On Fri, Jun 17, 2011 at 3:05 PM, V G wrote: > On Fri, Jun 17, 2011 at 1:44 PM, RussellMc wrote: > >> > How large have you made the R1/R1 combination and still gotten decent >> > voltage tolerance over "room" temperature range? >> > >> > The input bias current is rated at 2 to 4 uA +- 0.8 to 2.5 uA over >> > temperature range 0 - 70. >> > >> > Lots better than a zener but maybe would drain the battery fairly >> > quickly unless the R1/R2 pair was very high and then input bias curren= t >> > shift eats away at accuracy. >> >> His app is a charger. >> Power will be on most of the time and even if off is not going to >> cause vast problems over charger residency type periods. >> >> 18650 LiIon are about 2000 mAh + . >> >> 1 mA drain at 24 hours =3D a bit over 1% capacity. >> 100 UA - ~0.1% >> 10 uA ... >> >> In order to avoid thinking about temperature compensation I suggested >> he move the voltage back another 0.1 V to 4V. >> This reduces capacity but (probably) allows simple float and greatly >> improves cycle life. >> Also avoids having to do current tail level triggered charge termination= .. >> >> > > Okay so I figured out another mechanism to charge. Uses a few transistors= , > but simplifies the charging itself. > > I'm guessing many will find this design messy and unnecessary. Please > comment. > > Here's my circuit: http://solarwind.byethost7.com/pic2.png > > The left side MOSFETs, when enabled, connect the batteries in series, > allowing regular discharge. > > The right side MOSFETs and BJTs, when enabled, do a parallel current > limited charge up to 4V. As far as I know, the BJTs won't allow the > batteries to charge each other when the BJTs are disabled. When they are > enabled, the batteries are charging anyway. My only concern with this des= ign > is reverse biasing the BJTs and the batteries charging each other when in > parallel charge mode. > > I plan to use SOT-23 transistors to keep it small. Battery discharge I'm > guessing will be maximum 1A (to charge my iPhone). Battery charge will be > around 2A per cell. > Whoops, just noticed that my NPNs should be low side. Will fix now. --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .