Can pinpointer detecting thin chain?

[Cazavor]

Also attempting to make a NUPI pinpointer here, my progress so far has been:
- 8 mm x 160 mm "400НН" NiZn ferrite rod from Урал-112 radio
- cut into three 53 mm long pieces by filing a groove around the rod and snapping
- cores taped together with electrical tape
- first layer 104 turns of 0.32mm wire
- second layer 79 turns of 0.32mm wire = total 183 turns in L1 giving 1.56 mH
- third layer 100 turns of 0.32mm wire giving 0.42 mH, combined whole coil inductance 3.57 mH, coil finished
- wired up the NUPI FINAL 2014 circuit on the breadboard, with some changes due to available components:
LM2931-5.0 regulator for low dropout
TR2 is S9014
TR1 is S8550 with beta of 260
100k resistor in series with the pots to find the correct bias point which turned out to be 136.6k
C4 is 2.2 nF ceramic
C3 is 10 nF polypropylene
C1 is 1 nF polypropylene
C2 is 22 nF + 10 nF polypropylene in parallel
- nice sine wave on L1, frequency approx 15.3 kHz

Really impressed with the circuit considering the low amount of parts. The sound is very nice. The power consumption is just 1.2 mA with no metal around. But the PC buzzer takes 80 mA at full blast. That'd be 0.7W from a 9V battery! Sure doesn't sound like that. Must be possible to reduce. With loose balancing I detected 1 euro reliably from ~5 cm. With on the edge balancing more is possible, but I didn't bother yet due to higher instability/drift.

Also I checked on the scope that if I put a metal object closer to the coil the oscillator amplitude decreases and beeping starts. If I put a larger object really close, then the oscillator collapses and the sound stops completely. Any way around that?

[WM6]

Nice building process presentation.

[Polymer]

Now that I have measuring equipment and more experience, I have revisited this project.
Made a new coil to the NUPI uH specs by measuring whilst using the Chemelec schematic.



This pinpointer runs extremely stable, compared to the ones I made before.
Running at ambient 35?C whilst lying in the sun and tested in the fridge at 4?C.
No more sensitivity to movement in the air when waving around (compass effect).


No more reaction to probing pressure in the ground.
Distance for a nickel is about 2 1/4 Inches.


Not as far as the previous ones with optimal bench conditions, but waaaay better out in the field.

No more constant adjustments of a finnicky little critter.


The ferrites I have used now are NOT from am radios.
They are from inductive welding. 5cm long x 9mm diameter in this case.


I have also NOT potted the ferrite coil in the pipe this time.
Its "suspended" in the pipe with two wound foam strips front and back of the ferrite.
So the ferrite is now mechanically decoupled from twisting, bending, pressing when probing.


Have also used proper pots instead of the tiny 10 turn ones this time.

[Cazavor]

Good to hear it works for you. What type of components are you using to achieve such stability? My attempt which you can see a few posts above was drifting like crazy, forcing me to scrap the project. Plus the overload problem, which I presume you don't experience?

Also I found UK patent GB2104660A from the 80s which seems to describe a similar circuit using the same oscillator and frequency divider. Maybe it could help someone understand the principles behind this pinpointer project.

[Polymer]

Good to hear it works for you. What type of components are you using to achieve such stability? My attempt which you can see a few posts above was drifting like crmelec.
I cant find the file azy, forcing me to scrap the project. Plus the overload problem, which I presume you don't experience?

Also I found UK patent GB2104660A from the 80s which seems to describe a similar circuit using the same oscillator and frequency divider. Maybe it could help someone understand the principles behind this pinpointer project.

I use PP capacitors for the 33nF & 1nF capacitors. It looks like all the box capacitors you are using are PE, they are smaller than the PP ones.
Instead of a ceramic for C4 I have used a PE capacitor. Not sure if these components make such a difference.

What definitely is very different, is the ferrite I now use.

Breadboarding and having the coil wires flopping around also makes it more unstable.
Cant find the file right now, I have been using a printout which has been in the box "to do" for years now.
See if you can find the schematic from Chemelec.

I know how frustrating and amazing the NUPI pinpointer is. Dont give up.

I will build another one the next few months and hope to achieve similar results hoping this one is not some random luckiness.

As you can see, I do not fully understand this myself yet.

The transistors I used are 2n3904 and 2n3906, nothing special.

Ha, the patent you posted is from Whites ... perhaps Carl might like to comment.

[6666]

I use PP capacitors for the 33nF & 1nF capacitors. It looks like all the box capacitors you are using are PE, they are smaller than the PP ones.
Instead of a ceramic for C4 I have used a PE capacitor. Not sure if these components make such a difference.

What definitely is very different, is the ferrite I now use.

Breadboarding and having the coil wires flopping around also makes it more unstable.
Cant find the file right now, I have been using a printout which has been in the box "to do" for years now.
See if you can find the schematic from Chemelec.

I know how frustrating and amazing the NUPI pinpointer is. Dont give up.

