Please help a newbie! Device signature = 0x000000

[BigJim71]

This is my first time playing with a microcontroller. Feel free to smack me upside the head as long as you teach me something.

FIRST: Are you using a USBtiny from a kit or from DIY? - I built the USBtiny from the adafruit kit.

SECOND: What OS and computer are you using? - I'm using an old P4 Windows XP box.

THIRD: Which programming software are you using, and what chip? - I'm using AVRDude and I'm trying to use an ATMEGA328-PU chip.

FOURTH: Did the USBtinyISP ever work (even once) or is it your first time trying to talk to a chip? - This is my first time trying to talk to the chip.\

FIFTH: Post screenshots of what is happening



SIXTH: Post clear, in-focus, not-blurry, easy to see photos of the top and bottom of the kit.





Other relevant information: I crossed some wires when I first started playing with the programmer and I fried the D2 diode directly behind the USB connector. It's been replaced and the green LED is lighting up. I ordered and installed a new programmer chip just to be safe. I checked all the resistors and diodes and they're ok. I'm not sure how to check a capacitor, but everything looks ok. I built a simple target board using these instructions: http://www.evilmadscientist.com/article ... rgetboards I've got only jumper wires, a 6-pin header, and an IC socket on the board. No resistors, caps, or anything else.

As you can see above, I'm getting a "Device signature = 0x000000" error when I try to talk to the chip. I've read that this error is often caused by an improperly set bit clock, so I tried using -B with delays ranging from 1 to 100. I tried setting the baud rate with -b 19200. I've triple checked every solder connection on the target board. Everything is connected where it should be and I have continuity from the 6-pin header on the target board to the appropriate chip pins. No shorts either. I've got 5VDC from the USBtiny powering the chip and I've connected it to the appropriate analog pins too.

I've got good connections between the PCB and the programmer 6-pin header. I checked the pins with a voltmeter and I've got about 0.1VDC from the Reset, SCK, and MISO pins to the ground pin, 5VDC from Vcc to ground, and 0VDC from MOSI to ground. I don't have an oscilloscope to check further.

I'm hoping I just did something dumb on the programming end of things and it's not the buffer chip. Thanks in advance for your help.

[adafruit_support_bill]

A short bad enough to toast a diode might have caused more widespread damage. But it would be hard to check for a damaged buffer without some sort of scope.
Let's make sure we cover the basics first. Maybe a photo of your target board would be helpful.
Is the target device powered? Is this a 'factory fresh' chip? Or does it have a pre-burned bootloader?

[BigJim71]

I'll get target board pictures when I get home from work today. The target device is being powered by the USBtiny. I've got the jumper in place. This is a brand new chip. It shouldn't have anything on it.

[BigJim71]

Red = Vcc - Pin 7 jumpered to pins 20 & 21
Black = Ground - Pin 8 jumpered to pin 22
Brown = MOSI - Pin 17
Green = Reset - Pin 1
White = SCK - Pin 19
Yellow = MISO - Pin 18

For further clarification, the blown diode didn't burn or smoke. The green LED stopped working so I started testing components. I found D2 transmitted both ways when using a diode tester.

[adafruit_support_bill]

Check your pin connections on the 328. It looks like you might be offset by 1.

[BigJim71]

I think I've got it right. The pictures above show kind of a funny angle.

[adafruit]

you may need a crystal, especially if its an arduino-bootloaded chip

[BigJim71]

Can you please explain why I would need a crystal? Neither your Barebones AVR Dev. Board nor Evil Mad Scientist's minimalist target board use one. I've mimicked your board by adding the cap across Vcc and ground and the LED to the SCK pin. For what it's worth, the LED lights for a second when I try to contact the chip, but still no good.

[BigJim71]

I've added a 16.000 MHz crystal and I'm still getting the same device signature error. Any suggestions? Is it possible to test the buffer chip without a scope?

[adafruit]

the spokepov program has a test dongle procedure which could be used

[BigJim71]

Thank you! That program is exactly what I've been looking for. Pin 1 (DataOut) won't go high during the test. The other pins pass. Is this most likely a buffer issue?

[adafruit]

could be, check your soldering and the schematic, see where the signal is lost

[BigJim71]

I got it! Neither chip is bad. The trace between the R4 jumper and Pin 4 on the 6-pin header was bad. A little piece of green wire fixed everything. Thanks for all the help and putting up with my newbie questions.

[adafruit]

great!