Competition!

[Tinkerer]

With the KW detector detector we have a very simple PI for beginners to experiment with.

For a higher level experience, I propose the attached project.
I could not upload the whole document, as it is about 10Mbites, so I have printed out the first 6 pages.
If there is no way to upload the whole document in one go, I can split it into chapters that are small enough to upload.

Very little has been done to look at the target return signals of PI detectors during the ON time. This document could be a primer for that.

I spent myself a considerable amount of time investigating ON time signals. I would be happy to share what I know about it.

[Tinkerer]

Hi Thinkerer,

thanks for the pictures. [COLOR=inherit !important]


[/COLOR]

Here is the KW_1 detector on a shipwreck. If I remember right, it was about 1996 on the wreck of a French galeon that wrecked in 1678 on the reefs of Aves Islands, off the coast of Venezuela.
This is the small coil, about 20 cm diameter. I could also change to a 32cm diameter coil for deeper targets.

[GeoMax]

Hi Thinkerer,

thanks for the pictures. [COLOR=inherit !important]


[/COLOR]

[ivconic]

3 time constants?
I don't know how relevant this is; with such detector and coil I will hunt only for described targets.
I am not interested in any different and larger targets then what I described already.

[ivconic]

specify a low, medium and high target size, shape, weight and time constant

Very well said

ivica, it is your turn to specify the precise purpose you desire for your detector. Without precise specifications, we are walking blind.

FelezJoo PI, whoever made it so far will remember the claim from the author how it is roughly based on Minelab patent.
Later on I realized it is kinda true claim.
In the patent they talk (sorry for rough interpretation, long time, I don't remember well) about 3 "vectors" aka 3 time constants.
And in FelezJoo PI code there is a function to "calibrate" to the attached coil, by measuring 3 "numbers" which actually are the 3 time constants.
So the author gave the best ranges of those 3 numbers so diyers to know how good their diy coil is matched.
There is "debug" screen you get on detector, displaying those 3 numbers and one can check the coil with them.
This is related, but it took me years to realize that!
Having said that; I think is good to return to that patent and read it with bit more attention.

[ivconic]

A few years ago, I read an article on Eric Foster’s web site that explored the optimum coil discharge time constant for stimulating a particular target time constant. After World War 2 many UXOs had to be found buried in the ground. After removing the metal parts from found UXOs they did research to build metal detectors that would best find the small metal parts. Here is the main point of their research related to this project. The military research found that the best stimulus of a metal target of a given time constant would be a coil discharge time constant five times faster than the target time constant. Since the coil discharge time constant is driven by the coil inductance divided by the damping resistor value, minimizing coil seen capacitance will allow higher damping resistor values and thus a steeper coil discharge slope to better stimulate smaller targets.

When building this metal detector for competition, someone needs to specify a low, medium and high target size, shape, weight and time constant so the coil design and fine tuning can accurately and honestly be compared in the design submissions. It would be normal for different size coils to be used to optimize the detection of each target size and time constant. Here is where fine tuning the damping resistor value is necessary when changing coils for each target.

Select commonly available metal parts as low, medium, and high target time constants for the design submissions to specify in their final report.

Creativity in circuit design also needs creativity in coil design to optimize the final performance.

Joseph J. Rogowski

It will be easy for me to answer the question about the coil.
I prefer coils from 18cm to 22cm.
Target?
Very small silver and gold coins.
Very small pieces of silver and gold jewelry, from say 2-3mm to 10mm.
Irregular shape.
The coins, however, have a more regular shape.
These are thin pieces with a small mass.
Very difficult targets.
Time constant for such pieces?
Probably below 15uS and less.
I currently have no way of determining the exact time constant.
But from your post it is clear that such a detector and coil would have to be very "fast" and with the possibility of setting the delay below 10uS, or less.
We talk about the PI detector here now.
And at the same time, I'm also interested in what the chances are for VLF I/B non-motion with similar capabilities (to detect such targets).

[Tinkerer]

A few years ago, I read an article on Eric Foster’s web site that explored the optimum coil discharge time constant for stimulating a particular target time constant. After World War 2 many UXOs had to be found buried in the ground. After removing the metal parts from found UXOs they did research to build metal detectors that would best find the small metal parts. Here is the main point of their research related to this project. The military research found that the best stimulus of a metal target of a given time constant would be a coil discharge time constant five times faster than the target time constant. Since the coil discharge time constant is driven by the coil inductance divided by the damping resistor value, minimizing coil seen capacitance will allow higher damping resistor values and thus a steeper coil discharge slope to better stimulate smaller targets.

