My wife broke my megadrum, please help

[Ruiner46]

Did you measure voltage on the power supply?

Umm, no, but it worked fine for two weeks...

How did you check it?

Oscilloscope at my work.

All keyboard related pins always stay high unless buttons are pressed.

Actually, I meant to say the lcd display pins, not keyboard pins.

[Ruiner46]

Does anyone have any info or experience with programming the bootloader while in-circuit on the V2.8 PCB? Can I load the bootloader by using the ICSP_Atmega pins somehow?

[dmitri]

Does anyone have any info or experience with programming the bootloader while in-circuit on the V2.8 PCB? Can I load the bootloader by using the ICSP_Atmega pins somehow?

It is described in http://www.megadrum.info/content/programming-mcus but if you had bootloader programmed, Atmega won't loose it no matter what keys you press or what sysex you send to it.

[Ruiner46]

Update:
Signs of life! I made a cable and loaded the bootloader back onto the Atmega using ponyprog. Now it powers up to the bootloader. When I push down and it displays "Send Sysex now", the MIDI light comes on right away. However, when I try and send sysex firmware using MIDI-Ox, MIDI-Ox will hang forever and apparently not send. The MIDI LED never turns off or flickers after I push down on megadrum.

I also tried using MCT, but I get an error everytime I try to update the firmware.

Any ideas?

[dmitri]

Update:
Signs of life! I made a cable and loaded the bootloader back onto the Atmega using ponyprog. Now it powers up to the bootloader. When I push down and it displays "Send Sysex now", the MIDI light comes on right away. However, when I try and send sysex firmware using MIDI-Ox, MIDI-Ox will hang forever and apparently not send. The MIDI LED never turns off or flickers after I push down on megadrum.

I also tried using MCT, but I get an error everytime I try to update the firmware.

Any ideas?

The ideas:
1. Not correct MIDI ports selected in MIDI-OX or MCT
2. Incorrect frequency selected in the bootloader.
3. Broken connectivity or shorts between Atmega (pins 14/15) and PIC18F2550 (pins 17/18)
4. Something broken with drivers installtion in Windows.
5. Partially broken Atmega or/and PIC18F2550

Since you can directly program Atmega, program it with MegaDrum firmware via ISP using ponyprog (.hex file, not .syx file!). Once programmed test MIDI communication using MIDI-OX or MCT. Do you get two way MIDI communication with MegaDrum firmware?

[Ruiner46]

Dmitri,
Thanks so much for your help. I tried multiple times with MIDI-Ox, and suddenly once it worked correctly!! I think the key was that I wasn't powering up holding the left button maybe? It was still going to the bootloader screen though. I saw a post where you said that to allow some troubleshooting with the bootloader, the LED will flash while you hold the left button. So, I powered up holding the left button and watched it flash for a while. Then I pressed down and sent the firmware erase sysex and the LED came on, then went out and it said erased. Then I sent the firmware and it worked wonderfully. Thanks for all the help everyone, I'm so happy to have megadrum back again!

On another note, I was talking to some friends of mine who have experience with these processors. My friend said that these processors have a history of erasing their internal memory (not sure if flash or EEPROM) in a brown-out condition. I think my power supply may have dipped or something and somehow the Atmega started erasing memory until it ran out of power. I'm going to ditch my hokey power supply and go with something a little more robust to prevent these issues in the future.

[dmitri]

Dmitri,
Thanks so much for your help. I tried multiple times with MIDI-Ox, and suddenly once it worked correctly!! I think the key was that I wasn't powering up holding the left button maybe? It was still going to the bootloader screen though. I saw a post where you said that to allow some troubleshooting with the bootloader, the LED will flash while you hold the left button. So, I powered up holding the left button and watched it flash for a while. Then I pressed down and sent the firmware erase sysex and the LED came on, then went out and it said erased. Then I sent the firmware and it worked wonderfully. Thanks for all the help everyone, I'm so happy to have megadrum back again!

On another note, I was talking to some friends of mine who have experience with these processors. My friend said that these processors have a history of erasing their internal memory (not sure if flash or EEPROM) in a brown-out condition. I think my power supply may have dipped or something and somehow the Atmega started erasing memory until it ran out of power. I'm going to ditch my hokey power supply and go with something a little more robust to prevent these issues in the future.

I abuse my Atmegas and never loose the bootloader. The only way the bootloader could disappear is if the fuses for Atmega are programmed incorrectly.

[Ruiner46]

Well, my wife can break anything lol. Synthex programmed my Atmega, so I can't say for sure on the fuses. However, I definitely lost the bootloader since reprogramming it clearly brought it back to life.

[Ken Forgettable]

I abuse my Atmegas and never loose the bootloader. The only way the bootloader could disappear is if the fuses for Atmega are programmed incorrectly.

As people will be plugging and unplugging the USB cable at will - is there a reason why you don't use the BODEN Fuse?

644 data sheet: 'An EEPROM data corruption can be caused by two situations when the voltage is too low. First, a regular write sequence to the EEPROM requires a minimum voltage to operate correctly. Secondly,
the CPU itself can execute instructions incorrectly, if the supply voltage is too low.'

Also why have JTAGEN and OCDEN enabled in 'What you need to know to successfully build MegaDrum.

644 data sheet: 'When the JTAGEN Fuse is unprogrammed, these four TAP pins are normal port pins, and the TAP controller is in reset. When programmed, the input TAP signals are internally pulled high'

644 data sheet: '4. Never ship a product with the OCDEN Fuse programmed regardless of the setting of Lock bits
and JTAGEN Fuse. A programmed OCDEN Fuse enables some parts of the clock system to
be running in all sleep modes. This may increase the power consumption.'

[dmitri]

EEPROM corruption has nothing to do with the bootloader in FLASH memory.

OCDEN and JTAGEN are software disabled by firmware.