atmega8 programming

[silverdog]

I burn Atmega8 chips for the CHANCE project

I use a USBASP cheap programmer
http://www.ebay.co.uk/itm/2xUSBASP-U...item43aa6b900e

and extreme burner software
http://extremeelectronics.co.in/avr-...r-programmers/

[Tepco]

I’m not Atmega expert, and i have no idea why someone have trouble to program this chip, i did it dozen times (including just now) in different projects so far using simple serial programmer, as posted here.

For “unlocking” Atmega chips, usually HVPP, high voltage parallel programmer is needed, but:

Using Ponyprog, making two fatal mistakes, locking up RSTDISBL and SPIEN is not possible, so to unlock the chip, try this: Apply external clock, similar to what is expected in actual circuit, from small dip “brick” oscillator, something hand made, or from signal\function generator (AC coupled, add some 100n cap in series), no need to be TTL level, volt RMS or so will do, to pin 9, remove xtal and caps and try.

Otherwise:

http://www.youtube.com/watch?v=EHOxc-FvaUs

Real high voltage programming, fuses erased 100%

[6666]

Tepco I tried this http://www.youtube.com/watch?v=EHOxc-FvaUs

but its too hard to get the chip back in the socket with the legs burned off

[satdaveuk]

I burn Atmega8 chips for the CHANCE project

I use a USBASP cheap programmer
http://www.ebay.co.uk/itm/2xUSBASP-U...item43aa6b900e

and extreme burner software
http://extremeelectronics.co.in/avr-...r-programmers/

I got a usbbasp on order, so together with the extreme burner should be problem solved.
Thanks very much, will let you know results

Regards

[6666]

Hey Silverdog which usb drivers did you use for this device ?
thanks


http://www.ebay.co.uk/itm/2xUSBASP-U...item43aa6b900e

[satdaveuk]

Thanks Silverdog, better late than never
Right, after all that hassel for £2.38 incl p/p you can read and write ATmegs8 and 48 know problem at all, and cant see any reason for problems within the ATmega family of other ics.
To read a atmega8 chip takes 4 seconds flat, to flash the hex file takes 10 seconds flat.
In both modes everything is varyfied incl keys.
Follow silverdogs links, everythings there regarding program etc.
The hardware driver im putting here because it doesnt come with the product and im not sure where I got it from, select X86 its does windows 7 and xp, proven by me.
The chips that I played with had been used and abused, how they still function is your guess as good as mine, they must be more robust than first thought.
Ive arased and reprogramed many times, even programed the chance project into them, then back to minipulse three version 1.10.
I still cant put my finger on why these other programmers were such a pain in the butt, but they were, and spent many hours trying different combinations etc and following other peoples advise as well as my own ideas.
Im going to donate a small sum of money to the lad who developed the xtream av programmer because its as it says powerful.
If you want to make a standalone PCB for the ATmega just use the one I put here earlier with a 8mhz crystal, its easy and it works I can assure you, proven by me.
You dont need any 5v regulated supply or reset button as its all fed via the usb, and the reset is done automatically when applicable by the extream program.
I do recomend installing LED to the +5v leg of the ATmega as proof of supply, but not a must.
Theres not much else to say apart from when opening extream all you need to do is select your IC, and make sure your on iscp mode,both are top of the page tabs, thats it dont mess with anything else because you dont need to, extream and your programmer does all the work for you.
Dont know if there is any instuctions but I certainly didnt read any, just went stright into it like a bull at a gate and bingo job done.
Anything you want to know feel free to ask.
Like to say a big thankyou and many thanks to everyone who contributed to this thread because without your help it wouldnt have kept going, and we all kept pushing to get resolve and suceeded.
Regards

[6666]

satdaveuk

Congratulations, thanks for the driver pack, I installed the driver that windows xp sp3 choose for the usbsp board.

