2nd Chance PI

[dantech]

That feature works properly. I have verified one.
Remember that every calibration/recalibration (eg. auto coil delay test during the start or filters reset) should
be done when the coil is raised up and away from metal objects.

[glowian]

Hi, got coil error 0us pulsing sound where to start fix problem??? coil 320uH 1.2ohm??

[DonCaffeDelaSkobalj]

Dobry dzień dantech.

Great project. Any new info.



If you can try to improve dynamic range of 2'nd Chance PI OP using some good and fast R2R rail-to-rail OP instead old NE5534.

Good examples are OP from microchip. My choice are MCP 660-669. Very high gain x bandwidth range of 60MHz, with very fast Skew rate 32V per microsecond are ideal for amplifying nano and microvolt signals from fast decaying targets like pure gold nuggets, golden rings, golden seeds, small golden coins at high deep, or normal size golden coins in highly mineralized ground.

http://www.microchip.com/ParamChartS...ranchID=110153

Also clamping diodes at should be schottky type smd diode. At output of OP directly to op-out pin should be some pure resistive load from 10 to 600 ohm to stabilize OP circuit in wide range of input impendances. S-parameters tuning. Good values from 330 to 620 ohm.

Transmitting part should be corrected. Instead old PFET 9640 with 11A drain-source current, 200V breakdown voltage and, 0.5 Ohm,drain-sourc4e channel we need something newer with continuous drain current over 50A, with peak impulse current over 200A and drain-source resistance in 0,0xx range,

For penetrating deep ground soil we need hyper sharp fronts of PI impulse. It means rising and faling times in several to tenths of nano-seconds. To get this we need high speed high current driver special designed for draiving high current mosfets/igbt transistors with big input gate-source capacitance Cgs from several nano farads. To very fast charge this parasitic capacitance we need very high current pulse direct from mosfet driver.

My choice is ixd630 30A peak pulse current

http://www.ixysic.com/home/pdfs.nsf/www/IXD_630.pdf/$file/IXD_630.pdf

Puting 3 in parallel we have 90A current pulse. Square this value multiple with coil inductance and devide by 2 and multiple with number of target kicks per second and you will know how much energy hit the gold target. This imply using very high electrolytic capacitor usualy connected to source of mosfet to supply driver and mosfet with enough peak current in range 100A-200A. Value bigger than 50 000 uF will be fine. This capacitor is main source of detector power and only he could release several tenths or biger enregy pulses in properly designed coil.

And for the end negative voltage pump LT1054 must be synched with ATMEGA328 directly to avoid wide range of interference problem due to uncorrelated currents in all parts of PI metal detector.

These are my little suggestions for you and other builders. Try it and see it. I don't ask nobody to belive me - but to understand me.

Reg,
DonCaffe

[baum7154]

Hi Don,

You have made some good suggestions.

The original Chance PI design used an OP37 amp that is much more capable than the NE5534. If you have not already done so, take a look at the original Chance PI schematic.

Regards,

Dan

[DonCaffeDelaSkobalj]

Hi Don,

You have made some good suggestions.

The original Chance PI design used an OP37 amp that is much more capable than the NE5534. If you have not already done so, take a look at the original Chance PI schematic.

Regards,

Dan

Thank you Mr.Baum7154

Dantech new Chance PI project look good and worth trying. If Mr.Dantech find time and want to improve Chance project i may give him new information how to hit GOLD and Silver coins with 100 kWatts and much more impulse power. Same techinque as is used in GPR-s ground penetrating radars and particle accelerators.

It is matter of work many very sharp nano impulses several kV amplitude with slopes even in picosecond domain. So instead using one or hundred microsecond impulses with low amplitude 400-500V, with big pulse width and slow fronts we are using for deep penetrating targets many nano impulses from several tenths to several thousands, packed in only one micro second impulse.

Similar technique used in Mineab PI detectors known as MPS - multi impulses in one period transmitting and sensing. And physicist Bruce Candy and Minelab know how to make good detectors.

Regards,
DonCaffeDeLaSkobalj

[dantech]

v026 is released

[phred01]

v002
http://chomikuj.pl/dantech/2nd+ChancePI

tried to download but site is difficult navigate in particular that payment is needed in zolties is there paypal available? secondly can the whole suite be download @ once rather than piece meal and asking payment each time?

[glowian]

Good morning

[phred01]

Thanx Glowian now I can upgrade the unit built a yr ago. Then will wind the coil for it

[phred01]

Just checked original cct compared to the 2nd chance some differences Flyback clamp, coil damping resistor & R11 from + bat to socket X3 pin 14. My main concern is the fly back clamp what impact does this change have.

