THE ELUSVE 1us TC TARGET

[Tinkerer]

Hi Jeff.
thank you for helping me verify the correctness of my reasoning.

The shielding of the coil and cable helps protecting the signal from EMI and static electricity.
I consider the shielding good enough, when I can touch the cable and the coil with my hand and when I can touch the coil with a statically charged PVC pipe (several thousand volt).without getting a significant interferance on the reasonably amplifield signal. Gain of about 1000.

At the moment of the Flyback, the voltage in the cable and coil is 1000 volt. The capacitance of the cable and the shield to coil is charged to that voltage, but there is a low resistance path for discharge. I see no problem.

The inductance: The cable and coil shield exposed to the di/dt of the TX current gets charged with eddy currents. These eddy currents take some time to discharge because of the low resistance path. This discharge combined with the cable capacitance sets up oscillations.
This is one of the reasons I like to use relatively low TX current 1A or less and use 60 to 100 turns on the TX coil.

All that has only significance when very high sensitivity and delays to first sample of less than 5us are desired.

By the way, I have developed a demodulation of the target signal also during the Flyback. The sensitivity is fenomenal, but I have not yet found a way to do the ground balance.

[PICON]

Hi Tinkerer,

I am interested in what you have seen regarding TX/RX coil oscillations and what you have tried to resolve them, as these oscillations can affect early RX sampling.


"The inductance: The cable and coil shield exposed to the di/dt of the TX current gets charged with eddy currents. These eddy currents take some time to discharge because of the low resistance path. This discharge combined with the cable capacitance sets up oscillations. This is one of the reasons I like to use relatively low TX current 1A or less and use 60 to 100 turns on the TX coil."

These oscillations, are they a decaying ring that is not frequency related to the RX or TX coils self-resonance and cannot be suppressed?



"By the way, I have developed a demodulation of the target signal also during the Flyback. The sensitivity is fenomenal, but I have not yet found a way to do the ground balance."

So, you have tried a ground balance technique and it’s not working?



Jeff

[Tinkerer]

Hi Jeff,

The oscillations decay fast.
In 4 days I will be home again. I have an open DD coil and TX and breadboard preamp on the bench. It is a good occasion to look at the scope screen input signals.
Yes we have worked on a digital ground balance, but got busy on another PI system and left the LFD demodulation (as we called the demodulation during Flyback) on the shelf.
I will be glad to return to it, if I find somebody willing to help me.

[pito]

0

if I find somebody willing to help me. = you should post coil schematic and photo

[PICON]

Hi Tinkerer,

"I will be glad to return to it, if I find somebody willing to help me."

So what type of help do you require?


Jeff

[Tinkerer]

0

if I find somebody willing to help me. = you should post coil schematic and photo

Pito, thank you for the driven shield picture.

You want to see a coils schematic and a photo. Please explain, because I do not understand what you want. I have posted pictures and parameters of my DD coil on the thread AMX COILS.

[Tinkerer]

Hi Tinkerer,

"I will be glad to return to it, if I find somebody willing to help me."

So what type of help do you require?


Jeff

Hi Jeff,
thank you for offering help.
About 15 to 20 years ago you helped me with a MACRO for my timing loops. Thank you for that.

At present I want to work on the AMX project. For this I will need a timing schedule and digital signal processing.

I also mentioned a project that we shelved. A Proof Of Concept (POC) was built and tested in the field. It showed hypersensitivity to the smallest targets and ferrous /non ferrous discrimination, but it also showed high sensitivity to the ground. It definetely needs a Ground Balance. This project demodulates during the Flyback, as well as during the whole cycle. I think the project has potential, but still needs some work.

[PICON]

Hi Tinkerer,

You have a good memory regarding the macro but I think it was about 10-12 years ago.

I have some time free at the moment so, what microprocessor are you using for AMX and do you have any code for it already?

Jeff

[moodz]

Just what is the elusive 1 us target and why is it so important

[Tinkerer]

Hi Tinkerer,

You have a good memory regarding the macro but I think it was about 10-12 years ago.

I have some time free at the moment so, what microprocessor are you using for AMX and do you have any code for it already?

Jeff

Hi Jeff,
I have been using a STM Nucleo 446RE for the microprocessor and ADC for the AMX. I have a proprietary code on loan. The owner of the code does not give the code to be used on an open source project. He wants to sell the program. So I need another code to be working on the AMX project, as it is a public community project.

[Tinkerer]

Just what is the elusive 1 us target and why is it so important

The purpose is the AMX project. I consider a 1us TC gold nugget to be the smallest target of interest for a professional gold digger metal detector.

[PICON]

Hi Tinkerer,

What do you need this code to do for starters?

1. TX ‘H’ bridge drive signals. What cycle timing do you require and what FET drivers and FETs are you using so we can work out how much “dead time” is required.
2. Internal 12bit ADC sampling sync to TX timing. What sample rate do you require?
3. How do you want to monitor the digital signal processing?
4. ??

Jeff

[moodz]

The purpose is the AMX project. I consider a 1us TC gold nugget to be the smallest target of interest for a professional gold digger metal detector.

I was more interested in a table of standard targets so that we can compare the performance to commercial detectors since gold nuggets are highly variable.

[Altra]

I was more interested in a table of standard targets so that we can compare the performance to commercial detectors since gold nuggets are highly variable.

I been using the 5mm and 10mm squares cut from a aluminum beer can. But aluminum thickness and grade will vary from can to can. I had an idea for standardize targets. We could make them using a pcb. Draw up board with multiple geometries and sizes. These could then be cut into separate targets and glued to wands. This would create a standard gerber file that everyone could use. Assuming say 1 oz. copper, people around the world could test and compare results

[pito]

Please explain, because I do not understand what you want. = example showing how coils are connected to PCB, here is missing shielding .



and how it is done