This is both a tutorial and a question set. If you know all this (or think otherwise on some points) then you know enough to answer my questions :-). Summary sets the scene. _____________ Lithium Ion cells are known for their demanding charging requirements. The penalties for transgression include "Vent with flame", low yield ordnance imitation and 100 cycle lifetime award. BUT they have high energy density and are deemed "low maintenance" if managed well in a user invisible manner. Modern ICs set about this task and usually succeed. Hazards include: - Over voltage on charge will destroy cell. - Over current on charge may/will violently destroy cell. - High temperature will ... - Low voltage on discharge will destroy cell. - Very low voltage on discharge will plate copper across cell so on recharge a low capability charger will produce ordnance. Within cell protection USUALLY includes. - Polyfuse to limit over temperature discharge. - Vmax FET to switch off over voltage - Vmin FET to switch off undervoltage - V_very_low series fuse blower to stop recharge attempts - Over pressure switch to prevent ordnancing - may also blow a fuse. Notable points: -Vmax is typically 4.3V for Graphite anode and 4.2V for coke. Coke is so last year. -Best lifetime gained by holding at about 60% charge and lower than higher temperature. -Overcharging fatal for most chemistries but Spinel electrode allows removal of all Li from electrode at charge completion so can be overcharged with some impunity. -Spinel cells have lower initial capacity and faster degradation rate due to different and more rapid oxidation mechanisms. -Coke anode cells drop greatly in last 30% of energy content and need to be discharged to 2.5V ish for full capacity. -Graphite anode cells much flatter at end and are still at 3.5V at 90 % capacity dropping to 3V at 100%. - Charging to less than max design V (ex 4.2V for Graphite rather than design 4.3V will reduce capacity but notably increase cell lifetime. - Deeper cycles = fewer cycles lifetime. - Quoted lifetimes are usually 500 cycles to 90% and maybe 600 cycle lifetime (?). 1000 cycles is claimed by some in some cases. But cells also have a finite hours lifetime regardless of cycles due to oxidation in cell. 2 years generally deemed max for typical types. Newish LiFe... cells in eg OLPC target a 5 year lifetime but have a lower gross capacity. - Cells charge to about 2/3 capacity in 1 hour at Imax and rising V and then to balance in 2 more hours on Vmax and falling I. 1 hour "fast chargers" tend to stop at start of I taper and only provide about 2/3 of capacity. FWIW this is probably about the best point to leave the battery for storage. Requirements for longevity and mechanical integrity include: -Low rate precharge to check for under voltage - may be able to trickle up to safe starting point. -Constant current up to voltage limit. -Constant voltage with current taper to current cutoff limit. -NO trickle but occasional boost if Vterm falls by some amount. _______________ So Li Ion one cell terminal voltage of 4.2V down to 3.5V and energy density are ideal for some applications if the price is right. Use of NiCd or NimH in a similar application would often not be well served by 3 cells as range lower. Using 4 cells gives V too high unless a SMPS is used. Given Icharge_available << Ichg_max_allowed If cost apart from cell is key, why should one not implement a very minimalist charger that keeps the cell safely away from no-go boundaries? Say charger is allowed to cost <= $US0.10 max in parts in large volume. Even a zener clamp will keep a graphite cell <= 4V2 if properly designed. Would almost certainly be << 4V2 to ensure this in all cases so capacity will be down or even much down. Low voltage cut-off can be arranged at extremely low cost with a bipolar switch and eg zener plus glue. If I chg << Ichgmax then should be OK as long as Vmax limited as above. Charge cutoff at end of charge taper seems the hardest at zero cost BUT if Vzener clamp or whatever is << 4V2 then it will sink ALL charge current as 4V2 approached so should have zero current below 4V2. A well sloped clamp curve would help. CONS This arrangement is very inefficient of charge energy if this matters. Vmax < 4.2V means low net capacity. As eg NimH has energy density > 50% of well charge Li Ion this could fall to near NimH capacity. In fact, as capacity is 2/3 at 1 hour at Vmax then stopping below Vmax may be <= 2/3 capacity. However, as battery will still current taper at < 4.2V the above is not certain. The bad energy utilisation at low Vmax_allowed may mean that NimH are superior in absolute terms. Only obvious (to me) technical advantage of Li ion over 4 x NimH here is higher terminal voltage of NimH. Prices may dictate one or other but it's hard to imagine that Li Ion should be cheaper than volume NimH prices per energy. QUESTIONS: Is a 10 cent charger viable? What wouldn't work that I have described? Why not use 4 x NimH? R -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist