That's me. Of course!

Hi all,

I have made the modification to the latest mono coil VLF bread board. It is working very nice.

DSP path:
Differential input: Signal = right-channel - left-channel (or vice versa)
Demodulation of Signal at frequency f1 -> I/Q vector 1
Demodulation...

​...

​...

What I think is there's no need for backhand insults ... in engineering there's risk analysis and there are opinions some people differ so be it.

You are absolutely right, and your point is completely understandable. However, the core design of this application was built to maintain maximum compatibility with previous software standards and existing user habits. There are a few technical reasons behind this specific architecture:

...

Dear contact me with email

To be frank, if we want to turn this into a real collaborative effort, it requires setting up a dedicated repository where the MXT firmware can serve as a sensible reference for analysis and learning. But how do you actually envision that happening in this environment?

How much useless...

We have a good IB-VLF base now.

Now back again to pure mono coil VLF:
- It must have high Q.
- Less power drain. High power efficiency.
- Must work with standard line output level (1 V rms) and with every sound card this time.
- KISS & cheap design.
- All...

Hello KRinAZ​
Thanks for the HEX file.
I've been following the topic since the beginning.
This is the first time I've seen this file.

Honestly, having fully reconstructed and analyzed the firmware, I can clearly see the limitations imposed by the old hardware platform. There is really no point in trying to reuse everything exactly as it is.

Modern microcontrollers with hardware DSP (like the STM32 G or H series) offer...

Hi ivconic,
Don't be so quick to thank me in advance for teaching! We haven't agreed on my fee yet. It's important to use our sense of humor!

arabic is the mova of modern Great Britain?

.pulse induction metal detector sچه کسی میتواند در ارتباط با برد پی سی پی اشکار سascho bulgariaاز بلغاری کمک کند

کسی شماتیک و پی سی پی مدار اشکار سازبلغاری را دارد...

PS:

On very mild ground mineralsation:
It can be operated at single frequency in differential mode.

On heavy to severe ground mineralsation:
Two frequency mode (f1, f2) -> 4 I/Q vector processing required.

Hi all,

this is my latest KISS Dual-Frequency Induction Balance VLF Detector Controller for sound card use.
Operated at moderate power consumption in high-gain mode of G6 (mid-level output, TX coil voltage up to 100 Vp).
The Q of the LC tank is still high enough (approx. 10)....

I can't be sure, the link is missing from my "order" list, because it was last year.
It looks like Aliexpress doesn't keep those listings longer.
I did not save the seller's name.​

Can I have a link to a legitimate seller? Thanks!...

By the way, this is the original HEX file, don't remember if it was uploaded elsewhere in these posts......

Well there has recently been an amazing series of developments on this project!

Regarding the copyright discussions - and just to make the point - I didn't intend this project to copy the original MXT firmware or hardware verbatim - but to understand it well enough to be the springboard...

I agree, from the info this thread has produced I would be 100% confident in repairing a MXT or a GMT. Wish I still had that GMT that died, I could now resurrect it...sigh...I got rid of it over 10 years ago...sigh......

​...

It's true, Mxt is an outstanding detector, I'm following this thread with bated breath, unfortunately Garrett buried the best projects in the depths,White, and apart from 24k they did nothing, it's a great pity, unfortunately, Vortex is unfortunately, in terms of its structure and programming, like...

Last year I ordered 30 pieces of J113 and they all arrived ... FAKE!
And then I ordered 20 pieces from another seller and they all arrived... OK!
You never know what you're going to win, like the lottery!​

How much?

Now here is a python harness that exercises the firmware and you can see if its working or not.
But its real power is that the AI can "see" this interface and debug the software in a loop without me having to flash and check it manually.

Files attached below openmd cref...

There has been enough posted on here to fix your detector ... ( files / schems / etc ). I think all the whites patents have expired but copyright has not.
The only real concern would be if someone started manufacturing MXT clones with the original or derivative firmware ... then Garrett may complain...

I've also bought ICs from Aliexpress that don't work. Nothing can be done, they're cheap and people get scammed, there have been some that work, you can't be sure until you test them....

These changes would not infringe the patent rights of the manufacturer.


...

Here is the semantic C verision of the Clean Room firmware - derived from the clean room assembly....

In advised range... 0.38mH/1.7R

I forgot to mention:

The feed-in can be also made with rectangular pulses (PWM output of MCU). Take a simple push-pull driver (a two channel mosfet driver) with external power supply and you are on game.
The mosfet driver delivers enough power for this purpose.
This project...

Hi all,

oh well, I had expected a higher magnetic field strength at the inner corner regions. This is not as much as I have really expected.
So the rotation of 27 degree angle of the coil can be totally neglected.
But the 90 degree rotation makes still sense for the larger search...

Hi all,

ground loop systems are really not much practical to use. Especially on virgin ground.
But a larger portable frame coil is the next best option.

------
Back to my previous dual frequency VLF:

TX coil (either mono or IB):
Thick litz coil...

ivconic can you give your coil details?
​​
thank you

Aliexpress is death. I from Turkey and i used aliexpress 10 from years. AliExpress is banned in our country. Now I can't even log into my old account.

