TMBartman 
Hello -
I will be needing about 20 - 30 volts with roughly 10mA of drive current from 3 AA batteries (4.5V).
Does anyone have experience and would recommend an IC to use? I'm pondering buying eval boards then thought - why not ask in here first?
I got a few from Mouser then found the pin-outs are so small I cannot breadboard them to test first. Digikey has many but not in stock. Any past experience?
Thanks all!
Tom
aero1 Take a look at the Linear Tech LT3494. Available from DigiKey but not Mouser.
geoelec If 10 mA will be the current through load from the source will be (30V/4.5V)X0.01A=67mA plus conversion losses.
If the circuit includes an AVR then a PWM driven boost circuit is just a transistor, FET, coil, diode, capacitor, and firmware. Connect Vcc to the boost voltage through a 15-25 volt zener, start up using the raw battery through a blocking diode and ramp up PWM to a Vcc of 5 volts. That will provide arbitrary current at a stabilized 20-30 volts, at the cost of maybe 100 milliwatts wasted through the zener. If the AVR draws more than a few milliamps it would be more efficient to sample the boost voltage through a divider or use the separate boost chip.
icarus1 If there is a microcontroller in your circuit then for so small DC/DC you could use it to keep a low price for your whole product.
Just google for Step Up (or Boost) Converter circuit. In my opinion you need a PWM output, an Analog Input, 1 x N-MOSFET (R gate + R pull down), 1 x Schottky Diode, 1 Coil and, 1 Capacitor and a Resistor Voltage Dididr. I believe that for such a low output current you could get more than 95% if you make the calculations of the passive components.
Also if your circuitry is been supplied by a primaty cell battery then pay special attention to the current consumption. A current always flows via the coil and the voltage divider to GND. A simple method to solve this problem is to contol using an open drain (NPN or N-MOSFET) to control the R2 of the voltage divider. Also, in this case, a Schottky diode is needed from the analog (cathode) pin directly to Vcc (anode) to protect the Analog input from high voltage when you release the R2 to save energy.
So when you want to measure the Step up voltage you have to enable the R2 and then disable it after the measurement.
Also, if there is a load always connected to the Step Up output, then a current will always flow even when the step up is not functioning. That's why you need another switch before step up (a P-MOSFET).
The current consumption and the efficiency are problems that you have in both cases a descrete Step Up ic or a microcontroller.
I hope this helps.
TMBartman Thanks for the ideas. Unfortunately I do not have a microprocessor on the board so I was looking for boost controller that someone may have had a good experience with.
ezcomp 
We've used LTC1871, it's a very nice part but at $5.37 in single quantity from digikey, it might be a little more expensive than what you want.
oddbudman 
MC34063? Might not like working at <3V, or be very efficient, but it should work and is cheap.
oddbudman
nachus001 
LM3578 ?? works from 2V to 40V , PDIP-8, built in pass transistor. USD 2.06 at DigiKey.
Nachus
