Tweet
Here is the latest schematic diagram revision of my detector
Regards
Pawol/Poland
Regards
Pawol/Poland
-
-
Hi dantech,
How long is you programming for AVRs?
Regards,
1843Comment
-
Hi 1843,
Why do you ask?
I programm AVRs, I think, 4 years time. I do it when I really need it
Comment
-
Ok, 4 years programming is good.
You waste your time by writing programs for the such PIs.
I suggest you to write programs for Goldscan4 or Sniffer-XR71.
Have fun,
1843Comment
-
This is my first detector. In 2 weeks time I changed completly firmware idea and it was completly rebuilt. Next stage will be sampling by means of ADC and discrimination. I plane still to base on ATtiny-1 environment because it works stable and one is sensitive. It will be a question of another AVR ,ADC and LCD (I find it will be possible to use AVR internal ADC and get delayed samples within a few pulse cycles). To get proper disctimination I see I will have to use some automatic gain control for receiver.
BTW. I publish it because it is easy to make for many people still having good parameters
Comment
-
sorry , but bad german copy on your hand .
Comment
-
sisco:
Hi,
Like you shouls see earlier my MD is modification of Attiny-1. Firmware is mine but schematic i based one one. Look at the earier post:
Quotation:
"Here you have the latest firmware revision for my detector (based on ATtiny-1 project from pulsdetektor)."
My modified schemtics is working and little bit modifiedComment
-
The latest firmware for my modified attiny-1
Main changes: Switchable pulse width now is 50/100us. Delay1 range is 15-95us with 5us step. One is now quite sensitive for a gold. Code is optimized. Now it is possible to change quickly any width in the source code witch step 5us.Future version will take parameters from eeprom so anyone will be able to modify timing before uC programming.
:020000020000FC
:1000000008C018951895189518951895189518956D
:10001000189507E701BF0FE101BD07E101BD00E44D
:1000200005BF0AE107BB00E008BB89E093E1A4E05B
:100030006BE071E0382FB09901C0392F02E008BBA6
:10004000232F2A9500000000000000000000C9F7DF
:1000500000000000000046D02A2F250F2A9500003E
:10006000E9F7000000E108BB272F2A9500000000F7
:10007000000000000000C9F700000000000032D0BE
:10008000262F2A9500000000000000000000C9F79C
:100090000000000008E008BB272F2A9500000000A0
:1000A000000000000000C9F70000000000001AD0A6
:1000B0001CD027D02FEF2A1B251B2A950000E9F71B
:1000C0002FEF231B271B261B271B2A950000000050
:1000D000000000000000C9F728E02A95F1F70000B1
:1000E0000000A8CF00E008BB089501E207B900E0D6
:1000F0000FBF08EC06B93699FECF45B108E006B946
:0C0100000895542F569556955695089575
:00000001FFComment
-
The latest release taking width parameters from eeprom. You can change values directly in .hex file (containing eeprom data of course) or in the PonyProg editor window (eeprom starts at h400 adress space). Every value in the eeprom is hexadecimal, so youhave to recalculate them manually.
Meaning of bytes in eeprom and formulas to calculate widths: (data from eeprom content you see below)
byte0 - hex value 77 / dec value 119 - OSCCAL value,fine tuning for internal CLK=1,6MHz or for PULSE freq. = 500Hz
byte1 - hex value 09 / dec value 9 - PULSE width value1 = 5us+(byte1*5us) = 50us
byte2 - hex value 13 / dec value 19 - PULSE width value2 = 5us+(byte2*5us) = 100us
byte3 - hex value 04 / dec value 4 - PRE-DELAY1 value = 5us+(byte3*2.5us) = 15us
byte4 - hex value 0B / dec value 11 - DELAY2 value = 5us+(byte4*5us) = 60us
byte5 - hex value 01 / dec value 1 - SAMPLEs width value = 5us+(byte5*5us) = 10us
Important - avoid value of 00 (zero), some of values combination can be not allowed eg. total sum of delays will exceed 3200us (500Hz)
OSCCAL value can be read by means of PonyProg or you can try some values to get PULSE freque near 500Hz.
As a PRE-DELAY1 I mean the lowest width of main delay. So total main delay (delay1) is sum of PRE-DELAY1 and delay coming from variable resistor (+ up to 80us with 5us step)
I did not test it on live MD but it was simulated very well. Previous release was good and current changes were very small.
flash content
:020000020000FC
:1000000008C018951895189518951895189518956D
:1000100018950FE101BD07E101BD00E405BF0AE14C
:1000200007BB00E008BB00E07CD021BF01E079D035
:10003000822F02E076D0922F03E073D0A22F04E04B
:1000400070D0622F05E06DD0722F382FB09901C0AB
:10005000392F02E008BB232F2A9500000000000082
:1000600000000000C9F700000000000046D02A2F61
:10007000250F2A950000E9F7000000E108BB272FB3
:100080002A9500000000000000000000C9F70000F1
:100090000000000032D0262F2A950000000000004A
:1000A00000000000C9F70000000008E008BB272F8F
:1000B0002A9500000000000000000000C9F70000C1
:1000C000000000001AD01CD027D02FEF2A1B251BC0
:1000D0002A950000E9F72FEF231B271B261B271B60
:1000E0002A9500000000000000000000C9F728E089
:1000F0002A95F1F700000000A8CF00E008BB0895A2
:1001000001E207B900E00FBF08EC06B93699FECF4F
:1001100045B108E006B90895542F56955695569561
:0E0120000895E199FECF0EBBE09A2DB308952D
:00000001FF
eeprom content
:10000000770913040B01FFFFFFFFFFFFFFFFFFFF57
:10001000FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF0
:10002000FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFE0
:10003000FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFD0
:00000001FFComment
-
lz4lo:
what do you mean by pointing that PI? One is not my variant of PI.Comment
-
Dobra robota Dantech. Good work!
Comment
-
Sonar: Where are you come from?
I hope you made my variant of pi and one works well.
Comment
-
Many people in Poland buy this PI as kit. Jabel is known as 'nail master'. This simply md work realy good - but Jabel rain of nails is close here
I didnt bulit this one but i will try your 'ad-on'...
Im come from Poland - here the best dynamic metal detector is ALL-1...
Jabel kit cost here is about 45-50 dollars
ps. more about this PI... you can find here: http://www.exploracja.cyberdusk.pl/p...forum.php?f=43Comment
-
Hi,Originally posted by sonar View PostMany people in Poland buy this PI as kit. Jabel is known as 'nail master'. This simply md work realy good - but Jabel rain of nails is close here
I didnt bulit this one but i will try your 'ad-on'...
Im come from Poland - here the best dynamic metal detector is ALL-1...
Jabel kit cost here is about 45-50 dollars
ps. more about this PI... you can find here: http://www.exploracja.cyberdusk.pl/p...forum.php?f=43
it's very common problem with all PIs. Also very complex ones suffer from lot of iron detection, steel and minerals with magnetic crystals inside...hot rocks, fired bricks etc etc
I also noticed that the more the sensitivity to good targets (e.g. silver and gold) the more the sens to mag crystals that make you crazy in infested soils.
A really good solution for this problem doesn't exist (or I don't know about) and I've tested more than 30 different PIs... some bought units ... some built by my hands with lot of care on anything... circuit... components...tuning... coils... anything.
It's principle of operation that's too related to magnetic influence of iron stuff and other trash.
So I think they all (PIs) are good for beach or just in really few inland sites; in all other places they aren't any good, at least for me. I don't want find an unexploded 250Kg bomb at 3meters underground...
So don't blame Jabel... are all about the same with nails.
Kind regards,
MaxComment
Comment