2nd Chance PI

[CDI]

I did not try bigger coil but in theory it should work well.
Every process parameter can be customized in the menu in the very wide range so I do not see technical problems to use "frame coil". Of course you have to wind it by thick wire. to keep low resistance as possible and to keep inductance 300-400uH. The coil should be simply fast because pulse width is 120us and the coil should be able to saturate in this time
I saw the poeple has made frame coils for classic ChancePI

Thank's dantech, i will try 50 cm coil or less... but now i'm waiting for pcb and parts.

[ripsdevala]

Any problem, any progress?

Good morning all,
Not yet, i like to get 0.8mm wire instead, for stock , btw i dont know what diode should be place on the keyboard D1 - D4, or its just 4148
Do you have pic top of your pcb, i like to make sure all is already on place.

[dantech]

No, I do not have any picture inside. The detector is completed and closed in the box

[dantech]

Firmware v018 is released
Now it is possible to change main pulse width in the range 50-200us to fit pulse width and saturate properly the coils with different speeds and sizes (eg. very small basket coils and large frame coils).
"Restore to default" feature is moved into menu.

[dantech]

v019 is released.
Anti-commercial lock changed. After first 100 runs there is no permanent lock but it is necessary to wait additional 300s or reflash cpu again

[ripsdevala]

Is this winding ok? Have gap like this?

[dantech]

should be okay

[ripsdevala]

should be okay

Ok, i will continue then, many thanks

[ripsdevala]

My board not work, just blank lcd, no sound, anyone have voltage reading?

[dantech]

1. Remove all ICs
2. Check power supply lines +5V, +12V, -12V. Power consumption should be circa 30mA
3. Check shortcuts on PCB near ATmega328
4. Verify Atmega328 fuse bits and flash memory content
5. Put ICs back
6. Check oscillogram on crystal pins
7. Check signals on ISP bus (eg. clock signal for ADC)
8. Check LCD connections
9. Check LCD contrast resistor

[dantech]

7. Should be SPI not ISP

[dantech]

v020 is released

[ripsdevala]

1. Remove all ICs ....done
2. Check power supply lines +5V, +12V, -12V. Power consumption should be circa 30mA
+5 ok, alittle hot 78l05, +12v where?, -12v ?
3. Check shortcuts on PCB near ATmega328.... No short
4. Verify Atmega328 fuse bits and flash memory content
...done, all ok
5. Put ICs back... Done
6. Check oscillogram on crystal pins... What is that??
7. Check signals on ISP bus (eg. clock signal for ADC)
....Never hear that
8. Check LCD connections ... All ok
9. Check LCD contrast resistor.... Cant find R24 to adjust contrast

Im not an electro expert

[dantech]

Ad. 2 +12V/-12V you will find for instance on pin 5 and 8 of LT1054. 78L05 should not be hot. Check power consumption. Something is wrong

Ad. 6 and 7
You have to use an oscilloscope. Signals on crystal and SPI bus tell us that CPU is alive and tries to communicate with external device (ADC)

Ad. 9 I do not know what pcb you have but in my case I have resistor 1kohm between pin 1 and 3 of LCD

[ripsdevala]

LT1054 pin5 -10.4v, pin8 11.5v

Add 1k vr on pin 1 and 3, lcd is apear black dot, but not visible enough.
I use your pcb layout, lcd with the keyboard