ARMD (ARMRADIO based Metal Detector) VLF IB PROJECT

[moodz]

I asked Carl to remove those 4 posts as they are clearly out of place.
And it is because of my poor understanding of your idea.
I am sorry for that. Will not be repeated.

No one should be criticised for being passionate about something. You have made alot of very important points that will have to be considered at some point and we need people with your passion.

In summary I guess the fastest way to get up to speed is to use the discovery boards .. if we have to wait for custom boards interest will drop off.

moodz.

[dbanner]

I take it you intend on using a photo resistive device to implement the AIB?
Or AIB on a typical coil arrangement with just a single Rx coil?

[moodz]

I take it you intend on using a photo resistive device to implement the AIB?
Or AIB on a typical coil arrangement with just a single Rx coil?

Being a PI man ... I have a loose understanding of AIB however the bucking coil or balancing winding current feedback must balance without affecting phase ?
Would the changing resistance of the photo resistive element affect the phase ( and thus target ID ).
Not sure ... if you have some ideas on how to implement AIB please do.

I was thinking of a variable current source ( high impedance ) in the FB path that would not affect phase as the frequency changes ( AIB is flat from around 1 khz to 60 Khz ).

moodz

[dbanner]

Two Rx coils both wound in same direction co-located with buck coil.( Above and below buck coil)
Rx 1 is under-nulled(more windings)
Rx 2 is over-nulled(less windings).
Capacitative effects could be cancelled in such arrangement.
And photo device using feedback with as close to a linear response with some sort of linear response correction circuitry.
Ground mineralization elimination would be dynamically implemented.
Maybe a dual gate photo transistor could be used with a difference amp?

[dbanner]

Seems like AIB is best implemented with 2 Rx coils. Noise and phases can be cancelled in front end analog circuitry. Software would do most of the heavy lifting for photometric nulling and ground mineralization elimination while obviating overload.
I can't envisage such a scheme with just a single Rx coil.

[moodz]

Seems like AIB is best implemented with 2 Rx coils. Noise and phases can be cancelled in front end analog circuitry. Software would do most of the heavy lifting for photometric nulling and ground mineralization elimination while obviating overload.
I can't envisage such a scheme with just a single Rx coil.

Sounds good. Have you an idea in mind for the circuit implementation ?

[moodz]

Attached is the ARMRADIO project that will compile with the ( V1.14 ) STM32CUBEIDE development platform.
Its a port of the original sources and it does not really use the HAL ( hardware abstraction ) current libraries.
A STM32F4-discovery board is required for this version. The MCU is an STM32F407VG.
There is a schematic in the project to connect a rotary encoder / LCD and some pushbuttons.
Remove the .zip extension from the z01 / 2 / 3 file names. Its a big zip file brokent down to 3 parts due to forum upload size limits.

This is for making a radio not a metal detector

[dbanner]

Nulling of imbalance with single Rx coil arrangement. Uses a μC

[Altra]

Dbanner, This Whites Electronics patent by John Earle looks interesting US9989663.pdf

Maybe Carl worked on this project?

[Carl-NC]

I worked on several ideas for electronically nulled loops and discussed them with John, I don't remember that one in particular. I'm going to break this topic out into it's own thread as there are several ways to approach the problem, depending on what you want to accomplish.

[Altra]

When you open a new thread, I have a nulling method to present.

[moodz]

The ARMRADIO base ...provides no means to sync the transmit signal it is recieving. ( which is need for the VLF tx coil )
The "Local Oscillator" LO signal in the AR code is a computed vector that is recomputed in real time and multiplied by the incoming signal vector from the ADCs to produce the I an Q vectors for phase and amplitude processing.
The task then was to modify the code so that the LO is brought out via a DAC to the real world so we can use it for the transmit signal in the VLF detector.

( it also means we have goodies like chirping or sweeping the signal without phase error being introduced )
A tone multiplex system will suppor the AIB function.

The stm32f407 has a 12 bit DAC and below is a spectrogram of the tone from DAC2 which is the internal LO ( real ) vector of the IQ demodulator.

The transmit frequency in this case would be 46.5 Khz.

To transmit and recieve multifrequency ... its a simple calculation change in the LO vector code.

[moodz]

...found a bug the trace had a sampling overrun by 1 count . Much better spectral performance for DAC2 with the bug corrected.

Note this is a real trace off real hardware. not a simulation.

[moodz]

The digital local oscillator is brought outside to the real world for TX Coil signal.

[dbanner]

Tone multiplexing. Does this mean that no external circuit will be required for AIB?