idea for ultimate VLF detector

[reptooyep]

Hi,
I've bee thinking since few month of a vlf detector working with 2 MCUs (ATMEGA328P for example).
The first one would generate the transmit signal (PWM), acquisition (12 bits SPI ADC) and generate interrupts.
The second one would takes care of the signal processing and VDI display when an interrupt is received.
The idea comes from the very low price of ATMEGA and its memory limitation.
Before starting the project, i want to collect your ideas about that.
I like the KROT2 TX circuitry

[kt315]

xlt on steroids. good plan although...

[reptooyep]

Hi,
I started yesterday making the layout.
The schematic is in my head, i'll edit it soon.
The two ATMEGA328p will be linked together by serial link.
The TX/RX MCU uses tho PWM lines, the first one is for TX generating and the second one drives a transistor and a RC cell wich will provide a negative DC supply for the VCA gain adjustment. The more signal width = the more negative signal = the more gain.
I'm planning to use a SPI 12 bits ADC or maybe, the 8 bits internal ADC will be ok ... i don't know. The KROT project use internal ADC so, it should work like this. I'd like to have advices for this please.
The second MCU will drive a 160X128 pixel LCD with touch screen... sounds good.
The 1st MCU out the clock on CLK0 for the second one set to external clock.

The all project will be FREE and the sources code will be available for everyone.

[kt315]

maybe do you want to continue QED project? only question.

[Skippy]

For info: The Teknetics T2 uses 19-bit ADC. Sure, with careful design you can reduce the number of bits needed, but if you're trying to make an 'Ultimate' VLF, then I think you're going to need a 16-bit ADC. Lower bit-count converters will be OK for the development of the idea, though.

[reptooyep]

The higher bit resolution also means the lower speed ?
Exept if using a 50$ adc, which have non sense to use with chip MCU.
the thread name is not good, ultimate is too much but using two MCU sounds good to my ears.

KT35, i have to check the project you told me to look at...

[kt315]

Rep, i check your email address.

[reptooyep]

Thanks for info,
Two MCUs PI detector will be my second challenge.

[Funfinder]

The problem with all those microcontrollers is:
They are susceptible for electro-smog!

And even if you use this stuff just for a little bit of "data-play" (signal-calculations etc)
the ultimate VLF detector would be one which is completely immune against e-smog!

And it should be immune against ground-mineralization, too.
Which means that it perfectly measures the GEB ultrafast and subtracts it in realtime from
the metal-targets.

Perphaps some uC's should be integrated already directly into the coil!
The more closer to the raw signals, the better.

[Skippy]

Good converters aren't expensive nowadays, and you don't need high speed either, 20 ksps ( 10 ksps, 2 channels ) should be enough.
Here's a decent 16-bit one, the AD7687:
http://www.analog.com/en/products/an...rs/ad7687.html
http://uk.farnell.com/analog-devices...ter/dp/2399976
http://uk.farnell.com/analog-devices...-me-pd-mi-acce

[kt315]

The first one would generate the transmit signal (PWM), acquisition (12 bits SPI ADC) and generate interrupts.

if you do http://www.geotech1.com/forums/showt...th-larger-coil! then this part has to be in a coil. then bluetooth transmit signal in main block.
if you plan to make a simple coil depth will be low.

[reptooyep]

Never any DIY MD would have better performance than a DEUS or other. That's my opinion. But maybe we can make something
Better than a Garett for few dollars ? That's the challenge.

[kt315]

garret ace-250 uses very simple concept. all does analog part, uchip used only for LCD.

[F117]

When you plan to connect 2 microcontrollers via a serial link, consider using bluetooth.
Serial to bluetooth is cheap.
Instead of receiving the signal on a second microcontroller, you could also receive it on a cheap tablet with a more elaborate GUI.

https://www.aliexpress.com/item/HC-06-Bluetooth-serial-pass-through-module-wireless-serial-communication-from-machine-Wireless-HC06-Bluetooth-Module/1850590704.html?spm=2114.01020208.3.10.EuPuZo&s=p& ws_ab_test=searchweb0_0,searchweb201602_5,searchwe b201603_3&btsid=c28dba74-1aeb-4643-882e-2d9a68bceb7a

[reptooyep]

I have 10 BT boards waiting for a project.
I'm thinking of a coil and simple electronic linked to my android phone by BT.
I'm afraid of BT signal perturbation in coil nulling. I've been told that the neerest of the coil is the best for electronic. Maybe a good shielding would get rid of that.
The biggest problem for me is to find free time for learning a new ANDROID SDK. I know Atmel Studio pretty well so, it's just to save time.
I tryed AppInventor but i'm not sure you can create such an App ? I have to try again.