Parallel operation is never recommended BUT will work well enough [tm]. Start with all new and they should track down with minimal intercell losses. YMMV :-). Worse - CR2032 are grossly underrated for the task. 4 x CR2032 ~~=3D 20 x 12mm in series. You can probably get a 3 NimH series stack of about that size. capacity would be well down on CR2032's BUT rechargeable with ease. Where did you get the current specs from? Are you planning to use the same brand as you have specs for? Max current rating might vary substantially between makers. eg Sanyo say 4 mA cont and Energizer "suggest" 0.2 mA. Sanyo datasheet suggests that 3 mA continuous is about the sensible limit at 25 C and above. http://www.biltek.tubitak.gov.tr/gelisim/elektronik/dosyalar/23/CR2= 032pil.pdf Energizer consider 7 mA for 30 seconds a day in 12 bursts is "daring" http://data.energizer.com/PDFs/cr2032.pdf Good discussion. http://interactive-matter.eu/2009/08/driving-circuits-from-a-cr2032-lit= hium-coin-cell/ At the currents you suggest (30/4 to 50/3 mA range =3D 7.5 - 17 mA) you ar well above their continuous rating and can expect poor lifetimes and substantial current drop in use. A single AAA NimH with boost converter would be less "extended by this role. Eben and AAA alkaline. There are many many many LiIon "prismatic" (aka flat) cells around - the smallest would likely tolerate this role well. LiIon is very easy to charge reasonably well. Keeping Vmax charge slightly lower than usual reduces cell capacity, removes any need for temperature compensation on full charge, and vastly extends cycle lifetimes. f you set Vmax slightly lwo - say 4.1 or even 4.05 V/cell (4.0?) charging can be made 100% foolproof - plug it in for as long as you like and it just assymptotes to Vmax at a safe value. An competenyt charger can bew two R's in series (R1, R2) to the battery from V+_charge_in with a zener clamp to ground beyween the two ressistors and an eg TL431 clamp at the battery. Zener set charge rate max via (Vz-Vcell)/R2. There are issues with this if the cell goes very low but they can be essentially eliminated with a v small amount of design. TL431 needs two resistors to "program" it. R1 limits worst case current for high Vin. Whole charger can be 4 x R, 1 x zener, 1 x TL431. Compact and cheap. Slightly more complex can do better again, but the basic version is fine. IF the usage % s low a PV panel may allow charging from roomlight or distant daylight. Probably acceptable size PV panel gives 50 mA at 1 sun =3D 0.5 mA at 1% sun= .. If you draw 25 mA operating you need 50:1 + duty cycle. 1 hour of 1% sun per minutes of use. A few minutes in OK daylight (say 25-50% sun) gives 1 minute use per 1 to 2 minutes of sun. Desklamp close can give > 10% sun (up to 1 un with abusive brightness and heat). Russell On 30 May 2011 20:36, V G wrote: > Hey all, > > This is regarding my optical guitar tuner idea, where a circuit will be > powered from CR2032 coin cells. The circuit will be drawing (I'm guessing= ) > around 30-50mA due to 2-4 LEDs pulsing at sound frequencies. > > I know that CR2032s are designed for around a 1mA continuous draw, and up= to > 10mA pulse draw, but people make LED keychains out of these things all th= e > time and those LEDs draw 20mA or so continuously. I want to power the tun= er > with these cells, but I don't want the performance of the batteries to > degrade as much due to the relatively high current draw. Therefore, I'll = be > putting these batteries in parallel (about 3 or 4 of them in parallel). > > Is this okay? What kinds of issues will there be? Will there be issues of > the batteries charging into each other? Things like that? > -- > 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 .