felezjoo PI(the best pulse induction metal detector that I made until now)

[ivconic]

the same problem MR ivconic DEMO!!!

You must delete eeprom content anyway.
Otherwise i don't know how to help you.

[faycel]

You must delete eeprom content anyway.
Otherwise i don't know how to help you.

thanks

I will use atmel studio and arduino avr isp mk2 as programme and i will report the result

[ivconic]

Hex file i provided in another post (EEPROM_CLEAR_IC.zip) is deleting eeprom completely.
You have to program it into Atmega and put power ON for few seconds.
After few seconds eeprom is completely clear.
Than you'll program the same Atmega with FelezJoo PI hex and that's all.
FelezJoo PI code is using eeprom for storing variables.
On each powering ON first it reads those variables.
First variable to read is the "counter" variable.
If you reach the counter's limit; code will stop and display "DEMO!!!" on LCD.
Or if you tried to hack the code or any other way messed with it; again it will stop the execution and displays again the "DEMO!!!" message on LCD.
Author of FelezJoo PI is great programmer and very smart man.
He protected his code pretty well.
Earlier versions of code were easier to hack and to disable counter.
Each new version is carrying new methods of protection.
For a start is just enough to respect regular way suggested by author, until you are sure that you are having fully working detector.
Everything else will lead you to headache and problems with detector and how it is functioning.

[faycel]

Hex file i provided in another post (EEPROM_CLEAR_IC.zip) is deleting eeprom completely.
You have to program it into Atmega and put power ON for few seconds.
After few seconds eeprom is completely clear.
Than you'll program the same Atmega with FelezJoo PI hex and that's all.
FelezJoo PI code is using eeprom for storing variables.
On each powering ON first it reads those variables.
First variable to read is the "counter" variable.
If you reach the counter's limit; code will stop and display "DEMO!!!" on LCD.
Or if you tried to hack the code or any other way messed with it; again it will stop the execution and displays again the "DEMO!!!" message on LCD.
Author of FelezJoo PI is great programmer and very smart man.
He protected his code pretty well.
Earlier versions of code were easier to hack and to disable counter.
Each new version is carrying new methods of protection.
For a start is just enough to respect regular way suggested by author, until you are sure that you are having fully working detector.
Everything else will lead you to headache and problems with detector and how it is functioning.

thanks ivconic the problem now is solved
thanks for your help
I erase everything flash and eeprom and with new fuses all is ok

[faycel]

now its time to do this in real pcb and coil
thanks ivconic

[ivconic]

now its time to do this in real pcb and coil
thanks ivconic

[gurdal]

[QUOTE = faycel; 229762] şimdi bunu gerçek pcb ve bobinde
yapmanın zamanı ivconic [/ QUOTE]

hi ivconic,

hi Faycel

Congratulations

You have succeeded

[slimabidi]

Congratulations

You have succeeded

[jladre]

[QUOTE=ivconic;229135]I just completed this one.
So far it performs excellent.
Later on i intend to take it outdoor for final testings.



Any news??

[ivconic]

Any news??

Yes. First time i took it outdoor for initial testings; it was nightmare!
Very unstable, producing false signals all over the place.
Total miss.
Quite opposite from the first device i made last summer.
Than i emailed author asking for help and suggestions.
He gave me several suggestions and i did few of those.
Most important change that made huge difference is coil cable!
Damn!
I used stereo microphone cable and it was the cause of great instability.
Related to this; my sketch from another post is obviously wrong, it should be kept in mind.
Author suggested me RG-58 and i replaced the cable. Much better situation.
Than i went outdoor again to check it.
Situation was much better, but still i was not satisfied completely.
Than i did the following;
1) Put detector on bench, fixed coil in air, away from any metal and interference.
2) Monitored all the time debug numbers, how they changing with every next step that i made,
3) I subtracted one turn from a coil and got 0.384mH inductance, measured on the end of a coil cable.
4) Replaced IRF840 with IRF740; resulting in slightly better debug numbers.
5) Replaced 2u2 tantalum capacitor with ordinary plastic 150nF non-polarized one; resulting also in slightly better debug numbers.
5) Added 180 ohms resistor in series with existing 100 ohms; resulting very good debug numbers, almost excellent.
6) Removed IC socket from LF357N place.
Than i went outdoor again. Same spot in forest.
Results; excellent! Very calm, no false signals at all. Behavior excellent. Discrimination works as hell! Better than Golden Mask 1+ which i carried along.
But discrimination range is moved now and starts at "15" adjustment.
Depths? I can't tell what would be the limits, i buried several shallower targets around. All were detected with very strong response and coil significantly lifted above the ground.
It is really irrelevant now at this stage to measure exact ranges and depths. It is deep as much as is possible with 36x31 coil.
Certainly is much deeper than Golden Mask 1+ that i had with me on that occasion.
...
Conclusion? Tough to say anything relevant now.
Some steps that i made are directly opposing from what author suggested. I guess it is because of the unacceptable tolerance between components that i used here and components that he used there while writing a code for Atmega.
It is very hard to replicate the same conditions.
We talked about this before too, for example; the LF357N... i am having 4 of those and each one when put on ic socket is causing quite different debug numbers!
So it is a case from case situation.
One must have patience and involve effort to "trim" up own device. There is no general rule nor general advice or suggestion.
Tough!

[ivconic]

Oh i forgot to mention one important thing more!
Pulse Width is playing very important role in calculating debug numbers.
Initially PW was set at 150.
Than i checked the debug numbers by rising PW up to 220 and on each step i got better debug numbers and much calmer and more stable behavior.
When PW is set at 180 and above; detector behaves perfectly! Just splendid!
But be aware that this is the situation closely related to coil specs too.
If you get different behavior than it's probably due differences in coils.

[gurdal]

Many thanks ivconic,

[jladre]

It is great that you settle right away the problem ivconic..
Your tips will definately a big help for those got problem getting the right debug numbers.. the irf 740 will also give you a lway a better performance.
The main things that need to check is the coil cable andvthen the coil. My tips to the other members who have difficulty to get the right debug numbers is fucos first on the coil. Do not shield your coil, for checking your coil if it is correct use the same wire used in coil and make it as a temporary cable. Do not twist just let it go parallel. Then check your debug number, and callibrate it by adding or subtracting turns on your coil..after getting the right numbers replace the temporary cable with a right one. RG-58 is advisable or like my cable below.
In my case i solve always tge debug number by just on coil cable and coil.

[jladre]

[jladre]

There is a simple trick to get the debug number. Just attach a very small piece of metal to the coil then press rebalance. Continue to add or subtract piece of metal until you get tye number.