Do you use XP ?

when you look at

Control panel
System
Hardware
Device manager
Libusb-win32 devices
Usbasp
Driver

Do you get driver version 1.2.2.0

thanks
6666

[satdaveuk]

satdaveuk

Congratulations, thanks for the driver pack, I installed the driver that windows xp sp3 choose for the usbsp board.

Do you use XP ?

when you look at

Control panel
System
Hardware
Device manager
Libusb-win32 devices
Usbasp
Driver

Do you get driver version 1.2.2.0

thanks
6666

The driver version I used in the pk was 1.240 the latest, which worked on my windows XP also 7
If you get any problems just go back into devise manager and click tab update driver software, then go into the pk and click on 1.240.
I cant comment on the older driver because never tryed it.

Regards

[6666]

satdaveuk

which usbasb programmer did you end up ordering ?

[satdaveuk]

Hi 6666
The very same one that Silver dog posted on here.
That hex file thats on the Pulsedtektor site which I posted here has flaws in it, Firstly it doesnt set the low fuse to external crystal, but thats not a problem as its easy to do.
Secondly which conserns me is it fires up saying minipulse3 then version 1.05, then goes to saying starting metal detector, then thats it just stuck on saying starting metal detector, so need to work on that one.
Tryed other hex files which I obtained over a period of time and they work but not as the mnipulse 3 was intended.
On all hex files the range control and the pulse width pots dont seem to make any difference which all directly relate back to the ATmega8 chip.
The hex files programed by a chap called Sisco seem to work but no graphic images on LCD, maybe that was intended, its diffecult at times due to alots in german, so often trying to read between the lines.
The problem is all the threads on this project dryed up a couple of years ago infact at least that, we are talking last software update 2006, so seem to be working alone, theres things about this design that dont add up, just wish there were other constructers involved to discuss it with.
I cant possibly give up on this project due to time and money already spent, and the factor that I dont like to abandon a project just cause it doesnt work, In my opinion everything is workable.
The plus side to all these problems are its been one hell of a learning curve for me far as atmega chips and how they tick, considering I new bugger all about them a month ago, and also this is my very first ever pulse induction project, which thought was going to be easy peazy , so jumped in at the deep end so to speak by going *** backwards.

Regards

[6666]

satdaveuk
thanks for the info on the flaws in the hex file
we three have the same usbasp programmer from same vender
the slow boat from china arrived today with new AT8 chips
but my usbasp programmer is acting up so it needs some work.

So close

cheers

[6666]

Today I had some success I think
as the USBasp programmer was giving trouble I went back to the Bascom simple programmer and Burn O mat.

I burned the flash, it worked according to the session printout.

When I went to burn the fuses, the values that tepco said were on the website
would not burn into the AT8, it just failed, the AT8 refused to accept them
they were L3D , HF9

So tried the values I got from the online fuse calculator and the AT8 accepted them
there were L3D , HD9.

And I used all the power pins, Gnd 8/22, Vcc 7/20/21

I have not tested the chip yet, will try it later, it may not work,
but here are the Avrdude session files, just for interest,


FLASH

C:\WinAVR-20100110\bin\avrdude.exe -C C:\WinAVR-20100110\bin\avrdude.conf -p m8 -P lpt1 -c bascom -U flash:w:C:\Mini Pulse3\MINIPULS3PART1\M8HEX\MINIPULS_3.HEX:a

avrdude.exe: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude.exe: Device signature = 0x1e9307
avrdude.exe: NOTE: FLASH memory has been specified, an erase cycle will be performed
To disable this feature, specify the -D option.
avrdude.exe: erasing chip
avrdude.exe: reading input file "C:\Mini Pulse3\MINIPULS3PART1\M8HEX\MINIPULS_3.HEX"
avrdude.exe: input file C:\Mini Pulse3\MINIPULS3PART1\M8HEX\MINIPULS_3.HEX auto detected as Intel Hex
avrdude.exe: writing flash (2664 bytes):

