This sounds like a sensible approach. Difference between a full and a depleted Lipo is just 1.2v max per cell, so I think 1 sec overlap shouldn't hurt the reserve battery too much. I'll do some testing to understand max current being taken from full battery when overlapped to confirm it is still on product margins (10C=15A for this reserve battery). Thanks 2008/2/19, Dr Skip : > You could use FETs or even low power relays. Don't do any fancy switching, > since you can overlap the batteries for a small instant for the switching. > > Use either a comparator or a lm3914 (IIRC) as voltage level detector. Control > the relays as inverse of each other and use a 555 timer to overlap the > batteries for some small time (1 sec?). You could probably just put a few > hundred uF cap in parallel with the first relay so it stays latched a second or > two after the switching. QED. ;) > > > Ariel Rocholl wrote: > > I would like to connect two batteries in parallel to service a RF > > receiver. Both are Lipo batteries but battery A is high capacity and > > main battery whereas battery B is reserve low capacity battery. > > > > The average consumption of the RF receiver is about 2A, battery A is > > 5A/h and battery B is 1.5A/h. > > > > I do not see an easy way to: > > > > 1) have battery A depleted then (and only then) > > 2) switch to battery B till it is depleted too. > > > > I would assume a large capacitor plus a very fast switching transistor > > (IGBT?) may help to make it transparent to the load to a certain > > extent. > > > > But perhaps anyone knows a better way to do this. Any help really > > appreciated, I wouldn't like to reinvent the wheel but, except for low > > current ICs in Maxim, I didn't find anything else doing this already. > > > > TIA > > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- Ariel Rocholl Madrid, Spain -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist