'Power Brown Outs & meeting massacre'
I had the same problem, but fortunately my product was in the
early prototype stage at the time. The PIC's main CPU and
oscillator sections start up at a lower voltage than does the
memory, thus locking the chip up upon brown-out or power-up
when the supply doesn't ramp up quick enough.
The watchdog timer will only protect you against software
lock-up, not against the CPU core itself going off into the
twilight zone, which is what it sounds like you have here.
Without getting into too much detail on the various power-up
requirement specs, the internal power-up timer (PWRT) and
oscillator start-up timer (OST) specs, you can simply enable
the PWRT if your power supply will always ramp up in less than
28ms AND the supply shuts off at low voltage instead of going
into un-regulation mode.
If your power supply can't do that, then simply toss a Motorola
MC34064 onto the /MCLR (active low RESET) pin, along with a cap
and resistor or two if you like. This is a little 3 pin
undervoltage sensing IC, and will hold the '74 in reset unless
the voltage is above 4.6 V. It comes in many different
flavours like wider temperature ranges, different cutoff
voltage like 4.3 or 2.7 volts, TO-92 or SO-8 packages, etc.
Similiar devices are available from many other manufacturers.
Oh, and they cost way less than $1 U.S. each in low quantities.
The microchip databook shows some other cheap alternative
brown-out protection circuits.
Of course, Microchip has realized this deficiency in the
standard '74 as well, and have introduced the '74A as a result.
As far as I can tell (note the subtle disclaimer) the only
difference is the addition of a brown-out reset circuit (BOR).
When enabled via the BODEN bit (love those acronyms) the BOR
will hold the device in reset once VDD falls below about 4V,
and will keep it in reset until VDD has remained above 4V for
more than 28 to 132ms (72ms nominal).
For the BOR to kick in and reset the chip, VDD must remain
below 4V for at least 100usec. I assume that Microchip in
their wisdom has made this delay less than the time is takes
the chip lock up, though I could find no indication or
guarantee of this in the databook.
So basically you have three choices: Improve your power
supply, add in something like a MC34064 (happy rework & PCB
changes), or go for the 74A's and save your '74 stock for a
future project. Now get in there and call another meeting to
say you've got a fix. Ok, maybe test it first.
______________________________ Reply Separator _________________________________
Subject: Power Brown Outs
Author: pic microcontroller discussion list <MITVMA.MIT.EDU> at PICLIST
Date: 4/3/96 7:34 PM
Here's a brain buster.
We have a stand alone CO gas monitor (safety equipment) that uses a
If the power supply is momentarily shorted (milliseconds), the PIC does
not recover, but locks up in random modes.
We have the Watchdog timer ON and CLRWDT is only given one time in the
main program loop.
Any ideas? I was massacred today in an engineering meeting as this
instrument was on the verge of its first major shipment.
Regards, James Musselman, President
Radix/Cobalt Instruments, Inc.
PO Box 897
Clovis, CA 93613 USA
Check out my home page http://rdx.com
More... (looser matching)
- Last day of these posts
- In 1996
, 1997 only
- New search...