Hi all,

Quick progress update - I'm officially starting on the software side of the Spectral-G4 (my multi-frequency VLF project with an AI "spectral fingerprint" discriminator). Hardware decisions are locked, so now it's time to write code.

--- The toolchain ---...

Private thread sounds good. Maybe split the firmware research/discussion from the general project.

I am extremely grateful to you for being willing to teach me, thank you very much!
Could you do me a huge favor and explain how to use an oscilloscope?
Little is clear to me from what you are talking about!?
Thanks in advance!​

​

Hi ivconic,
For you, experienced guy in metal detectors, is very strange to write for the problems with RX amplifier for this very simple detector. It is easy to see with oscilloscope if the signals on the output the RX amplifier stage haves some problems. In this project, the author make attempt...

​...

Here is what I understand... It is legal to reverse-engineer code for the purpose of understanding how the code works. First Texas did this with the Quest detector to determine that they had exactly copied the T2 code. But it is probably not legal to then use that code to develop a derivative product,...

Agreed, a round coil is the most efficient structure at generating a magnetic field. A square coil with the same size and N will always have lower inductance.

It also happens on the RX side: a round coil is the most efficient structure at coupling to the target's eddy field. An equivalent...

Maybe a new topic?

All these "stars" and idiotic reviews are obviously fake too!

​...

Very dissatisfied I wrote an official complaint to Aliexpress.
After a few days, I received a reply that "I am not right and that my complaint is rejected"... !??
So much for how Aliexpress protects its customers.
A gang of thieves, liars and robbers! Generally speaking....

I had the pcb, I had most of the materials.
It's boring so I amused myself by making another one.
In principle, it works immediately.
But this time this one doesn't work well enough.
Because I don't have the original LF357.
And I don't trust buying it from local sellers....

The only problem will be the free-running LC-oscillator (phase/frequency instability may be too much).
But this can be replaced with a programable frequency oscillator (crystal source, adjustable to ground mineralization, via MCU).
With blutooth-link.

Hi Paul,



No problem.
It will be a fully different technology. It's a VLF not pulse induction. Nor synchronized to the transmitter.

A very simple one:
High power, single frequency, stable and free running LC-Oscillator with a big ground loop (10 - 20...

Its been done and patented Aziz .. woody did one some time ago....

I think the point of the recent discussion is missed. Why do Clean Room Builds exist.

I have done a clean room build. I have expressed an honest opinion. I have released the files.

nuff said .... its a public forum - expect robust discussion.

Extremely useful topic, especially for people like me. I, as an old and stupid person, very often do not even know what to ask. In this topic I found answers to many questions that I would never have thought of asking.
Therefore, I thank KRinAZ for starting this topic.
I also thank Carl-NC,...

side view cross-section...

I am not convinced that the mere analysis of firmware, building a RAM map, or the semantic reconstruction of a program's behavior automatically constitutes a derivative work. If that were the case, a significant portion of reverse engineering, security research, documenting legacy systems, or implementing...

Thanks for reviewing the code ... If the arch is different then the clean room has done its job by not creating a "derivative work".

The fact that you are building the RAM map will be a red flag in regard to derivative work. I have no idea if Garret would chase you or anyone...

I reviewed your code. What you have is a completely different program architecture.

It looks like you generated an alternative assembly firmware.. That can be a valid clean-room reimplementation, but it is not the same as recovering the original firmware.

Your code does not...

Hey Moodz,
Amazing performance.
You are an extremely smart guy, hats off to you.​

two more...

Hi all,

now the comparison of magnetic field cross-sections at different z below the coils.
I have reduced the number of palette colors to have discretised magnetic field strenght regions.
And I had to change the scaling range.

File names describe the coil name...

Hi all,

lets think of big, bigger and max!

I wonder, whether it is possible to operate the High-Q VLF detector as a Ground Loop VLF detector.
5 m or more diameter coil layed on the (severe mineralized) ground. The operating frequency can be set to the ground conditions...

Hi all,

this is the target response comparison of the 18 inch round vs square mono coil (see zipped Excel file in the attachment).
I am still using my 20 mm diameter standard target.

My coil bundle is 8 x 8 mm in this example. It is very compact. This is the worst you...

This is a fundamental issue. It fits fully to the discussion.
The AI gives you a lot of food for thinking and asking the right questions. You decide the path to go. AI helps in finding the path you want to go.

Believe me. AI is the best tool ever.....

AI doesn’t replace thinking - and I honestly have no idea what your comments add to this discussion....

And my strong advise to the people out there:
Learn how to use the AI for yourself.
Learn how to use the AI for yourself.
Learn how to use the AI for yourself.

It is a peanut, if it is costing some money. But there are ways to use it for free too....

Good idea .... of course the danger of clean room implementations is that there is a certain "personality" in the detector that can be lost with a functional replication ( but not original code )....

You can approach it the way you described and even recreate the detector’s behavior from the manual, schematic, and engineering report. However, my original goal was to create semantically reconstructed code while also testing how well AI could handle such a task.

As for the ownership...

