• EEVblog #141 – AA Alkaline Battery Capacity Measurement

    How much Wh capacity does a typical Alkaline AA battery have? Dave intends to find out with his new Gossen Metrawatt Metrahit Energy multimeter

    Be Sociable, Share!

      About EEVblog

      Check Also

      EEVblog #820 – Mesh & Nodal Circuit Analysis Tutorial

      EEVblog #820 – Mesh & Nodal Circuit Analysis Tutorial

      Dave explains the fundamental DC circuit theorems of Mesh Analysis, Nodal Analysis, and the Superposition ...

      • Pingback: Tweets that mention EEVblog #141 – AA Alkaline Battery Capacity Measurement @ EEVblog – The Electronics Engineering Video Blog -- Topsy.com()

      • Bearman

        I have found that the battery holders have a rivet that holds the wiring and spring in place on the battery holder frame. For critical applications with these holders I have gotten into the habit of soldering the rivet to the spring and the wire so I have the best electrical connection possible.

        I have also found that many commercial gadgets that fail use this same technology for the battery springs and power leads.
        Most times just soldering the spring to the wire has repaired many of these faulty gadgets.


      • http://e-motion.lt elektrinis

        Some batteries have a way better characteristics than others. For example, A123. These have extremely low ISR and almost no characteristic drift due to temperature or discharge current. They hold up to 30C (50A for 2,3Ah cell) discharge rates! Here are my tests:
        (you might want to use google translate)

      • Anthony

        35.5 degrees Celsius = 95.9 degrees Fahrenheit

        Google is my friend. But yeah I am a Yank as you say, And I to wish we used metric. You know NASA lost a few mars probes because some idiot forgot to convert to metric. Or so I heard. I totally believe it though.

        They probably misread a data sheet hehehe =). Or didn’t read one…

        • Jay

          I have nothing against metric in general, but for everyday, non-scientific temperature measurement, Fahrenheit is actually better. See http://en.wikipedia.org/wiki/Fahrenheit -> Usage for details. Both Fahrenheit and Celsius are arbitrary scales. For scientific use, Kelvin is it.

          • Ramón

            An example:

            I can wash the clothes in the washing machine at 30 ºC, and maybe I could wash at 60ºC or 90ºC, 90ºC is a good temperature!, so somebody could say I am washing three times 30 ºC (3 x 30ºC = 90ºC), but this is not true!

            The “real” scale of temperatures doesn´t start at zero ºC, because 0 ºC is an arbitrary value, that was chosen because water becomes frozen. Temperatures start at -273,16ºC, that is absolute zero (Kelvin is an absolute temperature scale), the temperature when water becomes frozen corresponds 273,16 K.


            30ºC = 273,16 + 30 = 303,16 K.
            90ºC = 273,16 + 90 = 363,16 K.

            And the shift from 303,16 K to 363,16 K is about 20% more, but not three times.

            What matters if 90ºC is better….

            • Ramón

              Hey, hope all is well, i wanted to say -a question- if washing at 90º C is better than 30º C?, but then we would have to speak about what “better washing” means… it was just an example….

            • Leo Bodnar

              Typically chemical reaction speed doubles with each 7..11 degrees C increase in temperature. So where you use K or C is irrelevant.

          • Karl (not that Karl, the other Karl)

            Really, I can’t hear this ” is better” rubbish any more. Typically such a claim is then “proven” with some artificially created examples, carefully doctored to provide the desired “proof”.

            I tell you what, using the same standard of what constitutes a “proof”, I can construct “proofs” to demonstrate that any random metric unit is “better”.

          • Logictom

            I don’t think you can say either is better but at least with Celsius you have everyday reference points, ie boiling the kettle and ice

            • http://www.eevblog.com EEVblog

              I agree. Celsius is far superior from that standpoint alone.

              • Frank

                How lucky you are. Here, in France, we’re at -1°C. I envy you,

                Anyway, wery good tutorial, thank you !

      • Ramón

        Pretty good info Dave, Thanks.

      • Richard

        That’s a good discussion of the factors to consider. But it seems like, in the end, it’s sort of like measuring jello with a micrometer. Because even if you measure the exact battery life of one battery in one usage scenario, it won’t precisely apply if you change the usage scenario OR the battery. Even going to another battery of the same model by the same manufacturer may change things a bit, to say nothing of switching manufacturers. To further complicate matters, most interesting devices are neither constant current nor constant power, but instead they vary the load at various times as needed, sometimes in somewhat unpredictable ways.

        So in the end, you sort of end up doing a lot of estimation and hand-waving, applying fudge factors. Still, it’s good to be aware of the factors that affect battery life, and good to know how to back up the hand waving with at least a little bit of factual evidence.

        • http://toolhacker.com Wartex

          >it’s sort of like measuring jello with a micrometer

          BAHAHAHAH!!! Great analogy.

      • http://toolhacker.com Wartex
      • Matt

        Since it was so hot wouldn’t the temperature affect your experiment? The datasheet graph was made at a temperature of 70 degrees F. Was your ambient temperature 95 F or so at the time of your measurement or would it not make a significant impact on the results?

        Great information and keep up the great work!

        • http://www.eevblog.com EEVblog

          No, as the measurement video was shot several days before as part of the previous episode. It was actually around 21degC overnight whilst it was discharging.

      • walter delbono



      • Pingback: Battery Capacity Measurement - Hacked Gadgets - DIY Tech Blog()

      • http://TALKINGELECTRONICS Colin Mitchell

        In one week I had two AA Alkaline cells go “dead” in two different devices.
        They are much more unreliable than the old “dry cell.”
        Take my warning. Do not rely on them for emergency equipment. All the cells were new and just when the device was needed, the output was zero.
        The cells were different brands and not the cheap variety.

      • Pingback: ToddFun.com » Blog Archive » Links for Beginners in Electronics()

      • Michael K.

        Thanks for clarification on these complex theme! Very helpful.

        BTW I found this episode on the Gossen Metrawatt main page: http://www.gossenmetrawatt.com/

      • http://digital-escape.blogspot.com/ cda

        What about if you let the battery without any load for some time. You can get some more power from it afterwards, can’t you ? On the other hand how would you characterize the type of load on a consumer product such as a photo camera for example? Is there anything “constant” ? (there’s some flash usage, a focus motor …etc.) And you usually don’t use the camera non-stop until the batteries die. How could you possibly measure the capacity of the batteries in such condition ? (attaching some I/V logger and use the camera as much as you can…?)

      • Pingback: Battery Capacity Measurement | Mobile News()

      • Pingback: Briefing – Episode 41, Batteries » A* to G()

      • Robert Tennant

        I bought Supacell primarily for the kid’s toys after buying Duracell and Panasonic for too long, they always let me down and last 5 minutes! A false economy! Supacell on the other hand are excellent value for money and last the longest, I won’t be bothering with Duracell etc again.

      • Dave

        However the Ah measurement is useful for counting charge. When you work with Lithium Ion cells, you can count and log the current with time (Coulomb Counting) and estimate what the level of discharge is.

      • Pingback: ccn2785xdnwdc5bwedsj4wsndb()