Re: [avr-chat] Strange problem with ATMega32A-PU

[Royce Pereira]

Oops, in my earlier reply please read :-

"(The 'while(1);' does *not*  cause a  reset, so WD is not started by default)."

--Royce.


On Thu, Apr 28, 2011 at 8:31 PM, Royce Pereira <address@hidden> wrote:
> Hi all,
>
> Thanks for your replies.
>
> I got the new chips, but they too showed the same symptoms, So I guess
> the original chips were good.
>
>
> With divide by 8 issue, I was not able to find any suggestion in the
> data sheet that the clock is divided by 8. There is also no fuse that
> determines this (in both M32L & M32A)
>
> The watchdog is also not enabled at reset ( See below)
>
> I finally disabled the Timer2 (my system tick) interrupt, & put a
> while(1) ; after the LCD welcome message (The 'while(1);' does cause a
> reset, so WD is not started by default).
>
> Now I get the welcome message.
>
> I've compared the interrupt vectors, Timer regisaters etc... no clue
> yet. I'll tinker around a bit more...
>
> -- Royce.
>
>
>
> On Thu, Apr 28, 2011 at 7:19 PM, Gjermund Stensrud
> <address@hidden> wrote:
>> Another thing that has changed with newer versions of the AVR chips is the
>> Watchdog, it was disabled on reset before. On newer devices it needs to be
>> disabled in the beginning of your program. That is if you have a long
>> initiation phase.
>>
>> I'm not sure about this particular chip, but on the ATmeaga1284p-PU I had
>> this problem.
>>
>> But I agree, the clock divider fuse is more likely to be the problem.
>>
>> - Gjermund
>>
>> On Thu, Apr 28, 2011 at 2:37 PM, Graham Davies <address@hidden>
>> wrote:
>>>
>>> Royce Pereira wrote:
>>>
>>>> Searched the ATMega32A-PU data sheet. No 'CDIV' found.
>>>
>>> OK, but Dave did write "I seem to recall ...", so he's just pointing you
>>> in a direction, not giving you text to search for in the data sheet.  Look
>>> in Section 8, System Clock and Clock Options.  You will find that indeed the
>>> ATmega32(L/A) parts are shipped with the clock divided by 8, just as Dave
>>> said.
>>>
>>>> Atmels migration from M32 to M32A document says nothing about any fuse
>>>> etc.
>>>
>>> That's because there's no change here.  Maybe Dave wasn't suggesting that
>>> there's a change, but rather that there's more to bringing up an ATmega than
>>> loading the program.  Perhaps you should compare the fuse settings for the
>>> parts that work and the new parts, that don't.
>>>
>>>> I am suspecting a case of bad ICs.  I'm awaiting delivery
>>>> of a couple of M32A's from another source. Will try
>>>> those & update the list.
>>>
>>> I think that would be great.  I hate it when the OP does not post the
>>> outcome.
>>>
>>> Graham.
>>>
>>>
>>>
>>> _______________________________________________
>>> AVR-chat mailing list
>>> address@hidden
>>> https://lists.nongnu.org/mailman/listinfo/avr-chat
>>
>>
>> _______________________________________________
>> AVR-chat mailing list
>> address@hidden
>> https://lists.nongnu.org/mailman/listinfo/avr-chat
>>
>>
>