This is a good underestimation.

It is much much more. ...

And can you imagine, that 1 TB of data (it is a small hard disc), which contains 2 trillion parameter neural network (4-bit floating point number) contains the whole world knowledge and the processing scheme for the knowledge?
We are heading to the multi-trillion parameter networks. Faster and...

... bigger than the invention of the wheel IMHO ...

Hi all,

some words about AI and the so called "hype".

There is no hype in AI at all. Absolutely no. It is real.
It is a damn fnk real thing.

There won't be real investments in billions/trillions dollars if this would be a "hype".
...

Actually when I used to work ( LOL ) .... one of my engineering tasks was protecting intellectual property. If Garrett do own the whites IP rights then blatently putting up copies / reverse engineering the firmware is a violation of copyright and could result in a takedown notice to this forum at a...

[email protected]

Full code with Android app available...

The real question is: what exactly did the AI accomplish here? xDD

Because in my experience, the reality looked much less “magical” than the hype posts suggest.

Claude Code + Opus 4.7 could spend hours analyzing the firmware and still not fully understand everything. The...

​
​


​​​...

Hi Sergey,



but I can do it. There are more ways to do it. Note, that I have 2 / 4 I/Q vectors for processing.
A robust mono/IB-coil VLF design in severe ground conditions is possible.
Even with an extremely High-Q design.
...

...

this is what the AI said ...

​...

The gardener is always the murderer. In this case to the secrets of the source/binary code.


BTW, regarding the averaging of the phases in the code.
Never average the phase angles. Do it in the I/Q part and average the I and Q parts. Than calculate the phase, when...

BTW,

regarding the rotation angle of the square coil:

Rotation angle of 0 degree:
A rotation angle of 0 degree will do it either. The coil mounting is simple. The small shallow target is beeing hit twice the time when swinging the coil at in front and rear position. When...

This AI is a sly one.
Interestingly, there were no compilation inconsistencies due to the unspecified "paging" for the programs in the second page, called by the main program -
call dsp_filter
call dsp_ground
call dsp_metrics ...
But since everything works, there...

Done .. complete clean room build in pin for pin chip ( 16F876A ) for the MXT schematic. They still make the 16F876 in a DIP chip so its good for all you oldtimers.
Math is 32bit and CORDIC. 100% assembly code.
24 bit ADC support ( would need to be an SDI chip .. and some small additional...

Hi Tinkerer,



the coil software does take it into account.

We have three regions to look at it:
- Near-field zone (z distance < coil radius R):
We have an intense magnetic field strength inside and nearby the 4 corners of the square coil compared...

The AI did all this whilst I was gardening ... its prepping the docs now.





​...

it lives ! ( on the simulator )

this means the code compiles and runs ( on a 3rd party simulator not mine ).

=== adc_sample=40000 : printable runs (>=6 chars) in SPI stream ===
.......OpenMD 1.0 BATT 133 .................................................. .....................................................

... here is the main asm file ... the AI is doing the simulation and debugging now.

;================================================= =============================
; openmd.asm -- OpenMD : clean-room VLF metal-detector firmware
; PIC16F876A, 8 MHz HS. (c) my own original authorship....

I'm really impressed.
I can rightfully call you "The Lord of Artificial Intelligence."

.. about 15 minutes in ... I am outside mowing the lawn

​...

Well thanks...

Instead of porting the code ... I have instructed the AI to build a "clean room" version for the 16F876A chip. It will support 16 or 24 bit ADC and uses 32 bit math not 24 bit.

Its working on it now .. estimate 30 minutes to complete.

...

I completely agree with you. I was also worried about whether such reengineering was ethical and legal, indeed, Karl could give a competent answer. .

I also completely agree that LLM is an extremely effective tool (with proper structuring of tasks, questions, data... given to it).
...

you are right ... but it also comes down to understanding how the AI works and how we can use that tool ( the AI ) to do alot ( all ? ) of the heavy lifting for us.
Broadly speaking there is a new technique in programming using AI that I will call "CONTEXT ENGINEERING" ... this has evolved...

<High-Q LC-tank (of approx. 30 in my case).>
A change in the ground signal level causes a phase shift in the resonant circuit due to the change in its parameters, not due to the induced signal from the ground/targets. This phase noise will interfere with target detection and cannot be compensated...

FOUND ONE!

Hi moodz,
I still think that AI wouldn't do that much useful work. What AI wrote to you is true and it generates that, based on statistical data, close enough to the code given to as a model, and not because AI understands the code and processes. My opinion remains the same - if you want to have...

Have you taken into account the uniformity of the coil field?
With the round coil the field(lines) is the most uniform and gives the biggest depth.
With the square coil, the field is less uniform. In the 4 corners the field it "pinched". The "pinching" produces more...

Hi all,

I've realized that Google's AI has been telling me a lot of nonsense.
We cannot do without the coil software.
The coil models have been made.

The results are very interesting..
I need to find a way how to visualize the results ..
The square...

You need to rotate the coil by 27 degree now.


I am preparing a new coil comparison between a 18 inch round and square coil....

Yes, it was....

1970, and I own one:

​...