PICKINI V4 - an easy to build, self adjusting PI detector

[ripsdevala]

Imposible, sorry

[ripsdevala]

Hi Bernard, is it possible to use direct stepup 5v tp4056 single powerbank charger without 78L05?

Thanks

[BC547]

The current hex file that can be downloaded is for PIC16F1824 specific, since it uses specific hardware resources of this device.
No idea if you will find the same HW functionality (32 MHz clock, timer gating, comparator, PWM, DAC,...) inside the 16F616.
The C source code can also be downloaded
http://users.telenet.be/willaert/MD/pickini/software/v4/Pickini4.hex

[folharin]

Does anyone have the hexadecimal file ready for pickit3? Pic 16f616

[Qiaozhi]

Alguém tem o arquivo hexadecimal pronto para pickit3?pic 16f16

Please read the forum rules -> Basic Rules of the Forums
and make your posts in English.

[folharin]

Alguém tem o arquivo hexadecimal pronto para pickit3?pic 16f16

[bernte_one]

nick Holas has made a smd pcb already, Format is better to fit in a pipe
look his last postings

[ripsdevala]

Ehh.... You are right, forgot to rotare

[BC547]

@ripsdevala

I only saw this today: the symbol of the LM318 on your PCB in message #403 is upside down.

[ripsdevala]

Nice repto, addition to my collection.

[reptooyep]

This version has only one SMD componment. Here is mine :

Here is the modified lay file, i've added the prog pins and the board works fine in my underwater detector.
PIKINI SMD.zip

[ripsdevala]

I think this is correct

[ripsdevala]

Anyone have smd layout with pickit connector to flash?

[BC547]

@biciliux: hopefully you understand how this detector works now. If not, met me know.
Regards,
-Bernard

[BC547]

This detector works with "temporal sampling", a fancy term for pulse width measurement.
By closing the FET, current starts to flow through the coil. This current is abruptly cut off by opening the FET again.
We then get a huge flyback pulse ( 100ds of volts ) that decays back to 12V in this case.
When there is a target in the magnetic field of the coil, it will react to this field by emitting a small magnetic field caused by "Eddy currents" in the target.
This small magnetic field comes some time after the main field, depending on the type of metal. This is the so-called time constant. This is why you need a fast coil with very little capacitance to detect gold, which typically reacts very fast to the field.
So, when there is a target under the coil, the flyback pulse will take a little longer to return to 12V, due to this extra target field.
With every type of metal under the coil, the duration of the flyback pulse will be longer.
When you look at the block diagram, the clipped and amplified flyback pulse - we are only interested in the lower part, close to steady state = 12 V in this case - is fed into an analog comparator.
The DC level of this amplified pulse and the slicing level of the comparator are set by software = calibration routine after switching on the detector. This makes it independent of the power supply voltage / coil / electronics variations.
Pulses to the FET are applied every 2 ms. The pulse width of the comparator output is measured using an internal timer that is gated by this output.
The timer runs at 32MHz, so with a resolution of about 33 ns. The pulse duration is about 100 microseconds, so with each sample the timer goes up to about 3300.
One sample = one pulse width. The "sensitivity" potmeter determines how many of these samples are added up to have more resolution.
One complete sample consists of the sum of a number of individual flyback times.
The samples are fed into an array to have a moving average of the sample value. This eliminates drift of the measurements due to (slow) changing ground conditions.
Each new sample is compared to this moving average value. The difference determines the frequency of the audio output. This varies between 1Hz and 1kHz in a logarithmic way.
Have a look at the source code - it will clarify more than a lot of text here.
http://users.telenet.be/willaert/MD/pickini/pickini_en/V4/software.html