Writing | ################################################## | 100% 1.05s

avrdude.exe: 2664 bytes of flash written
avrdude.exe: verifying flash memory against C:\Mini Pulse3\MINIPULS3PART1\M8HEX\MINIPULS_3.HEX:
avrdude.exe: load data flash data from input file C:\Mini Pulse3\MINIPULS3PART1\M8HEX\MINIPULS_3.HEX:
avrdude.exe: input file C:\Mini Pulse3\MINIPULS3PART1\M8HEX\MINIPULS_3.HEX auto detected as Intel Hex
avrdude.exe: input file C:\Mini Pulse3\MINIPULS3PART1\M8HEX\MINIPULS_3.HEX contains 2664 bytes
avrdude.exe: reading on-chip flash data:

Reading | ################################################## | 100% 0.86s

avrdude.exe: verifying ...
avrdude.exe: 2664 bytes of flash verified

avrdude.exe done. Thank you.



FUSES

C:\WinAVR-20100110\bin\avrdude.exe -C C:\WinAVR-20100110\bin\avrdude.conf -p m8 -P lpt1 -c bascom -u -U hfuse:w:0xD9:m -U lfuse:w:0x3D:m

avrdude.exe: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude.exe: Device signature = 0x1e9307
avrdude.exe: reading input file "0xD9"
avrdude.exe: writing hfuse (1 bytes):

Writing | ################################################## | 100% 0.00s

avrdude.exe: 1 bytes of hfuse written
avrdude.exe: verifying hfuse memory against 0xD9:
avrdude.exe: load data hfuse data from input file 0xD9:
avrdude.exe: input file 0xD9 contains 1 bytes
avrdude.exe: reading on-chip hfuse data:

Reading | ################################################## | 100% 0.02s

avrdude.exe: verifying ...
avrdude.exe: 1 bytes of hfuse verified
avrdude.exe: reading input file "0x3D"
avrdude.exe: writing lfuse (1 bytes):

Writing | ################################################## | 100% 0.00s

avrdude.exe: 1 bytes of lfuse written
avrdude.exe: verifying lfuse memory against 0x3D:
avrdude.exe: load data lfuse data from input file 0x3D:
avrdude.exe: input file 0x3D contains 1 bytes
avrdude.exe: reading on-chip lfuse data:

Reading | ################################################## | 100% 0.00s

avrdude.exe: verifying ...
avrdude.exe: 1 bytes of lfuse verified

avrdude.exe done. Thank you.

[satdaveuk]

I jumped the gun saying there was a problem with the hex file as infact there was none.
The reason why I wasnt getting past the starting detector message on the lcd display is because if you look at the main pcb circuit it needs a pulse from Q2 to kick start it up.
The pulse was missing because by mistake had fitted a directional diode in placement D4 .
Worth keeping in mind with both D3 and D4 both suppressor diodes ending in OCA, which are bi-directional if ending in OA there directional.
everythings working as should be now with test coil.

Two points to remember after loading hex all will work fine with a standalone board.
You can wire LCD module as per second sch posted previously.
By doing that on first switch on you will get: Mini puls 3, followed by: version 1.05 Followed by TESTING.... Followed by Starting Detector.
After that you wont get anything else because its waiting for a pulse on PBO of the ATMega 8 from Q2 to start it up.
If your going to build this project you will need to set the lower fuse to external crystal up to 16mhz, which is so easy peezy using Extream.
Go to fuses option looking at lower fuse setting will be OXD9 and OXEI by default which means its set to a 1MHZ internal crystal.
Now all you have to do is change the OXE1 to OXEF, which means external crystal up to 16MHZ then just below it is a write box, tick it and write, job is done everything else stays at default as from factory.
You will find now that the messages on LCD display flick from one message to the next alot quicker and more stable which is to be exspected due to the external 8mhz crystal has now come into play.
Personaly wouldnt dream of using overdude to flash chips, its to long and drawn out and many mistakes can easily be made, when all you really want is flashing a chip and changing just one fuse.
The whole process takes me less than one minute from opening the program to finalizing.
Hope this helps

Regards

[Tepco]

Save your Atmegas and programmers for something far more interesting.

