Searching \ for 'Problem with PIC and WDT wakeup' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/microchip/devices.htm?key=pic
Search entire site for: 'Problem with PIC and WDT wakeup'.

Truncated match.
PICList Thread
'Problem with PIC and WDT wakeup'
1998\05\01@170419 by Frank Schmidberger

flavicon
face
in a datalogger application we are using:

PIC 16C715
CCS C-compiler
PCF8583 RTC
crystal oscillator at 4 MHz

To save battery lifetime the device is mostly in sleep mode. We use the WDT
to wake it up periodically. When a wakeup happens we check for a RTC alarm
(e.g. the RTC pulls down a port pin of the PIC), do something (and reset the
RTC alarm condition) and then go back to sleep mode. If there is no RTC
alarm we go to sleep immediately. This approach works great most of the
time, but some devices do very seldom stop to work. In this cases the RTC
alarm pin is down (e.g. the RTC still worked and did set the alarm
condition) but the PIC does not react on the alarm any more. There are two
ways to restart it: 1. Disconnect the power supply, 2. Touch the oscillator
input.

At the moment we have no idea whether it's a software or hardware problem.
We think that there are the following possibilities:

1. The oscillator doesn't restart upon WDT wakeup.
2. The WDT stops to work and therefor does not generate a wakeup any more
3. The program counter gets corrupted

Any ideas

Frank Schmidberger

1998\05\02@195139 by Mike Keitz

picon face
On Fri, 1 May 1998 19:32:46 +0200 Frank Schmidberger
<spam_OUTFrankSchmidbergerTakeThisOuTspamT-ONLINE.DE> writes:
>in a datalogger application we are using:
[...]
> In this cases the
>RTC
>alarm pin is down (e.g. the RTC still worked and did set the alarm
>condition) but the PIC does not react on the alarm any more. There are
>two
>ways to restart it: 1. Disconnect the power supply, 2. Touch the
>oscillator
>input.

It sounds like the oscillator is not starting.  Since touching the pins
helps, you probably need more capacitance.  It seems almost certain, but
you can be sure it's not a software problem by connecting an external
clock.  You could also try pulsing the MCLR pin low to guarantee a
wakeup.  I don't think this disturbs the oscillator if it's already
trying to run.

It's good to stress-test the oscillator by varying the supply voltage.
Use the actual circuit board and crystal, and a test PIC programmed to
just toggle an output pin to show that it's running, then sleep and WDT
wake up.  Replace the usual supply with an adjustable voltage.  Vary the
supply voltage and make sure the oscillator starts and runs at all
reasonable voltages.


_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...