[wxsig] Crashing and random characters

[William Beals]

Guy (and all):

Let me digest this for a bit.  While I don't know of it, there is a chance I
do have something on the edge from a timing or voltage level standpoint that
is going over the edge on the newer chips.

will

-----Original Message-----
From: wxsig-bounces at lists.tapr.org [mailto:wxsig-bounces at lists.tapr.org] On
Behalf Of John Bennett
Sent: Monday, July 21, 2008 7:48 PM
To: TAPR Weather Station SIG Mailing List
Subject: Re: [wxsig] Crashing and random characters

John Koster wrote:
> So am I drawing the right conclusion?  The processor chip from the T238+
> kit was bad?
> 
> On Mon, 21 Jul 2008, Guy Story KC5GOI wrote:
> 
>> I wanted to let the list know what I have found out and it took some 
>> patience.  I took advantage of the fact that the only difference between 
>> the older kits I have been using and the newer kit was the circuit 
>> board.  It reduced the variables.  I have had faith in the code that has 
>> been released so I started there.  Since the microprocessors are the 
>> same throughout the projects life span, I took a microprocessor that has 
>> been running for multiple years and flashed it to the Wills most current 
>> version and placed that on an older board.  One week later I still had 
>> the heart beat.  That ruled out sensors, cabling, code, older board (I 
>> could duplicate the crashing problem on the newer processor on the older 
>> board) and pointed to the processor.
>>
>> I then swapped the processor to the current circuit boards, everything 
>> else the same.  Ten days later I still have the heart beat and readings. 
>>
>> I am not sure how soon I can get it back in service but W5NGU-3 will 
>> soon start to give wx data from 300ft again. 
>>
>> There is one fatality, the baro sensor is not reverse polarity 
>> protected.  I am sure that is already known but just encase someone sees 
>> this post and does not know it, I am putting it in.
>>
>> 73
>>
>> Guy
>>
>> _______________________________________________
>> wxsig mailing list
>> wxsig at lists.tapr.org
>> https://lists.tapr.org/cgi-bin/mailman/listinfo/wxsig
>>
> 
Wow. A bad MCU is not unheard of, but not in the quantity this would 
indicate. I wonder if Motorola, in their infinite wisdom, changed some 
of the timing parameters on writing to the EEPROM in a revision of the 
chip. Might be worth looking into. I would sooner think that than a bad 
bunch of MCUs.

John
n4xi


_______________________________________________
wxsig mailing list
wxsig at lists.tapr.org
https://lists.tapr.org/cgi-bin/mailman/listinfo/wxsig