[phred01]

found answer in the tread will have remove & change components

[phred01]

I have traced back to the original web site in Russia and found this interesting post

Location: Poland	Translated by google I need more feedback from field! v1.2.8 - Restore digital filter algorithms originally used by Fandy - Optimization of digital filters - Failure of the OK button to reset the filter. Now the reset should only be done during start-up and change the guard time in the menu Due to the continuous construction of the dynamic "background image processor" with a weak signal from the object coil, the signal strength will be seen on the LCD. Output of the second coil on the sensitivity recovery side. The same overall functionality and sensitivity reaches the detector only after the start. If the threshold is set high barriers This detector operation may resemble some type of dynamic detector operation. With strong signals, the effect is unlikely to be significant even for small signals, if it has information activities, this is not a problem. The dynamic nature of the construction work "background image" generates a small signal is masked use of the "barrier". Touching the coil to ground can generate a signal for a short time, but after a while the processor changes the "background image" Start the detector and change the protection period we do when the coil is in the air, and not on the ground. It is suggested to put "Volume raise" on 1/8 Last edited by pawol : August 13, 2016 16:18; edited 1 time in total.
to come back to the beginning

pawol

Post subject: Re: Chance Posted: Fri: Jul 08, 2016 23:21

Registered: Wed: Jan 30, 2008 5:59 pm Messages: 100 Location: Poland

firmware Chance PI v1.30 Slowly this is beautiful List of changes: v1.2.9 - code optimization - Limiting the maximum height from the frequency of the generated sound. Sometimes the frequency was too great and could no longer hear it. Now the range of frequencies generated This is 100Hz and 200Hz for a distortion object of 5.5kHz - Changed the masking algorithm. Currently, there are 4 masking modes: 1 - colors only 2 - only iron 3 - only gold and bronze 4 - only silver, copper and aluminum v1.3.0 - Small optimization digital filter - Added submenu "VDI1 - left limit" and "VDI2 - right limit." Set there and the boundary criteria VDI1 VDI2 is used in masking mode. Mumer VDI is the number of LCD segments VDI on the indicator in the top line is numbered from left to right. The area on the left side of the LCD to "VDI1 is the left limit" area for gold and bronze and the "VDI2 - right limit" area to the right of the LCD is an area for copper, aluminum, silver. The area between "VDI1 is the left limit" and "VDI2 is the right limit" is the area for the iron. Last edited by pawol : August 13, 2016 16:18; edited 1 time in total.

[phred01]

from russian web page translated
pawol
Post subject: Re: 2nd ChancePI
Сообщение Posted: Fri: Oct 28, 2016 8:48 pm
Не в сети

Registered: Wed: Jan 30, 2008 5:59 pm
Messages: 100
Location: Poland
v012 is available

Because of question from users I have to clarify a little bit ...
Entire Q6 solution was a bit "fictious" in the original Chance PI. It did not work properly in fact
did nothing usefull instead of additional and unwanted 100mA power drain
So I do not drive Q6 in the 2nd Chance and I have changed the way of coil driving.
For instance the original Q6 solution drains 180mA but my solution drains 80mA.
Is there any difference?
Like I said Q6 solution was not succesfull
Should not be difference if Q6 circuit is placed or removed from PCB when using my software.
In my software Q6 is driven as inactive whole time. The ICL7660 / LT1054
is enabled whole time also.

So, simplifying, my release should be backward compatible. I do not test it at all.
For purity reasons I have removed Q6 circuit from my pcb completly and new schematic diagram and pcb.

By the way, a few things good to know about 2nd Chance PI:
- OK button pressed whilst starting restores default settings
- Menu button pressed whilst starting direcly into menu without battery test.
It can be usefull when you set battery U min level higher than voltage battery you have.
Normally you will see "low battery". So use this method.
- Default settings are proper in my case. Everyone should test the device and enter own settings

Posted on: January 09, 2018, 12:31:10 pm

---

2nd chance video
youtube.com/watch?v=AdzgmCRVShM

[dantech]

Hi,
Firmware v27 is released:

https://chomikuj.pl/dantech/2nd+ChancePI/Firmware

[fekemil]

I'm nearly finished my circuit. Not too elegant, some ugly soldering, but looks short circuit free. And I have some questions
What is the purpose of R26? (Volt meter adjust) What value should be seen on the voltmeter while turning the knob? And where can I measure?
The other thing is the Atmega328P. I ordered an USBasp programmer, which I think good to program this IC. I have a blank IC, do I need program bootloader in it? I think the answer is no, because this is not Arduino. The buttons and the programmer will arrive shortly. I have to learn to program this chip so far.
Thanks