I'm needing about 108 NiMH AA for 3off Replica Ever Ready B104 45V=20 battery packs. (36 per pack) It's for a miniature suitcase spy style Radio set (RX & TX) Any recommendations on sources with cheap postage that are going to last=20 well? Service Life and cost rather than mAH capacity are the issues. Also thoughts on getting more than 3500mAH @ 1.25V nominal by=20 paralleling multiple NiMH? Other Issue. These are miniature Military Russian valves (tubes to=20 Americans) and *REALLY* are 1.2V or less filament. Ideally maybe 1V to=20 1.1V for longest life and consistent performance. Fresh an NiMH is almost 1.5V (1.48V I have measured). Flat is rapid=20 after 1.0V or 0.9V? The VERY early 20's radios used 2V cell and rheostat on filament. It=20 would be nice to do this without transistors or ICs, so I thought a 3=20 way switch, two values of resistor and last position a short to keep=20 filament supply around the 1V to 1.1V. Any better ideas. If I *DID* use a regulator, 1.08V out for 1.1V to 1.5V in and then less=20 than 50mV drop as cell heads lower is not something off the shelf? Total consumption might be 700mA to 800mA @ 1.15V (The Valves are 16mA,=20 24mA, 54mA and for "P.A." valves 220mA). Normal HT is 44V nominal (need=20 not be regulated, so 36V (near flat) to 51V (fully charged via diode, 3=20 x 45V packs diode paralleled) and on Transmit relays connect the 3 packs=20 in series (no diodes) boosting HT on P.A. from 45 to 135V ( 108V near=20 flat to 159V fresh) with no diodes and one pack feed +45V to rest of radio. The RF class AB push pull valves are used for speaker on Receive and RF=20 PA on TX. Only 44mm x 11mm! --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .