Hi,
I finally found an old PC with RS232 port and connected my doublechecked arduino on breadboard via ponyprog variant serial programmer.
As the old pc is not able to run arduino IDE I am using avrdude for the job and getting no response from the chip. I went as far as conecting LED's to each ICSP pin and it seems the chip is not answering back.
The LED's flash and seem to indicate this: 1) Reset disconnects from ground 2)SCK & MOSI blink 3) Reset grounds
avrdude reads: device signature 0x840100 and fails.
with -v -v -v parametres i see the avrdude is sending the same data over and over to the atmega but has 0 as a response. Upon connecting the MISO to +5 the response changes to FF, so the Serial programmer seems to work ok.
So as far as i can judge, there is no response from the chip back to PC.
Does anyone know what voltage is supposed to go onto these pins (using +5)? Could the chip be dead and how can i test that?
It is chip without bootloader, fresh from a shop. it's connected with 16MHz crystal - could it need to be bootloaded without the crystal first?
That old computer has probably higher (more official) RS-232 voltage levels than for example modern laptops.
I don't know if the MISO (0V...5V) is enough.
Can you try such a simple programmer circuit for the parallel port ? Some of us have made that, just in case other things fail.
tyglik:
It is chip without bootloader, fresh from a shop. it's connected with 16MHz crystal - could it need to be bootloaded without the crystal first?
If it is fresh from shop, then crystal is not neccesary, but I would connect it anyway because probably you are burning the bootloader with 16Mhz and after bootloading you are going to need it.
Is there a pull-up resistor on reset pin?
Disconnect pin 21 from Vcc, just leave it alone.
I don't think this is a voltage problem, serial programmer Pony prog was made to do this stuff with old RS232 port. (I've used similar programmer but for PIC microcontrollers and works fine).
Setup looks OK, I think. Zener diodes should cut off signal from RS232 to TTL level. Crystal is not problem as mart256 mentioned. It is not used by default. Maybe stupid question: Is your chip powered? Provide a real photo of your setup.
Hi, first of all, thanks guys for the replies.
The chip is powered by +5 and GND directly from the pc cables (not on the picture though, you can see the two pins sticking out on the bottom right corner, where the power connector goes)
I also disconnected the pin 21 now, but don't have time to try it this morning (do you think it will make any difference?).
As for paralel programmer, i do not have the connector handy.
The real photo is attached.
As Peter_n wrote remove LEDs it is just indication and circuit will be more simple. Do you have Vcc and GND on the other side rails (missing on photo)? Where are Rx&Tx connected (pins 2 &3)? - need not. Wires from programmer? Provide another photo.
Try minimal circuit. Double check wiring. You've used stranded wires. Are they well plugged?
The leds i added only after avrdude kept failing.
The reset IS pulled high with 10k resistor (it's in the photo & also in the schematics).
Indeed i have power in the other two rails (the reset led would not work if i didn't (for one).
I'll add the decoupling capacitors, thanks for the tip.
RxTx pins are connected to another circuit on that breadboard (FTDI serial programmer) - good point here i better disconnect it.
Paralel port is unfortunately out of question.
Will try to bootload it again with all these changes and report back.
thanks & regards.
So i removed the LED's, disconnected ARef from +5 (AVcc), added decoupling capacitors, removed the tx, rx connections to FTDI and tested (tried to upload bootloader). The same error, the same output.
I further removed the crystal and the two capacitors (hoping to have it run on it's internal clock tested again - again, the same error, the chip does not return data back to the avrdude.
Then i despaired, then i connected the crystal back again and it's two capacitors and went to try. Didn't work.
Then i connected an LED to the power source and found out it's not securely in, replaced the connector pins and just as an afterthought changed the parametre of avrdude -c to "-c ponyser" from previously modified version of it (having the ICSP programmer without what parametre to use, thinking the negative reset must be wrong in the ponyser setting (and it didn't work anyway originally) and ... it works!
Apologies to all and many thanks for attempts to help!
Thank you
P.S.:
Having it working, I did some testing with the LED on pin19 of the chip (or also called D13 or SCK) and specifically the resistor. With up to 470R the avrdude would fail, then I skipped straight to 1K and that worked again.
So for noobs like myself out there, remove the LED or increase the resistor and get proper connectors!
Otherwise the schematics mentioned above work just fine.
lol