When building this metal detector for competition, someone needs to specify a low, medium and high target size, shape, weight and time constant so the coil design and fine tuning can accurately and honestly be compared in the design submissions. It would be normal for different size coils to be used to optimize the detection of each target size and time constant. Here is where fine tuning the damping resistor value is necessary when changing coils for each target.

Select commonly available metal parts as low, medium, and high target time constants for the design submissions to specify in their final report.

Creativity in circuit design also needs creativity in coil design to optimize the final performance.

Joseph J. Rogowski

specify a low, medium and high target size, shape, weight and time constant

Very well said

ivica, it is your turn to specify the precise purpose you desire for your detector. Without precise specifications, we are walking blind.

[Tinkerer]

Hi Thinkerer,

thanks for posting this great PI. Would be interesting to see some pictures of the actual detector and/or assembled PCB as well.

It is amazing that these PCBs have survived all these years. I guess you could slap this circuit together on a breadboard and learn the basic functions of PI.
However, simple as it is, it was doing it's job at the time, so long ago.
After that, we can see how we can improve it.
Lots and lots of improvements possible.

[bbsailor]

Totally agree!

A few years ago, I read an article on Eric Foster’s web site that explored the optimum coil discharge time constant for stimulating a particular target time constant. After World War 2 many UXOs had to be found buried in the ground. After removing the metal parts from found UXOs they did research to build metal detectors that would best find the small metal parts. Here is the main point of their research related to this project. The military research found that the best stimulus of a metal target of a given time constant would be a coil discharge time constant five times faster than the target time constant. Since the coil discharge time constant is driven by the coil inductance divided by the damping resistor value, minimizing coil seen capacitance will allow higher damping resistor values and thus a steeper coil discharge slope to better stimulate smaller targets.

When building this metal detector for competition, someone needs to specify a low, medium and high target size, shape, weight and time constant so the coil design and fine tuning can accurately and honestly be compared in the design submissions. It would be normal for different size coils to be used to optimize the detection of each target size and time constant. Here is where fine tuning the damping resistor value is necessary when changing coils for each target.

Select commonly available metal parts as low, medium, and high target time constants for the design submissions to specify in their final report.

Creativity in circuit design also needs creativity in coil design to optimize the final performance.

Joseph J. Rogowski

[GeoMax]

Hi Thinkerer,

thanks for posting this great PI. Would be interesting to see some pictures of the actual detector and/or assembled PCB as well.

[ivconic]

Here is, I hope, finall schematic.
Guess what... I will make it.
I like such projects!
...
Just one more thing; bone conductor... kinda mechanical transducer right?
I would like there speaker instead it.
Would not be a problem to rearrange that part.
Thanks!

[ivconic]

OK, now I understand what you mean. A "Figure 8 coil".
There are quite a few variations possible.
Which variation is the best?
The best variation is the one that best fits your purpose.
Before deciding on what coil, we must decide on the purpose of the detector.

The best electronics become useless with a bad or wrong coil.

Totally agree!

[Tinkerer]

Tony I tried to redraw your schematic in a way more suitable for my way of undrstandings.
Though there are few things you can help me to understand.
I put marks (circles, elipses) on the parts I would like to clarify, their values.

I will look it up. I actually still have one or 2 assembled PCBs.

Tentatively:
C1 - clock timer 120 to 150pico F
C2,3,4,6, = pico F, the last number being the number of zeros = 104 = 100,000pf = 100nF or 0.1mF

? at the end = bone conductor clicker. Set at about 2 Hz for no target. Proportionally increases to about 120Hz with target saturation.

If I reember right, the detector was running at about 3000cycles per second.

[Tinkerer]

I was thinking on "8" shape (twisted at center) and not on size like 8" (inches).
"8" shape coil has it advantages as more immune to noise and interferences.
But it is good that you explained.

OK, now I understand what you mean. A "Figure 8 coil".
There are quite a few variations possible.
Which variation is the best?
The best variation is the one that best fits your purpose.
Before deciding on what coil, we must decide on the purpose of the detector.

The best electronics become useless with a bad or wrong coil.

[ivconic]

Tony I tried to redraw your schematic in a way more suitable for my way of undrstandings.
Though there are few things you can help me to understand.
I put marks (circles, elipses) on the parts I would like to clarify, their values.