VLF MD with digital signal processing : Bee-Buzz 1

[boilcoil]

Ha, ha, the cost of the money transfer will be greater than the amount.
But let's not spam Atul's topic.

[Marchel]

Now I see that the author of the mole metal detector also used the STM32H7 in the new version and writes that the detector achieves better results.

http://www.radiodom.org/index/krot_stm/0-32

[Tinkerer]

Carl continues to "annoy my vanity" with constant RTOS "slapping"!
It's as if he knows that I'm weak here and that I haven't done it before!
I'm going to have to get mad, sit down and learn it once and for all!
But if I've already decided to start with Nucleo 144... then I don't need it too much!

https://www.freertos.org/Documentation/00-Overview

[Atul Asthana]

This represents about +/- 3.5% which isn't enough to make a difference in the target responses. MF detectors run a minimum of a 3x frequency ratio and preferably 5x. For 6kHz, you would want the second frequency to be 18kHz.

It is possible to switch a resonated coil between 2 frequencies but it is a slow process. It takes some time for the coil oscillation to settle after switching frequencies. This is why all MF detectors use linear ramps instead of sinusoidal oscillations.

I searched and found some oblique references to using first derivatives of smplitude and phase with frequency (slope) [d(amplitude)/d(frequency) and d(phase)/d(frequency) ] for estimating target depth, size and material identification, though it requires creating empirical formula to actually utilise this info. but none the less, it improves on the anslysis, though tecommendations are to use more frequencies.

I was wondering if there are any patents on this, that I can refer to.

[Atul Asthana]

You can make fast samples with slow ADC, but you need to store the signals in buffer capacitors. Of course, the number of samples is limited, but the sampling speed is as fast as you want (it depends only on the time constant of the loading circuits).

Something like this:
[ATTACH]n432404[/ATTACH]

Of course, this is a PI metal detector, but it illustrates the principle.

PS: and costs $15

sample and hold !
I recall using these with slow 8 bit adcs, some 35 years ago, using lf356 (as far as I can recall).

[Marchel]

Today I had time and tried the method that I use in my metal detector on an STM32 BluePill and it seems that direct sampling would work without problems. In my test I tested a TX frequency of 20kHz and I made 16 samples which were then resampled 64x. The main loop runs on RTOS and includes all filters, magnitude and VDI calculation so this is the majority of the code. I attach a printscreen where you can see samples from the TX signal processed in Demodulation(); the frequency of the RTOS loop is about 312Hz.

[Atul Asthana]

Today I had time and tried the method that I use in my metal detector on an STM32 BluePill and it seems that direct sampling would work without problems. In my test I tested a TX frequency of 20kHz and I made 16 samples which were then resampled 64x. The main loop runs on RTOS and includes all filters, magnitude and VDI calculation so this is the majority of the code. I attach a printscreen where you can see samples from the TX signal processed in Demodulation(); the frequency of the RTOS loop is about 312Hz.

excellent excellent

thats a great confirmation.

I recalculated the entire md for 9 khz, improving the Qs and tightening the bandwidth + increasing the siot rates from 256 to 512. my calculated enob is 19, but practicslly, it should be slightly better than 16-17 bits.

[Marchel]

excellent excellent

thats a great confirmation.

I recalculated the entire md for 9 khz, improving the Qs and tightening the bandwidth + increasing the siot rates from 256 to 512. my calculated enob is 19, but practicslly, it should be slightly better than 16-17 bits.

If you are interested, I can add complete files for Keil ARM and STM32 Cube MX.

[Atul Asthana]

If you are interested, I can add complete files for Keil ARM and STM32 Cube MX.

I will be grateful,

thanks

[Marchel]

I will be grateful,

thanks

Here is the complete project for Keil ARM and the file for displaying data in the MCUViewer program, which is in the Media folder.

https://www.mediafire.com/file/qbp8x...ector.zip/file

https://github.com/klonyyy/MCUViewer

[ivconic]

Marchel I sent you a pm.
But as always I have problems with the pm system on the forum.
I have no feedback that my pm was sent to you.
Please check if you got a pm from me.
Otherwise, if you want p2p contact; contact me on email: [email protected]
I apologize for breaking into the topic like this and disturbing your work!

[Marchel]

Marchel I sent you a pm.
But as always I have problems with the pm system on the forum.
I have no feedback that my pm was sent to you.
Please check if you got a pm from me.
Otherwise, if you want p2p contact; contact me on email: [email protected]
I apologize for breaking into the topic like this and disturbing your work!

Yes, I received a message from you and I replied to you.

[ivconic]

However, now I only see your first pm from before!?
That's what I'm talking about, those problems with the pm system on the forum.
Here I tried everything and failed to see your second pm?

[Marchel]

However, now I only see your first pm from before!?
That's what I'm talking about, those problems with the pm system on the forum.
Here I tried everything and failed to see your second pm?

I wrote you the last message about 05:05 PM

[ivconic]

Oh I finally figured out what the problem is with pm!
If I click on "Inbox" your message is not there, it simply does not exist.
But if I click on the above "Message Center" then your second message appears!??
Does this defy logic?