The battery question I really dread

-April 04, 2013

No, the question I dislike is not "what's the story on those Boeing Dreamliner batteries, anyway?" And no, it's not "are those lithium batteries I bought for my kid's toys going to catch fire?"

Instead, it's when someone hands me a couple of basic cells (AA, AAA, C, or D) and asks this simple question: "Hey, is there any life left in these?" Somehow, they expect a clear answer: "Yes," "no," or "some," end of discussion. [Why there is no A or B cell designation is the subject of a future column.]

But they don't want to be told that it is a fairly complicated question, depending on the battery chemistry and the intended use. Back in the old days, when carbon-zinc was the only battery chemistry in widespread portable-product consumer use (with some alkalines as well, and lead-acid in cars and fixed installations), you could check a battery by measuring its no-load, open-circuit potential, and come up with a reasonably accurate answer.

Why? Because those cells had a discharge curve that was somewhat linear, and their cell's open-circuit voltage corresponded to remaining energy. But for today's more advanced chemistries, especially the many variants of lithium-based cells, the output is fairly close to nominal - until they lose most of their charge. Adding to the challenge is that the remaining energy they deliver depends somewhat on the discharge rate, so even knowing how much apparently remains gives only part of the answer.

So what do I do? I have an informal test process that seems to work "OK enough." I have several high-intensity, high-drain flashlights with halogen bulbs (I use the Maglite ones). I put the battery cells in question in the flashlight which uses that cell size, and look at the light-output intensity. Sure, that's qualitative, but that's as good as I can do without a fancy setup.

The flat-output characteristic of new chemistries has had an effect on OEM product design, as well. Traditionally, the way the "remaining life" was assessed in these products was by measuring the cell's terminal voltage. To do that now with validity requires a fairly accurate voltage measurement, usually more costly than the product can tolerate.

So for products where cell status is critical, OEMs have gone to the more rigorous method of coulomb counting, actually tracking the electron flow out of the battery (and back into it, if it is a rechargeable cell). It takes more work, but advanced battery management ICs can do this with fairly good precision and at modest cost.

[Note that this is sometimes called "gas gauge" measurement, but I detest that slang term; it makes it sound as if the battery is being powered by a gaseous substance or even gasoline. I'm a little less uncomfortable with "fuel gauge," but I prefer to use terms such as "energy" or "charge."]

So, what's the hardest "simple" technical question you get asked? Is it "when will the project be done?" Or is there something else?

Loading comments...

Write a Comment

To comment please Log In