I am designing an instrument powered by two series AA NiMH, NiCd, or alkaline batteries, I need to display remaining battery capacity on it, so I was wondering what the best way to do it is.

I thought about a fuel gauge, but I am facing some trouble to choose the best one to function with the 3 battery chemistries (NiMH, NiCd or alkaline).

  • 1
    \$\begingroup\$ this is basically impossible, guessing the charge left in replacable RIMH or NICd batterties that could be anywhere from 600mAh to 2400mAh and have voltage 1.2V from 90% to 10% \$\endgroup\$ – Jasen Jan 4 '16 at 21:10
  • \$\begingroup\$ I'll add this on so that you understand a bit about why what you're asking is impractical. In order to even do what you're asking you need to know about the "general case" of each battery chemistry. These would then be stored in look up tables. So unless you're willing to find some pre-made tables and trust them, or willing to test for the numbers yourself, your question is not solvable. Then, even if you could get those tables, you'd have to magically know which one to use. You'd have to have outside information about which chemistry was actually inserted to use the correct look up table. \$\endgroup\$ – Dave Jan 5 '16 at 4:52
  • \$\begingroup\$ So, if I assume that all the batteries to be used are all NiMH. What would be the error if I put an alkaline instead? \$\endgroup\$ – MiguelB Jan 5 '16 at 16:40

Sorry, I can't add comments yet. But instead of trying to measure the battery capacity, why not simply measure the battery voltage?

For example, an AA Alkaline Battery has something around 2500mAh and their initial voltage will be just below 1.6V when brand new. You could simply assume they're 95% empty when they reach 0.9V. Here the datasheet from a Duracell AA: http://ww2.duracell.com/media/en-US/pdf/gtcl/Product_Data_Sheet/NA_DATASHEETS/MN1500_US_CT.pdf as an example, it should be very the similar for other brands.

Using the voltage you can tell approximately how much juice still left on the battery. If you want be more precise, just pay attention to the discharge curve, for Alkaline for example it has a quick drop from 1.6V to 1.4V, than a slow drop down to 1.2V followed by another quick drop to 0.9V.

Rechargeable AA batteries might have a different discharge curve (http://www.fenixtactical.com/manuals/sanyo-eneloop-aa-manual.pdf). As I can't think of a good way of detecting the battery chemistry, you could provide a dip-switch to specify the type or detect the voltage as soon new batteries are inserted and "assume" its chemistry.

At the end, just make a simple table containing: chemistry | voltage | %

Hope it helps!

  • \$\begingroup\$ That was my first idea, the problem is that the NiMH batteries have a very slow drop on voltage in a wide range of capacity... Thank you for your answer \$\endgroup\$ – MiguelB Jan 14 '16 at 18:50
  • \$\begingroup\$ Yes, other factor is the temperature, which could be used to correct the reading. Anyway, most projects will be fine with 5-10% error rate for battery reading. Here some interesting paper from Maxim: maximintegrated.com/en/app-notes/index.mvp/id/121 \$\endgroup\$ – Talk2 Jan 14 '16 at 23:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.