Bottom line: Physics.

Please refer to the "Power" thread, as it has a pretty comprehensive answer to the problem.

Believe me, we discuss this issue internally all the time, and it's not something we're simply ignoring.

As you can see in the "Power" thread, there are a *lot* of issues with power consumption, safety, cost, etc.

Just last week, Nokia recalled 46 million batteries.  This is just the most recent recall of dozens over the last couple of years covering hundreds of millions of defective and dangerous batteries.

It's not just a simple issue of putting out a cheap battery pack.  There's a lot of work to do to get something like this right.

If you know what you're doing, go for it - that's exactly the kind of hacking we're expecting people to do.

As a *company*, however, there are a lot of additional considerations that must be accounted for in order for us to manufacture around an official battery solution.

From the wikipedia page on Lithium:

Combine this with the fact that the cathodes are often made out of graphite/carbon (which also burns well in air...remember chernobyl?) and the fact that over-charging, shorting, or puncturing the battery can lead to thermal run-away conditions that lead to ignition--and the fact that conventional fire abatement methods don't put out lithium fires--and I think you have all the reasons.

In other words, you might void the warranty on your house by burning it down, and we're not responsible for that either.

Then again, the lithium ion battery feat is pulled off day after day in billions of laptops, cellphones, and PDAs worldwide. But as noted previously in this forum, millions of batteries get recalled on a routine basis, for a reason.

Polymer lithium ion, for what its worth, has a reputation of being marginally more dangerous than regular lithium ion. I don't recall the exact reason but I think it has to do with the fact that the polymer matrix will melt at lower temperatures.

I have some rechargable NIMH batteries I will be trying. I am going to use a set of 8, giving 12 volts and I purchased a radio shack casing which directly connects to the 9v connector. I'm going to time how long chumby will stay up, it should be cool. I just want to find out if chumby charges them if it is plugged in as well and if not how I could do that.
This should probably be in the power board. heh

NiMH has some voltage depression issues and doesn't have as good an energy density. It also doesn't handle the case of a deep discharge too well, which is actually possible given the current power architecture of the chumby.

Then again, these issues are lesser evils in my mind, especially since the chumby is neither space-limited nor weight-limited limited in its use case. I guess the big risk is just people returning the chumby because the batteries pooped out due to lifetime shortening effects inherent in these chemistries (or perhaps the lifetime is just shorter than expectation--after all, what's the expected battery lifetime for a device that's advertised as "always on"? even a flashlight advertised as "always on" would have unhappy customers based on too short a battery life), which runs counter to providing the most cost-effective platform to customers e.g., building in the cost of battery related returns into the pricing model would drive up the retail price significantly.

I think your approach of purchasing a battery accessory and plugging it into the chumby might actually be the ideal compromise here. Those who don't need a battery in their use case don't pay for it, and those who do need it can buy one, and the third party vendor (in this case Energizer) has to deal with any battery returns based upon dissatisfaction with the battery's performance.

I'd settle for a battery back-up on the clock.