What you think about device below? Yet another display sized VDI for detectors? Well, no. Complete display sized VLF detector with VDI, histogram, static or motion mode, auto gnd. compensation, variable frequency (4-16kHz in 10Hz steps), selectable audio mode, filter response, exclusion zone discrimination, 4 user selectable profiles ,coil nulling mode and more. Built using only 3 commonly available chips and handful of components, using internal Atmega ADC and software processing, from:

http://radiodom.ucoz.ru/index/selektivnyj_metallodetektor_quot_krot_m_quot/0-14

Only thing I can claim so far is, software works and this thing can actually detect metal, no performance tests yet. I modified electronics slightly, so far only software is intact, will made PCB for it (non-SMD) when prototype phase is over and test it with various coils and conditions (pinpointer usage included). Interesting project. Maybe not the best machine, but simplest one to build so far. Work in progress, will be continued in another topic.

(More advanced version of this is “Quasar” already mentioned on this forum, but more complex.)

[Qiaozhi]

Save your Atmegas and programmers for something far more interesting.

What you think about device below? Yet another display sized VDI for detectors? Well, no. Complete display sized VLF detector with VDI, histogram, static or motion mode, auto gnd. compensation, variable frequency (4-16kHz in 10Hz steps), selectable audio mode, filter response, exclusion zone discrimination, 4 user selectable profiles ,coil nulling mode and more. Built using only 3 commonly available chips and handful of components, using internal Atmega ADC and software processing, from:

http://radiodom.ucoz.ru/index/selektivnyj_metallodetektor_quot_krot_m_quot/0-14

Only thing I can claim so far is, software works and this thing can actually detect metal, no performance tests yet. I modified electronics slightly, so far only software is intact, will made PCB for it (non-SMD) when prototype phase is over and test it with various coils and conditions (pinpointer usage included). Interesting project. Maybe not the best machine, but simplest one to build so far. Work in progress, will be continued in another topic.

(More advanced version of this is “Quasar” already mentioned on this forum, but more complex.)

Very interesting design that uses a low-power audio amp as the transmitter.

Here's a translation from the Russian, with parts I couldn't decypher in RED.
I'm not sure what he means in the firmware history section by - CHRIST AROSE!, and "to look out for lovers in the night".


