My Pulse Oscillation detector project.

[Teemo]

Thanks for the information. I didn't mean to degrade your project concerning the sensitivity.
You have a great project! It's an original and clean design. Maybe treasure hunting in your future?

Regards

Yeah, I feel treasure hunting can be fun. I will try it out next summer.

[joop]

Teemo's method, not as sensitive as a pure PI. But may be able discriminate,
since different metals will affect the ringing frequency and decay.

I was exactly thinking the same when I first saw this project. I think there is a lot of potential here...

[satdaveuk]

Maybe with coil experiments you got a pinpointer, the component count is a plus

[ApBerg]

Hello,
Made the pcb, I did not have al the parts.......used a IRF730 , and some high volts diode and a 470 nf and 1000uF cap.
The coil is the red wire on a pvc tube with a ferrit from some 4 cm it in.
Sens for the coins some 3cm for the small gold coin and 5 cm for the large silver coin, the ferrit gives also some detection.

Nice as a pinpointer..?
...the ferrit core is also detected..!

Current was some 20 mA at 4.5 volts

Best regards

Ap

[Teemo]

Hello!
Very good job AP! I't is good to see.

Do you consider that it is detecting ferrite, good or bad thing? In theory it should not detect ferrite core very well.
Regards,
Teemo

[Teemo]

IRF730 may not to be the best MOSFET for the job. According to the datasheet it needs pretty high gate to source voltage. So it is probably limiting current.

[ApBerg]

IRF730 may not to be the best MOSFET for the job. According to the datasheet it needs pretty high gate to source voltage. So it is probably limiting current.

Hello Teemo,

Yes the current is only some 20mA..I like that, with your fet how high is then the current ?
The detected ferrite is giving more sound when it is removed from the coil, can it be that I have the wrong working freq.?

Found that the circuit can use many coils.. it works at many different one's. The last I used (on the photo) has a ferrite in it, by shifting that ferrite different uH 's can be made ,the circuit alway's works..

It is a nice circuit and if it can reject the ferrite,( ground effect) it is a good pinpointer, again thank for sharing !

I have not much testing done, the room were my scope stands is not heated...

Best regards,

Ap

[Teemo]

Hello Teemo,

Yes the current is only some 20mA..I like that, with your fet how high is then the current ?
The detected ferrite is giving more sound when it is removed from the coil, can it be that I have the wrong working freq.?

Found that the circuit can use many coils.. it works at many different one's. The last I used (on the photo) has a ferrite in it, by shifting that ferrite different uH 's can be made ,the circuit alway's works..

It is a nice circuit and if it can reject the ferrite,( ground effect) it is a good pinpointer, again thank for sharing !

I have not much testing done, the room were my scope stands is not heated...

Best regards,

Ap

Thanks Ap,
With my mosfet K2134 average current draw is about 64mA.

[Teemo]

Hello
I made new experimental version of firmware code, asm and hex available: http://digiwood.ee/en/metal-detector/firmware-page
It now uses frequency information also. Much less reaction to the ferrite.

AP, your current consumption is so small that, it is a miracle it works. 20mA, Only the LED takes half of that, and other half goes to the microcontroller and coil. This new version code might not work with that mosfet.

All the best,
Teemo

[ApBerg]

Hello Teemo

Nice! a update from the software, next week hope to test it. The led I used is a low current one, I think only some 2mA..
Well see if I have an other fet.....from the IFR730 I have plenty, so it is a first to use in experiments..

Best regards

Ap

[joop]

Hi Teemo,

I noticed an error on your website, the link on your firmware page called metaldetector110.asm is pointing to the file metaldetector100.asm

[Teemo]

Thanks for pointing out, I corrected the link now. Sorry, about that.

Happy holidays,
Teemo

[Teemo]

Hello.
I made some tests today to see how my circuit works with different capacitors for C1. I just got some new Wima MKP4 47nF 630V- polypropylene capacitors today.
And it turns our that differences are HUGE. Oscillation lasts much longer with good quality capacitor. With polyester (MKT) capacitor (47nF 400V), not sure what brand, I had first the circuit detected about 64 oscillations before the oscillation decays under the treshold. But with the new WIMA cap it detects 96 oscillations. So it really matters. And the decay value (t_result in assembler program) is twice with WIMA, than it was with the other cap. Frequency did not change much.

So it is essential to have a good capacitor for C1.

But also this was too good, for the circuit to work, because oscillation from the last pulse is not decayed yet when the new pulse is coming on. So with too good C1 and low resistance coil circuit may not work at all.

Teemo

[johnandles]

looking forward to build this project.
Just waiting for my programmer to arrive.
I have the Pic12f...
Anything I should be aware of about programming the Hex ?
Cheers
John

[Teemo]

Hi John,
There is nothing special about programming the hex. You may experiment with the parameters in asm file if you wish to tune: the sensitivity and compref(comparator reference voltage).
Looking forward to hear how you are doing.