On Wed, Jun 29, 2011 at 6:30 PM, Jesse Lackey wr= ote: > Hello all, I'm part of the design team for a high-volume consumer > product that needs to run for several months on a single li-ion battery. > As such, identifying quiescent current 'leaks' and checking that the > battery supplier is delivering batteries that meet specs is important. > > Here's what I'd like to do, and prefer to buy vs. build, if the price is > sane ($hundreds fine, $thousands not). > > I'd like to measure current into and out of the battery, the range is up > to 400mA going in when charging, and out of will be brief pulses of > 150mA when doing RF transmit and down to 50uA when in a sleep state. > > I use a sigma-delta A/D converter with a differential input for this purpose. The MCP3551 works well. With a 5V reference and assuming you can get around 20 bits of data, the per-LSB voltage is 5V / 2^20 =3D 4.77uV So with a 10 ohm resistor your lowest increment is about 477nA. In practice it will be pretty difficult to trust all of those bits, but you have plenty left over. Of course there's noise and error associated with this. Using a smaller value resistor will give you better impedance characteristics. I have to give Olin credit for this idea. --=20 Martin K. --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .