Really sensitive pinpointer for grain size objects

[Detectorist#1]

Hi Pito,
The coil's module in video seems very flat. In LVDT principle have to be used 3 coils in linear configuration. Where is the true? Or I not understand the video?

[Detectorist#1]

Hi Pito,
4mm steel ball is very strong target for any metal detector.
Send me please video with pinpointer configuration with 1mm steel ball as target. Thank you in advance!

[pito]

I just balanced 2 coils small black one Rx and under it is Tx coil, I didn't build LVDT, I used what I have to check the principle,

Tx coil is driven by NE555 generator , series resonance,


1mm steel ball = For now I don't have time to build a pinpointer, busy with 2 box machine.

[Carl-NC]

Hi Carl,
First, thank you for suggestions.
My experience in metal detecting is very modest but I have some question to you:
Increasing of the value of the safety resistor in the input of scheme (50 times to 10K) will increase seriously the current noise. I think this create problem. To fight with this problem, you need to increase the repetition rate to 2500pps.
What is the summary profit?
Also the peek current via input resistor with 200 oHm is 50 times bigger , but discharge of current carriers in diode via 200 oHm is also 50 times faster, or not?
What is "direct sampling" - this is new idea for me.

The peak coil current is proportional to 1/L, or 1/N². The peak TX field strength is determined by the ampere-turns, so it is proportional to 1/N. The induced RX signal is proportional to area-turns, or N, so the round-trip sensitivity versus coil inductance is pretty much a wash. However, a lower inductance coil allows earlier sampling (which helps) and higher inductance coils reduce power and allow a higher rep rate (which also helps). In designing the F-Pulse I got better results with the higher inductance, but only when the clamping resistor is increased. Yes, it seems that the clamping resistor should also be a wash but I almost always find that increasing it improves early sampling, and the noise increase is hardly noticeable.

Hi Carl,
Send me please video with distance performance of F-pulse with different sub grain targets in air. This will be very useful for me. Thank you in advance!

Wish I could, this is work I did several years ago and the prototype was stolen. I was at the point of adding a GB channel which required a new PCB. I've been wanting to get back to this but no time.

[Carl-NC]

I just balanced 2 coils small black one Rx and under it is Tx coil, I didn't build LVDT, I used what I have to check the principle,

Tx coil is driven by NE555 generator , series resonance

When I designed the White's TRX I started with this approach. However, placing the RX coils on the ferrite actually reduces sensitivity. It's harder for a target field to change the field that cuts through the RX coil. Instead, I wound the TX coil on a ferrite rod and wound the RX coil with an air core. Instead of a rear RX coil for IB, I instead reverse-wound the RX wire on top of the TX coil to achieve IB. The ferrite core results in fewer counter-RX turns needed for IB. Here is an illustration of the design:

[Detectorist#1]

Hi Carl,

Thank you for this explanations. I will test variant with high inductance coil and high value of safety serial resistor.
The construction of coil's unit of TRX pointer is very interesting. I also noticed that using of ferrite needs longer delays.

[Detectorist#1]

Hi Carl,
I know that TRX uses 12KHz IB principle. For the target's field is more hard to change something parameters of the coil with ferrite core.
It is interesting - the same happened with pulse detectors.

[Carl-NC]

I also noticed that using of ferrite needs longer delays.

It depends on the ferrite. Some are really bad, some work well.

[Detectorist#1]

It is good - we have wide area for experimenting!

[Detectorist#1]

Now I remember - at my first tests I used ready 22uH-47uH-100uH inductors with ferrite cores and I was surprised that parts with the same induction needs two times different delay for using.
Only different part was the type of ferrite.

[ivconic]

I don't know what motivates (or prevents) you not to draw a schematic clearly and legibly?
(not only you, I'm addressing all those (and there are a lot of them, especially Bulgarians, btw are you Bulgarian too?) who have the same bad habits, to draw the ugliest and most unclear schematics)
LM337 input is not clear to me btw...
Remark; very bad results for such a "complicated" schematic? A video demonstration of the work would say a lot.
Is it so difficult to draw the schematic a little more clearly, for example like this:

[ivconic]

"...recovering of one grain pieces from one inch distance without problem in air and in ground and salt water..."
"...my pinpointer have possibilities to recognize one grain object from one inch without problem..."

"...Remark; very bad results for such a "complicated" schematic? A video demonstration of the work would say a lot..."

I stand corrected!
I wrote this hastily because I'm under the impression of being in a bad mood after accidentally watching something very evil on TV.
Your results are actually very good!
Congratulations!
Proof that I am honest is my effort to draw a schematic clearer and more legible.
As I said, the only thing that is not entirely clear to me is the voltage input to the LM337.
All this motivated me (which is very difficult lately) to dig my pinpointer out of the drawer and test it.
For the first time!
I've had it for over 2 years, I've used it on the sites and I'm very happy with it.
But I've never looked for anything smaller than standard coins.
This video will show more than words. That's why your result with your pinpointer is more worthy of attention.
If we clear up some doubts about the schematic to the end; it will be a very interesting DIY project in these dull winter days.
Finally, I can trace your pcb design and see for myself how what is connected.
But it would be easier if you provided a sketch of the pcb layout with components markings and labels.

P.S.
Carl;
the proof of my superficiality and laziness is that until half an hour ago I did not know that F-Pusle has a "timeout" function!
Because I hated reading the manual.
I never read any instructions!
So now after 2 years I found out that F-Pulse has a "timeout"!!!

[Tinkerer]

How is it possible to achieve 3us with sich a high inductance/capacitance coil? I'm using 170uH but after attaching the cable and the coil shield I'm sampling at 5us, on top of the transient with a preamp gain of just 14.

Teleno
To achieve a delay of 3us to the first sample on a PI can be very challenging. Many factors have to be taken into consideration and many compromises have to be made.
As I am working on this problem at present, I would much appreciate if you would be so kind and help me finding, analyzing the problems, circuits and solutions we could try.

I am starting a new thread for this purpose: THE ELUSIVE 1us TC TARGET

[pito]

@Carl-NC
In your case strength of the field reaching the target is smaller, but air core Rx may compensate that ???

[Detectorist#1]

Hi ivconic,
We are neighbors - yes I'm Bulgarian. And you are real Serbian - with strong feelings!
First, thank you for good words for this project. Inside have not nothing new as schematic. I just picked up very carefully every component.
Your drawing of schematic is more readable. In my variant, I put many additional information for make more easy the repetition from DIY people.
It is important that the resistors around the coil have to be as 3pcs in series - the peek voltage is appr. 1000V. Variant with one high voltage resistor is inconvenient - too expensive or too big size of the resistors.
In your schematic you dropped module for battery low indication (yes it's optional).
LED diode in the input of LM337LZ is for "Power on" indication only.
Now, I prepare final versions of detailed BILL OF MATERIALS and sketch with PCB layout and component's placings.