Sable VLF Detector with VDI - Inspired by the Raptor project

[Olly]

Hi Guys,

After experimenting quite a bit with PI detectors it was only natural that at some point I'd want to take a closer look at Induction Balance designs.

It seems that the coil is much more critical and less forgiving with IB and so to give myself a platform for experimenting with various coil designs I came up with a design which takes inspiration from the 'Raptor' project but uses a Raspberry PI Pico for timing and VDI display which fits on a small 90mm x 60mm PCB.

I've got the first one up and running in the lab with very promising results. Detection distance is pretty good and the VDI seems to work well even though the code is still pretty rudimentary and I think there is much room for improvent with my coil which is my first attempt at building a double-D.

I've called this project 'Sable' and here is the schematic and a few pictures. If anyone is interested I can post the gerbers and Pico code such as it is.

[ATTACH]n414268[/ATTACH]



Regards,
Olly

[ivconic]

Looking nice.
LT1054 choice surprised me a bit.
It is old, tough to obtain and not cheap.
But that would be the easisest part to modify.
I am not familiar with Pico programming, though I do have one RaspB. 3+ for long time, forgotten in some drawer.
But I see there is Pico on local market and the price is affordable.
Any shorter video demonstration?
Can you be more precise about TFT model, specs etc.
I can't locate it in local ads.

[Olly]

Looking nice.
LT1054 choice surprised me a bit.
It is old, tough to obtain and not cheap.
But that would be the easisest part to modify.
I am not familiar with Pico programming, though I do have one RaspB. 3+ for long time, forgotten in some drawer.
But I see there is Pico on local market and the price is affordable.
Any shorter video demonstration?

The Pico can be programmed in Python or 'C', I use 'C' as it tends to be much faster.
I used the LT1054 as I have a few of these lying around. As its input is only 9V then the much cheaper ICL7660 or TL7660 may be used instead.
Here in the UK I get the Pico from the 'PI Hut' and it's around £4. They also have the 1.8" LCD display for around £10.

Here is a link to the LCD
https://thepihut.com/products/1-8-lc...pberry-pi-pico

I'll try get a short demo video up and running at some time.

[Olly]

Here is a short video showing the VDI in action for a number of coins as well as a pull-tab.
The bar graph on the right shows signal strength and as can be seen the VDI is relatively stable across a wide signal strength range.
Still very much work in progress with lots of tweaks to come and at the moment there's no -ve indication for ferrous targets yet.

https://youtu.be/7MiHW2Rm1mc

Regards,
Olly

[ivconic]

Ok, I don't have such a display, but there are standard ones of 1.8", with further "investigation" on the net, I will find a way to connect it.
Pico is around 4-5e here too. The video is ok, it seems promising.

[Qiaozhi]

I've called this project 'Sable' and here is the schematic and a few pictures. If anyone is interested I can post the gerbers and Pico code such as it is.

Regards,
Olly

Looks very good. I would be interested to see the RPi Pico code.
I bought one when they first came out and played around with Python, but haven't used it in a project yet.

[Olly]

Looks very good. I would be interested to see the RPi Pico code.
I bought one when they first came out and played around with Python, but haven't used it in a project yet.

Here is the code as shown on the video, please don't judge, I know it's very quick and dirty at the moment and certainly nowhere near as elegant as I'd like yet.
Definitely Work-in-Progress.

Sable_Pico_Code.zip

Operation is relatively simple: -
Double-Clicking the encoder pushbutton illuminates the top LED and then allows R sample phase adjustment via the encoder.
The next click of the button illuminates the next LED and allows X sample phase adjustment.
Adjustment of the Sensitivity level is after the next click.
The final click stores changes to the flash and returns to normal operation.

I plan to have these values displayed on the LCD while adjusting, also to have preset discrimination and sensitivity selection via the LCD. More of the work-in-progress stuff...

All of the timing generation is done without involvement of the main processors either via programmable PIO or with the built-in PWM controllers.

Regards

[Qiaozhi]

Here is the code as shown on the video, please don't judge ...

