In Shanghai - rushing ... Is this a one/few/many / vast quantity device? There many LiIOn charge control ICs. If it is one or few and not too caring of worst case you can cu some corners. If the cellas are protected intermnally (as you say youes are) the= n the corner cutting is usally covered by the protection. YMMV. Biggest is protection against deep discharged cells - system goes into low trickle or even with low duty cycle and may give up if no improvement after a while. Reasin is that a deep discharge cell may vent with flame[tm] if given norma= l charging current. Once you have extreme cases covered the basic LiIon cycle is very very simple and can be implemented with very little hardware. Charge at constant current of I max until Vmax is reached. Hold at Vmax until I min is reached. Stop. Temperature compensation needed for Vmax usually. Thinking up semi discrete hardwqae to do that is not too taxing. THEN you can handle charge while operating from PSU etc with ease. War of the orlds has started here. Presumably fireworks. Time to go. R On 30 October 2010 17:22, Josh Koffman wrote: > On Fri, Oct 29, 2010 at 11:56 PM, Josh Koffman > wrote: > > So here's what I'm looking for. I will be using a single cell, either > > 900mAH or 2600mAH. The unit will be running self powered, off the > > pack, connected via Bluetooth, or it will be connected to the computer > > via USB. If it's connected via USB to a computer, it would be nice if > > it could be doing some charging. In addition, to minimize the number > > of connectors, it would be nice to use the USB jack as a source for > > charging from a higher current wall adapter. If the charger doesn't > > provide it, I'll need a 3.3V regulator (I have that lined up already). > > I would like to be able to use the device even if it's connected to an > > external power source, so I can use those "backup" USB power sources > > if I find I'm running low on power. > > > > Ideal package would be surface mount, but something I could hand > > solder, QFP, SO, SOT, etc. Non exposed packages (like DFN) are > > probably a bit too hard for me at the moment. > > And if that wasn't enough, I guess having some sort of battery fuel > gauge might be helpful as well. This would be for the user to see, not > necessarily for the processor to track. Charging, and charge complete > LEDs would be nice. I'd settle for something like the bluetooth > headsets use - flashing red LED for charge, solid green for complete. > Flashing green for battery low. Or something like that. > > I might give up on this idea if it looks like it'll take too much room. > > Josh > -- > A common mistake that people make when trying to design something > completely foolproof is to underestimate the ingenuity of complete > fools. > -Douglas Adams > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .