting the preamp in the coil will help with EMI and such, but not so much with faster damping as most of that resides on the TX side.

[Carl-NC]

Thank you for the answers, they are very informative. So I will continue to work towards placing the receiving circuit in the coil. Does it make sense to consider transmitting data over a differential cable using specialized chips? Or can we limit ourselves to simple converters of a unipolar signal to a differential signal and vice versa built on operational amplifiers?

As far as I know, XP is the only company who does data conversion inside the coil and transmits it out digitally (and wirelessly). Everyone else has the preamp in the coil and the ADC in the control box. That's the way I would go. Otherwise, you will likely need to use something like LVDS to get the data to the control box.

[JoyJo]

I am considering a VLF metal detector with a DD coil. At the same time, I plan to use the full H-bridge as a TX, which will be located in the block. The receiving circuit is supposed to be placed in the coil. And transmit data over a differential line.

[JoyJo]

I was supposed to use something similar.
In this case, a differential buffer with G = 1 will be installed in the control unit in front of the fully differential ADC.

[Carl-NC]

I don't understand. Are you wanting to drive an analog preamp signal from the coil to the box, or an ADC digital signal?

[JoyJo]

I propose to connect the RХ coil to the receiving stage, convert the unipolar output signal into a differential one. The differential signal is transmitted via cable. In the control unit, the differential signal is sent to the buffer stage, and after the buffer stage, the signal is sent to an ADC with a real differential input.

[Carl-NC]

I see now. A differential coil signal may help a little but you will want a shielded twisted pair in the cable which might be hard to find. Might be a whole lot easier to just use a single-ended signal.

[JoyJo]

I have such a cable at my disposal. It meets all the requirements. it remains to add two more isolated cores to it for the TX signal. 1 pair for signal transmission, 2 pairs for supply voltage. And two more insulated conductors for the TX signal. Combine all this into one shrink tube and you will get a ready-made cable.