Perhaps with a gadget that stops working when the voltage falls below 1.45V per cell.
And such a gadget is one that no one would buy, and consequently no one would produce. Electronics designed to run off AA batteries already extract as much energy as is practical from them. From those discharge curves you can see that there is a gentle downward-sloping section, followed by a distinct "knee" at the end where the voltage falls rapidly. That is where the gadget would be designed to work down to.
Sure there's a few percent more beyond that, but adding a boost converter which attempts to extract that energy won't work - the boost converter itself will be consuming energy, and likely far more than is available past the knee, so this device could actually reduce how long your batteries last!
As for the miniaturisation - this isn't that small; similar boosters have been present in MP3 players, cellphones, and other battery-powered electronics for years. Fitting everything on a PCB with the diameter of an AA and less than 1mm thick is not that hard. Look at the ISL9113A for one example. I haven't laid out a boost converter in a design with that space constraint, but it turns out to be roughly of that area anyway and I'm sure if you did things like cutouts to sit the passives into the board, it'll be thin enough.
And such a gadget is one that no one would buy, and consequently no one would produce. Electronics designed to run off AA batteries already extract as much energy as is practical from them.
Indeed. If your gadget can make use of NiMH rechargables, then you're already having to deal with about 1.2V per cell, and that's when they're fully charged!
So if you need 3.0V or 3.3V nominal for your main supply, you are probably already using a buck/boost switching regulator. As long as the battery can supply enough current, your gadget will keep going, and drain pretty much all the useful energy from it.
And such a gadget is one that no one would buy, and consequently no one would produce. Electronics designed to run off AA batteries already extract as much energy as is practical from them. From those discharge curves you can see that there is a gentle downward-sloping section, followed by a distinct "knee" at the end where the voltage falls rapidly. That is where the gadget would be designed to work down to.
Sure there's a few percent more beyond that, but adding a boost converter which attempts to extract that energy won't work - the boost converter itself will be consuming energy, and likely far more than is available past the knee, so this device could actually reduce how long your batteries last!
As for the miniaturisation - this isn't that small; similar boosters have been present in MP3 players, cellphones, and other battery-powered electronics for years. Fitting everything on a PCB with the diameter of an AA and less than 1mm thick is not that hard. Look at the ISL9113A for one example. I haven't laid out a boost converter in a design with that space constraint, but it turns out to be roughly of that area anyway and I'm sure if you did things like cutouts to sit the passives into the board, it'll be thin enough.