Mauro Thanks for the advice about the Lithiums. The trouble with this particular design is that this is for a commercial product that draws about 25mA. Using expensive, hard to find batteries are not a good option in this particular case. So I either need to implement common old garden variety inexpensive batteries or I need to use rechargeable batteries. I can also not make the circuit run on 3V as suggested, since some of the parts used, like an MPX5100A pressure transducer, MUST have a 5V supply. The best solution so far seem to be Steve Lawther's suggestion to use the Microlinear ML4890CS-5 which is a combined booster / 5V linear regulator that can work down to 1V with low ripple and good effieciency. I think I'll try that and see how it goes. Thanks to everybody that gave me advice. Rgds Werner "Mauro, Chuck" wrote: > Werner, > > I'll bet you can find Lithium's in AAA size. Honest. > > At 2x3V, you get a very nice 6V supply. I would suggest a low dropout > regulator to power the system. The nice things about Lithium are: > > 1) ~ 3X energy density of Alkaline's, > 2) 2X the voltage in the same space, > 3) 10 yr. shelf life, > 4) Stable, relatively flat discharge curve. > > I realize they cost more, but this is a drop-in replacement meeting your > physical requirements, and with the addition of a low dropout regulator > to your design (in the 0.1V range), you should be able to realize great > battery life (this may take well thought out power management design > principles, such as power switching the A/D and using sleep mode for the > PIC depending on your static and dynamic current requirements). Perhaps > you could run your circuit unregulated for longer battery life? Or even > at a lower nominal Vdd (say, perhaps, 4 or 4.5 V). There are probably > several techniques that can be used here. Personally, I would seriously > look into running the whole thing at 3V, if possible. Look at the > common mode input requirement of the A/D circuit (are you using a > differential mode?). Can you control the input voltage range of the > system? Will the reduced dynamic range at, say, 3V still provide > meaningful results and enough bits of precision from the A/D? > > These are some of the things I would be asking myself and considering > if I was challenged with the design issues you've mentioned. > > I envy you. It sounds like a fun problem to solve! > > Good luck, > > Chuck Mauro > RAPID Engineering > KLA-TENCOR > > > -----Original Message----- > > From: Werner Terreblanche [SMTP:wterreb@PLESSEY.CO.ZA] > > Sent: Tuesday, February 17, 1998 3:05 AM > > To: PICLIST@MITVMA.MIT.EDU > > Subject: 2V to 5V conversion > > > > I found a nice enclosure that I would like to use for a PIC based > > product. Unfortunalely this enclosure will only accept two AAA > > sized penlight batteries, whereas many of the IC's in my prototipe > > requires at least 5V. I'm thinking of using two 1.2V Nicad > > batteries. Surely many other people on this list have also had a > > requirement for something like this some time or other. > > > > I know there are dc to dc converter and voltager doubler IC's > > available from companies like Maxim, but I would still like to hear > > the opinions of other list members about the usage of these types of > > voltage doubler IC's. Since there will also be a 5V, 12 bit A/D on > > the circuit, I need something that is not very noisy and still quite > > efficient. I have never had to make such a battery operated > > product before. What would you recommend or not recommend? Is it > > feasible to run a 5V based circuit from a 20 - 2.4V supply with dc > > to dc conversion, or is it not recommended? > > > > Rgds > > Werner >