To continue this monologue. (Perhaps this is the bane of living on the other side of the world, I am practically troubleshooting by myself while other people is partying, sleeping or having breakfast and starting their day). And finish with a good conclusion. Because finally after 1 day plus (or 22 days depending on how to look at it!) manage to get it working.
So let's have all the glory (and frustration) in point form.==1==
This thread begin as I ask the question of how to power Arduino BT. The original idea is to use a wall wart with rated output of 3.6V under load. However upon checking the voltage using multimeter, found that it is 6.2V instead. PaulS later point out that if connected to Arduino board with proper load. The operating voltage should be 3.6V. ==2==
Decide not to take the risk (in other words chicken out). So Instead of using the wall wart, use 2AA batteries. Multimeter show around 2.8V to 3.2V. Unfortunately I use crappy battery. The fact is that even though it power up, can't program the board. Failure to program is not because of the battery or the low charge hold by the batteries. But, since the battery run out very so often it give wrong perception about the problem. Beside this, the troubleshooting process is interupted by changing batteries, pairing the device again and restart troubleshooting. This is improved by using 4x AA batteries with each 1.2V and 2100mAH. In total I get 4.8V to just bellow 5.5V due to diode voltage drop. It is still a mystery to me why 4 x 1.2V give more than 5V.WARNING : Do not use 4 x 1.5V in series. This will result to 6V which exceed Arduino BT max voltage at 5.5V==3==
This is the most important point. The reason fail to program Arduino BT. My theory is that the Bluetooth chip (Blue Giga) is hogging Arduino serial communication. Which also use to program the onboard ATMEGA328 chip. This theory is formed by reading bellow threadhttp://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1230381702/all
Reply1 - Adilson Akashi
I don´t know if this problem is the same, but I read somewhere in this forum about a sketch that sends constantly messages from arduino´s serial port, and makes it uncommunicable to receive new sketches.
Is the actual sketch a loop of this kind?
That's why there is supposed to be a short delay after pressing reset... no? That delay allows you to UPLOAD a new sketch with the bootloader. If you wait too long... you miss the window and you will have a serial conflict as seen.
If you have a different bootloader that has the delay removed... oh well.
Reply13 - halley
Connect the board to the computer. Ensure you've selected the right serial port in the IDE. Hit the Reset button on the board. Hit the Upload Sketch button on the IDE before two seconds has elapsed.
When you hit the reset button, the board starts running the bootloader program instead of your program. The bootloader program watches the serial connection for two seconds, and if there's nothing that looks like a valid program-upload process being attempted, it will start running your program.
In essence pressing the reset button before immediately click the upload button solved the failure to program issue. There is 5 second window before the Blue Giga chip start to hog the serial com. At least this is what I think!==4==
But that is not it. There is a plot twist in this story. After successfully uploading Blink sample, still there is no bloody blinking LED. As noted in my earlier post. Even though the hardware page described Arduino BT as having built in LED. It miss that one LED connected to Digital I/O pin 13. Running the Blink sample on Uno or Duemilanove will result to immediate and visible result. But this does not apply to Arduino BT. Need to connect LED to pin 13 to see the result on Arduino BT. So after doing the necessary, finally there is blinky!
It is good to ponder. Why this issue is not highlighted in the Arduino BT hardware page? So many forum post where people are describing this same issue with Arduino BT and asking for help.
Best Regards & Happy Weekends