Help - My AW32 spoiled twice

[Sany]

I will be very appreciate if anyone could troubleshoot with me. I using this AW32 S08 MCU with 5V. The whole PCB with this AW32 was working fine. It just suddenly spoiled for no reason. And it happend twice for two of my PCB. I didn't run this MCU for more than 10 minutes everytime. Assuming there isn't any short circuit problem. No over voltage. I believe is the hardware circuit problem. I double checked all the pins and voltage everything look fine. Any other reason that can cause a MCU spoil? My PCB has these loads: GLCD, max232, max485, MC7805T, 4xEEPROM 95256, DS1305, buzzer, 16MHz crystal, relayand a EM card reader module. Thank you in advance.

[bigmac]

Hello Sany,

You do not mention the nature and extent of the failure.  The following are some things that come to mind, that you would need to check.

You will need to have some idea of the anticipated current draw for the board.  Do the faulty boards significantly exceed this value?  Do any of the devices on the board run unexpectedly hot to touch?

Do your pre-fault boards exhibit any of these symptoms?  If so, I would tend to progressively build a blank board (by hand), and measure the current draw before and after each IC is installed.  Perhaps I would start with the MCU, and then add the peripheral devices one by one.

Do you still have any communications with the faulty boards, via BDM module?  If so, you could test individual I/O lines, via debug, for remaining functionality.  You would particularly want to check that output lines are not "stuck at" either 0 or 1, and that you are achieving the full logic swing.

I have great difficulty in reading your "schematic" - it is not really a schematic, but a type of net list, since the interconnections between many of the major devices are not explicitly shown.  And the size reduction from an original A3 sheet size to A4 does not aid clarity.  So it is not obvious which MCU I/O pins are directly exposed to the "outside world".  It is also not clear whether you use any external pull-up resistors on unused, or exposed input pins.  This would possibly lessen the chance of damage caused by static discharge.

Regards,

Mac

[Sany]

ty for all the troubleshootings. I think I must have done something seriously damaged the MCU. Because I did some solderings before that. And some component onboard testings which probably over voltage and damged it.

[Jaux]

Is there any way you can send me more information on the physical layout of your board?

A good start will be if you can send me two high definition pics of the two sides of the board.

[Sany]

These links have all the pictures.

http://www25.brinkster.com/sanylcs/pcbback.html

http://www25.brinkster.com/sanylcs/pcbfrnt.html

http://www25.brinkster.com/sanylcs/sch.html

ty for reply.

[Jaux]

Some quick remarks:

1. The MAX232A normally uses 0.1uF caps for the charge pump ( MAX232 uses 1.0 uF ) 10uF is way out of the spec.

2. LP1 and LP2 I presume are "normal" LED's - with the bias resistors of 330E you would sink 10mA thru PTE0 and PTE1. Make sure you configure these as 'high output drive' in register PTEDS. This is a bit high for my liking - and I do not drive LED's directly from MCU pins

3. The physical layout look ok - but the schematic is difficult to read. Also I tend to be rather conservative with pin loadings and fan-outs - I have seen a couple of 'adventurous'
moves in the design - I would advise to check all pin loadings against the spec and make sure the chip can handle it.

Freescale MCU's are tough little suckers, and you must have done something serious to blow it. I see you have several unbuffered output pins directly from the MCU - what goes in there?