To Brownout or not to brownout? That is the question!

Will Reeve · Joined: 30 Oct 2003 Posts: 209 Location: Norfolk, England

PIC18LF452
4.5v, brownout enabled, sleeping, current consumption measured 38uA
4.5v, brownout disabled, sleeping, current consumption measured 2uA

measured using an uncalibrated Fluke 29 but won't me a million miles off.

I am in two minds to disable brownout for the production version, it will eek several more years out of the batteries.

I've never had a 'real' need for brownout as I don't do any housekeeping if Vcc drops, but it's seems safer as it keeps the device in reset if the Vcc falls below a set threshold.

I don't know enough about the chemistry of alkaline batteries to guess at the differences in the real world as self discharge may be more relevant?

Anyone make battery products like to comment? If the PIC get's it's knickers in a twist the first thing users will do is remove the re-insert the batteries anyway?

Keep well,

Will

Mark · Joined: 07 Sep 2003 Posts: 2838 Location: Atlanta, GA

Since you are running off batteries I wouldn't worry about it. The supply isn't going to dip unless the batteries are running low and you can signal to the user before they get too low.

SherpaDoug · Joined: 07 Sep 2003 Posts: 1640 Location: Cape Cod Mass USA

Is there any heavy load that might drag down the batteries, but then recover?

Something that comes to mind is an electronic door lock on a hotel room. It waits for hours for someone to swipe a card. When the card is swiped it processes the data to see if the code is correct. THEN if the code is correct it fires the solinoid to open the lock. The processing is all done by the time the heavy drain of the solinoid drags the battery down. When the solinoid is done the battery recovers. Here brownout might give better battery life.
_________________
The search for better is endless. Instead simply find very good and get the job done.

Will Reeve · Joined: 30 Oct 2003 Posts: 209 Location: Norfolk, England

There is a high(ish) drain under PIC control (40mA) but I'm using the low voltage detect interrupt during this to spot a low battery. It's more if the PIC gets locked in some way due to a low voltage, when swapping batteries maybe (there is some capacitance in the circuit, and with such a small current drain in sleep this will happily keep the PIC asleep for several seconds with no batteries). Admittedly I've never seen anything horrible happen during testing.

Keep well,

Will

theMagni · Joined: 21 May 2004 Posts: 48 Location: Victoria, BC

I work on really low-power boards too. I've got the brownout disabled, but the MCLR is attached to a low-voltage detector chip. I'm not permitted to get into details, but you should be able to find a chip that sets a line low when a voltage is too low. That lets the PIC die quietly and nicely should the battery level drop too low. When / if the power comes back up, the PIC re-initializes itself and grabs the most important information from the EEPROM.

The standby current on the board I'm working on now is ~5uA with everything attached. It goes up to 40-50 mA during the GPS acquisition times.
_________________
In the 90's, I promised myself I'd never use a so-called "smiley icon". I hate what I've become. ;)

valemike · Guest

I used the MAX666 voltage regulator with low battery detection. The Vout and Vlobatt threshold are both adjustable.

In the case of 9V alkaline batteries, the voltage gradually decreases towards around 6.7V-7V, and then sharply decreases past 6.7V. So i set my adjustable low batt detector on the voltage regulator to around 6.8V.

I have BOR disabled, not because i wanted to save current consumption, but rather because i'm running at 3.3V. The way things are set up, I am guaranteed to get a low battery warning before my chip ever gets into the uncertain voltage region where it just might reset due to too low of a voltage.