To begin, let's look at the description of modes-menu "
Begin by setting up a "virtual zero" reception path. Disconnect the coil, put the jumper with the land and seek С0 zero (actually the voltage at about + 23 foot controller 2.5 in) from the receiving tract, this procedure do in service mode "NL" (prošivkahof 7 and higher versions) by pressing the "+", remember the value "0". In earlier versions of customization make resistors R1/R2. As soon as we can customize and "correct" readings of voltage, like a sort of "-".
Continue to insist, for ease of tuning coil rekominduet′sâ R5 and R17 set to maximum values of resistances. Then pick them up to receive parameters sensitivity-stability.
If you need to configure the receiving sensor resonance RX (for non-resonant-without condenser С0, skip) and transferring TX coils into a resonance or near resonance condition. To do this, connect the transmitter coil TX, go to service mode (long pressing the "Director") and the buttons "+" and "-" find the resonant frequency of the maximum voltage at the coil TX. Amplitude voltage value should be approximately 30-60 b. Remember this frequency-f. Next raznosim (attached) receiving and transmitting coil (replacing a capacitor C1 at 10-50 Ohm resistor) and the same buttons find the resonant frequency receiving loop for maximum signal amplitude at the output it D2, 6 may differ from F in the + -20% farther you will leave the frequency resonance TX so more stable look coil.(It is recommended to configure the host RX ~ 10%: i.e. If you configure RX TX -10 kHz to 9 kHz.) I recall frequency F can range from 4 to 16 kHz. The recommended frequency of 7-12 kHz.
Find resonant frequency, you can change the number of coils or windings С0 capacitors and capacitance value C1. Next, fit into the shell and coils are compensating windings CX (if the movement winding coil DD) minimum voltage to 6 printed circuits D2 (you can do this with a voltmeter "Mole" in the X, Y-firmware 5 and above). For normal operation, it should not exceed 1B! But in general you should endeavour to obtain the lowest possible value. Expose our previously found frequency F or close.
Next, configure phase: when the copper coins of Y and x must increase, when the black metal (iron) Y-x is decreasing. If not, the "0" (discrete increments 45 Gr.) and trimpot resistor R16 (smoothly) change phase has yet to achieve the correct reaction to metals. If this procedure fails to pick up the container, C1 or modify С0 frequency transmitter (not necessarily to TX worked at the resonance frequency, but also very far away from it is also not recommended somewhere but not further +-200gc). Then, you need to set a precise phase for the correct measurement of the PTM postroečnym resistor R16. Configure phase better on ferrite-X minus Y or a b +. You can even set up a stage for the "standard" in the mode of dx, dy-for example on 5 USSR CPC +55 about PTM value. PTM for ferrite in the -85. -90.0 ~ Foil, pure copper +80-+90, all hardware in the negative. If possible, adjust the trimpot mixing circuit (s). There is a subtlety, after changing the frequency + -30 Hz. find a place where the least change (jump) values x and y!
Sensitivity settings: Resistor R17 is current coil TX, R0 can endanger ... The more current the more sensitive, as to the objectives and unfortunately and to the ground. Recommended current to a resonant system 30-60 Ma non tuned 60-90 Ma. To estimate current in TX can be the difference between consumption device with coil and without, then for a resonant system 10-20 ma non tuned 20-30-depends on the quality of the path. Resistor R5 is the sensitivity of the device in less than RX channel's value, the higher the sensitivity. Recommended values of resistors for MRI System 1 k not resonant 100om. Configure the device on the sensitivity you want to on the threshold (sensitivity) 1.2 no false positives (of course, assuming no interference). For a great sensitivity on air Chase, 20-25 cm. most!, for sensor diameter ~ 20 cm. Otherwise there may be problems with the thermal stability of the system ...
Resistor R7 input voltage readings are collected by the battery. (Convenient 6 batteries type pal'chyk or minipal′čik the United States consistently). The 7-th and higher prošivkah this can be done programmatically-Up mode. R13 resistor adjusts the contrast of the indicator and its nominal wakes up depends on the type of indicator.
Note: the mode buttons "+" and "-" change the frequency to TX with ~ 10 Hz. Button "0" changes the phase period of 45°.

A new tab under the dip chips. Board of healthy and proverenna! For better performance, the Board should keep the top layer of foil and connect it in several (marked blue) pixels. When mounting the led watch description, conclusions of the nutrition happens differently!
Products: you can replace MCP601, K140UD1408, LF357 NE5534, LM308. You can replace the ATMEGA8A microcontroller ATMEGA8-16PI


Simply programmed microcontroller, this can be read here


Fuse program:



Videos about Mole and VDI "


Video "Mole and the brick" "static" cut ground mode "G"
Available the new firmware without previous restrictions version 0.1


Who wants to listen to the sounds of the "Mole" or other CBMS on FM radio you can assemble a simple schematic.



Configure the desired frequency-pushing, squeezing the turns of the coil L1
R4 we select for lack of distortion.


New firmware 0.2 with two dynamic modes and vector cutting soil ...Will periodically set the primer button "0", first at the meter off the ground briefly "0", then search for a long time "away" value 0 coeff. of soil must be negative ( -10 -70)


15.09.2011
1. changes to the schema of the deleted one resistor (R4) and changed some values (red)
2. Modified Board in view of the new schema.
3. new firmware 0.3 for testing.
Increased performance measurements at 20%, changed the algorithm of ground in "D" and "G", now it is possible to measure the voltage above 10B.
New firmware and change in concept allows not resonant RX, but will have to sacrifice 20% range on ~ ...