I will build another one the next few months and hope to achieve similar results hoping this one is not some random luckiness.

As you can see, I do not fully understand this myself yet.

The transistors I used are 2n3904 and 2n3906, nothing special.

Ha, the patent you posted is from Whites ... perhaps Carl might like to comment.

Is this the one
https://www.geotech1.com/forums/show...205#post231205

[Polymer]

Is this the one
https://www.geotech1.com/forums/show...205#post231205

Yes, thats the one!
You can leave out the 22uF capacitor parallel to the buzzer.
I have used the 68k Ohm resistor and a 10k Ohm pot for coarse adjustment with inbuilt power switch.
The fine adjustment is a simple 480 Ohm pot, no parallel resistor.

Your needed resistor adjustment values will vary with the 4024 ic used.

Its all mounted on simple strip-board.

Good Luck and enjoy!

I will be travelling and using this pinpointer intensively the next months. I will see how it holds up.

Quick photo of mine:

[Polymer]

Forgot to answer the "overload" situation.

When metal touches the tip, my pinpointer squeals consistently after the beeping has gotten faster and faster when target getting closer.

So I dont have a "dropout" of the signal. If I remember right, the NUPI has several working points. It may have to do with this.

I like it this way. I then know I am "on top" of the target.

[Carl-NC]

Ha, the patent you posted is from Whites ... perhaps Carl might like to comment.

The patent describes a method known as "energy theft" or "loaded oscillator." It is in a family I call "proximity methods" that includes BFO and off-resonance. White's used this in their Coinmaster TR model in the 1980s:

http://www.treasurenet.com/forums/to...master-tr.html

In reality, it is not a TR. Targets create a load change to the oscillator which alters the frequency, amplitude, and DC offset. In this case, the offset is the biggest contributor to detection. In most other energy theft methods I've seen the output of the oscillator is rectified and amplitude changes are used to drive a separate VCO or other audio oscillator. The patent shows a very clever way to get the audio from the coil oscillator using a divider and then recombine it with the output offset to drive a comparator.

The patent is exceptionally well-written and exactly describes what is happening, and even has some light math if you want to understand it at that level. You don't find many patents written as well as this one. Hirschi was a pretty smart guy, I looked through some of his old engineering notebooks when I worked at White's.

[Polymer]

Thank you Carl.

Have read the patent and am puzzled by something written on page 2 / lines 32-37 ... "These eddy currents produce a secondary electrical field ..."

Is that not the definition of a PI detector? At least thats how I understand it.

[Polymer]

@Cazavor

Forgot to mention: The new coil I made for this pinpointer is shielded.

[Carl-NC]

Thank you Carl.

Have read the patent and am puzzled by something written on page 2 / lines 32-37 ... "These eddy currents produce a secondary electrical field ..."

Is that not the definition of a PI detector? At least thats how I understand it.

It's the definition of all induction-based metal detectors. Target eddy currents induce an EMF in the RX coil; for IB/VLF that's a dedicated RX coil, for PI it's the same coil as was used for TX 10us ago, and in proximity detectors (BFO, energy theft) it's the same coil as is continuously transmitting.

I re-read the math section, and I'm pretty sure there are major errors. Eq 6 does not look at all right, and Eq 7 & 8 are definitely not the derivatives of 6. If I get time I will try to grind through it and see what it should be.

[dbanner]

Further down, Q is taken at the resonant frequency, fine, but the equation for W ??? Shouldn't it be 1 divided by the square root of L*C ?
I probably arose on the wrong side of the bed this morning. Need coffee.

[dbanner]

Oh dear God, radians per second. I'm outta here!

[Cazavor]

I use PP capacitors for the 33nF & 1nF capacitors. It looks like all the box capacitors you are using are PE, they are smaller than the PP ones.
Instead of a ceramic for C4 I have used a PE capacitor. Not sure if these components make such a difference.

What definitely is very different, is the ferrite I now use.

Breadboarding and having the coil wires flopping around also makes it more unstable.
Cant find the file right now, I have been using a printout which has been in the box "to do" for years now.
See if you can find the schematic from Chemelec.

I know how frustrating and amazing the NUPI pinpointer is. Dont give up.

I will build another one the next few months and hope to achieve similar results hoping this one is not some random luckiness.

As you can see, I do not fully understand this myself yet.

The transistors I used are 2n3904 and 2n3906, nothing special.

Ha, the patent you posted is from Whites ... perhaps Carl might like to comment.

The yellow capacitors I bought from aliexpress named "Polypropylene correction safety plastic film capacitors" with no datasheet. Who knows what they really are. I measured them at 0.3 ohm ESR and no significant leakage.

I built a perfboard version of the original NUPI circuit so board is not the problem. Attached a few pictures including some of my finds with solely this pointer.

I might revisit this project in the future but not at the moment. I have my hands full with a PI detector and also currently I'll get by with my two 20? pinpointers from China, one is PI and the other energy theft.