2
\$\begingroup\$

Hi I am quite new to electronics, I had a project powered by 3xAA batteries in series (~4.5 V) with an LDO dropping it to 3.3 V. It went flat quicker than expected but what was very surprising to me was that the voltage of the three batteries was so different ~0.5 V, ~0.6 V and ~1.5 V total ~2.6 V. Is it normal for a battery pack to discharge so unevenly? Can someone explain why?

If it is relevant it sat most its time in the μA range with hourly very short spikes to ~150 mA.

\$\endgroup\$
8
  • 1
    \$\begingroup\$ It's not uncommon for the voltages of a series arrangement of batteries to yield the results you see. In a perfect world? Not so much. But reality impinges. So what can you do? Different battery manufacturers and product lines would be the first thing to examine. Try to keep all three batteries from the same date of manufacture (they self-discharge) or same manufacturer's lot number. That should help somewhat. But if you have to tolerate people stuffing random AA batteries into the holder, then you will have to tolerate this kind of problem. The current flows through all and affects chemistry. \$\endgroup\$
    – jonk
    Oct 25, 2019 at 6:46
  • \$\begingroup\$ OK so it is uncommon. I used three from the same manufacture (energizer) but can't guarantee the batch dates. I will investigate the cheap battery holder also \$\endgroup\$
    – AndrewT
    Oct 25, 2019 at 6:50
  • 2
    \$\begingroup\$ I doubt the holder is the problem. (I'm not sure how it could be the problem.) But if you are using fresh batteries from the same battery type, then you could honestly hope for better. However, if you are running them down completely, then I'd expect a much wider variation in the end. One of the batteries might have more energy storage and therefore may show good voltage remaining when then other two show much less. The curves, as batteries reach their end-point, is rapidly varying. So just beware of that issue, too. \$\endgroup\$
    – jonk
    Oct 25, 2019 at 6:54
  • 1
    \$\begingroup\$ "It went flat quicker than expected" - how long did it last? \$\endgroup\$ Oct 25, 2019 at 7:20
  • \$\begingroup\$ Are those voltages while the cells were still under load? \$\endgroup\$
    – HandyHowie
    Oct 25, 2019 at 7:44

4 Answers 4

2
\$\begingroup\$

You should use 3 cells from the same manufacturer, same model and same chemistry, out of the same batch or the same package. Of course the cells should not been used before.

If all cells have the same history and manufaction tolerances of capacity are small, such an uneven discharge should not happen.

But when the history of those 3 cells was different, uneven discharge is possible.

\$\endgroup\$
3
\$\begingroup\$

Assuming alkaline / zinc chloride / similar chemistry:

If you got eg 0.5, 0.6, 1.1 then it could be explicable by the fact that there is very little energy in the area under about 1V, so that if one cell had say 5% more energy content than the other two it would still be on the "final approach" to fully flat while the other two were essentially completely exhausted.

However, a cell at 1.5V has the majority of energy capacity remaining - probably 90%+ .
For one cell to be at 1.5V while the others are fully exhausted then they would have had only 5% - 10% of their new energy content at the start of discharge.
SO this is not a batch variation - two of the batteries were very close to dead at the start of discharge OR something else has happened not mentioned in your question.

If the batteries are 'alkaline' they would retain a substantial portion of full charge for many years. If they were Zinc Chloride or other similar chemistries then they may well have very little capacity after say 2 years of shelf life.

\$\endgroup\$
5
  • 1
    \$\begingroup\$ He may be measuring them open circuit, in which case nearly any battery that isn't completely exhausted will read around 1.5V. I'd suggest testing them under load; if they're still significantly different, then probably one of them was much younger or less used than the others. \$\endgroup\$ Oct 25, 2019 at 15:56
  • 1
    \$\begingroup\$ @CristobolPolychronopolis Battery open circuit voltages may tend to be higher to much higher than loaded. But for typical consumer "1.5V" cells there is usually a distinctive decline of voltage with level of depletion. I regularly check Alkaline AA batteries for 'degree of use' and can readily distinguish between new, somewhat used very used and "why is that still here". | Also, "brand new" batteries can have their age determined reasonably well from their slow decrease in open circuit voltage with age. A "factory fresh Alkaline tests around 1.65V - maybe 1.66V if very new. 1.60 is oldish ... \$\endgroup\$
    – Russell McMahon
    Oct 25, 2019 at 19:44
  • 1
    \$\begingroup\$ @CristobolPolychronopolis ... while 1.55 - 1.60V would be still unused but probably 2+ years old. \$\endgroup\$
    – Russell McMahon
    Oct 25, 2019 at 19:45
  • \$\begingroup\$ I will try fresh batteries and try again. I have added a WDT that will reset it if awake and running for more than a sec which should ensure it cant get stuck in a loop and drain the batteries. Thanks for the help. \$\endgroup\$
    – AndrewT
    Oct 27, 2019 at 8:10
  • \$\begingroup\$ @AndrewT Measure Vbat for each cell OC at start, loaded at start and whenever you remember after that. Having one or two of three die totally at the end while one has SOME charge is common enough, but 1.5V in one is unusual . ||What sort of battery are they? Model and brand? \$\endgroup\$
    – Russell McMahon
    Oct 27, 2019 at 10:03
2
\$\begingroup\$

The different voltages on the batteries is a little unusual but not really important. You say you have a LDO but don't give it's number or the amount of quiescent current. Many/most beginners don't realize that voltage regulators continue to draw current even when the load doesn't require any. This is the most likely cause of the rapid discharge of the batteries.

\$\endgroup\$
1
  • \$\begingroup\$ Hi thanks. It is a special LDO with 2-3ua quiescent current. Total of 9ua while in deep sleep and only awake 40ms every hour. I think one of the batteries must have been old. I have added a watch dog timer to reboot if awake more than a second and new batteries. Will see hope it goes. \$\endgroup\$
    – AndrewT
    Nov 19, 2019 at 23:45
0
\$\begingroup\$

This is common and apparently no viable study explains exactly why. Getting the batteries to discharge evenly is essentially impossible in a 'real world' application. In my flashlight test experiment the battery closest to the bulb always discharged soonest, the other batteries discharged inconsistently sooner/later. Using rechargeable batteries and changing their position didn't affect this result. Flashlight is an old incandescent type. My experience has been the engineers that dismiss me the quickest with a question are the youngest and the least imaginative and don't ask such questions themselves, I guess because it inhibits simply getting results.

\$\endgroup\$

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.