All of the timing generation is done without involvement of the main processors either via programmable PIO or with the built-in PWM controllers.

I'm not going to judge you on the code. My experience with the RPi Pico is minimal, and I've never experimented with the programmable PIO.

Where did you find details on programming the Pico in C? There a few books on Amazon, but they're quite expensive if you want a "real" book.

[Olly]

Where did you find details on programming the Pico in C? There a few books on Amazon, but they're quite expensive if you want a "real" book.

I used their official documentation which is pretty good.
https://www.raspberrypi.com/document...y-pi-pico.html

This is a pretty good getting started guide for just about all operating systems.
https://datasheets.raspberrypi.com/p...-with-pico.pdf

And the full C SDK documentation is here.
There are also examples of using the programmable PIO
https://datasheets.raspberrypi.com/p...pico-c-sdk.pdf

More detail on the internals and PIO architecture are available in the datasheet for the processor.
Chapter 3 has PIO details.
https://datasheets.raspberrypi.com/r...-datasheet.pdf

An area I use a lot is their online documentaion of the SDK and APIs here.
https://www.raspberrypi.com/documentation/pico-sdk/

Hope this helps,
Graham

[Olly]

Ok, I don't have such a display, but there are standard ones of 1.8", with further "investigation" on the net, I will find a way to connect it.
Pico is around 4-5e here too. The video is ok, it seems promising.

I think the interface is pretty common across a number of displays. I was thinking about trying something larger but the 1.8" is a decent size and fits in with my enclosure design nicely.

Here is the link to the Waveshare info on the LCD. I'm sure thay have others which will work but not necessarily plug directly onto the back of the Pico.
https://www.waveshare.com/wiki/Pico-LCD-1.8

Regards,
Graham

[Olly]

Here are the gerbers, I used JLCPCB to have them made.
sable_gerbers.zip

Due to the compact nature of the board some of the references on the silkscreen are not as close to their components as I'd like, so it's advisable to use the placement guides when assembling.
Front_Fab.pdf
Bottom_Fab.pdf

Regards

[Qiaozhi]

Thanks Olly.
Lots of information there to digest.

[Olly]

Busy printing an experimental 8.5" concentric shell at the moment.
Leaving a slot for the bucking coil above the RX coil so I can play with nulling/phasing etc. after I've potted the TX and RX coils.

[felixd]

Hey Olly.

Just found your project. I have just ordered PCBs.

Questions:

• Why are You not using RPi Pico as frequency generator?
• Coil recommendations?
• Since all data are available here (Schematics, Gerbers, Software) - Would You move this project to GitHub?

[Olly]

Hi felixd,

I'm using a Colpitts Oscillator as this results in greater efficiency and much lower distortion than force driving the TX via the Pico.
This does constrain us to a single frequency, but I wasn't looking to create a multi-frequency machine, rather I was just trying to further my understanding of a how a single frequency IB machine works.

The choice of coil is pretty varied and very much up to you and what frequency you want to operate the machine at. The default values used in the schematic were for a DD coil I made which has a 3mH - 3 Ohm TX coil and a 10mH RX coil. With these values the machine runs at around 10kHz as in the Raptor project.

I've also tried the detector with a TGSL type DD coil (6mH TX, 6.5mH RX) and adjusted the TX and RX caps in order to run at around 14kHz and this also works pretty well although not as deep as when running at 10kHz on my original test coil.

I find that it's quite tricky to wind and balance these coils so I've bought a coil from Aliexpress (Link below) to try.
This coil is intended for a higher frequency, but I'm going to adjust the TX and RX caps and try it out at various frequencies (8kHz, 10kHz, 14kHz and 19.2kHz which is what it's actually designed for)

https://www.aliexpress.com/item/1005...d=9vS4FaqFIbEA

I have no real intention of placing this on GitHub. In fact, I'm very happy to post the project here as this is the source of all the expertise and inspiration for the project in the first place.

I wish you every success with your build.

Regards,
Olly