15.10.2011
1. new firmware 0.4 for testing and no version 0.4 and + firmware 0.4 in (for transistor TX)
Added a fifth button for the mode of pinpointrera, now it is possible to reduce the frequency of up to 4 kHz, TX increased sensitivity through (only 0.4).


15.11.2011
1. new firmware 0.5 for testing.
2. mode (X, Y):
a) added, the function of voltage measurement in mV voltage unbalance of the coil.
b) added the ability to automatically adjust the minimum value of voltage unbalance coil-long pressing "0" (to use with understanding!).
with output values changed) form the initial phase are now configured: 0-1, 0gr-45gr, 2-90 g, 3-135gr
3. Change the initial state of 15 feet at the time of initialization microcontroller + 5V for parallel scheme on TX start, had to work with a consistent start is not affected.


15.12.2011
1. new firmware 0.6 for testing. 0.6 A-for the old-ATmega8-XXX (or anyone with a higher voltage without coil)
2. Change the splash screen during initialization, now shows # prfilâ (May 4), the frequency of the coil and the amplitude razbolansa.
3. Added function of selecting "profile". Number profile vyberaet′sâ mode (dX, dy) takes effect after the overload. In the profile are all selectable settings, frequency, phase, filtration, sound scheme and all that stuff ...
4. When you click the button "0" now display real voltage razbolansa reels in MV, while zeroing feature has not changed.
5. Changed filter-mode "statics" now it wakes to work successfully and with DD coils.
6. the "D" primer is "dynamically", "G" is "psevdostatičeski" ground.


How do MD? -Very easy! -Article


Video software-0.5, coil DD20 (resonance), 12 kHz, tests, quality telephone ...



Video "Mole brick and kopeck" software-0.6, coil DD20 (resonance), 12 kHz, demonstrates the automatic "cut" soil in dynamic mode, or test your unit in the winter.)


16.01.2012 Sub-rules article in light of recent observations.


15.02.2012-corrected inaccuracies Fixed schema, R17, R18 added resistors have changed some values, add connector mcs51 (as an option), measuring and digital earth posted (sooedinâût′sâ at one point you can through a resistor or inductor). All changes are irrelevant and do not bring significant improvements! Fees altered.


23.02.2012
The "men's day" new firmware 0.7
1. now remember and thresholds for different profiles.
2. changed the type of the discriminator-first black, then colored.
3. Added a new option in the settings-"NL, Up" Button "+" Select configuration "factor to 0 (if you disabled the reel was tantamount to a selection of R1/R2)"-"settings" select the coefficient for power "(equivalent to picking the R8/R7).
The settings are remembered by the button "0".
4. Voltmeter now measures and "negative" voltage.
5. the "G" again "static" taking into account the soil, after properly configuring the primer depth comparable to air ...
6 Removed auto-configure the processor minimum voltage balance coil.
7. other minor bugfixes.
Waiting for interesting reports about firmware!


15.03.2012 new firmware 0.7 (F) for testing.
1. all the same as in 0.7, fundamentally changed the "filters"-frequency of ~ 3-11 Hz-to experiment with.


CHRIST AROSE!
15.04.2012-new ROM 0.8
1. improved "histogram". Allows you to define pieces with complex geometric form, etc.
2. highlight mode-"LED" a third parameter "SG", eco-mode for indicators at the gap, and to lookout for lovers in the night.
3. improved "Statics". Now, in this mode you can search.
4. Entered "Nuisance" serves to localize the source of interference and the determination of its level. Is activated by pressing the button "Dir". Exit-off.
5. other minor bugfixes.


15.05.2012 Firmware 0.8 G-equivalent of 0.8, but may have to adjust color phase-R16.


It is customizable firmware to work on the ground in the mode of "G".
"D", in