65

I have a bag of about 50 non-rechargeable AA batteries (1.5 V) that I have collected over the years. I bought a multimeter recently and would like to know the best way to test these batteries to determine which ones I should keep and which I should toss.

Sometimes a battery will be useless for certain high-power devices (e.g. children's toys) but are still perfectly suitable for low-power devices such as TV remote controls. Ideally, I'd like to divide the batteries into several arbitrary categories:

  • As-new condition (suitable for most devices)
  • Suitable for low-powered devices such as remote controls
  • Not worth keeping

Should I be measuring voltage, current, power or a combination of several of these? Is there a simple metric I can use to determine what to keep and what to toss?

LeopardSkinPillBoxHat
  • 757
  • 1
  • 6
  • 8
  • 2
    Everything above 1.5 V is near new, and I'd toss anything below 1.2 (maybe 1.3V). – George Herold Aug 20 '14 at 13:46
  • 4
    Recycle them and enter a competition as person who most looks like Bob Dylan - you are bound to win (and collect a few dollars) to buy some new batteries with. – Andy aka Aug 20 '14 at 13:56
  • 2
    I agree that this endeavor is inefficient from a purely monetary, or perhaps even environmental, point of view, but regardless, I have enjoyed learning the answer(s)! – Jenn D. Aug 20 '14 at 20:31
  • @Andy aka. I am still laughing. Bob Dylan is still alive. – Steve Mar 04 '19 at 23:51
  • @Steve - for the record, that *is* a picture of Bob Dylan :-) And he is still alive, though doesn't look like that anymore. – LeopardSkinPillBoxHat Mar 05 '19 at 00:07
  • Toss the batteries if they say Super Heavy Duty because they will be full pf rice. – Audioguru Jul 11 '21 at 18:31

6 Answers6

61

**WARNING: Lithium Ion cells **

  • While this question relates to non-rechargeable AA cells it is possible that someone may seek to extend the advice to testing other small cells.
    In the case Of Li-Ion rechargeable cells (AA, 18650, other) this can be a very bad idea in some cases.

  • Shorting Lithium Ion cells as in test 2 is liable to be a very bad idea indeed.
    Depending on design, some Li-Ion cells will provide short circuit current of many times the cell mAh rating - eg perhaps 50+ amps for an 18650 cell, and perhaps 10's of amps for an AA size Li-Ion cell.

  • This level of discharge can cause injury and worst case may destroy the cell, in some uncommon cases with substantial release of energy in the form of flame and hot material.

AA non-rechargeable cells:

1) Ignore the funny answers

Generally speaking, if a battery is more than 1 year old then only Alkaline batteries are worth keeping. Shelf life of non-Alkaline can be some years but they deteriorate badly with time. Modern Alkaline have gotten awesome, as they still retain a majority of charge at 3 to 5 years.

Non brand name batteries are often (but not always) junk.

Heft battery in hand. Learn to get the feel of what a "real" AA cell weighs. An Eveready or similar Alkaline will be around 30 grams/one ounce. An AA NiMH 2500 mAh will be similar. Anything under 25g is suspect. Under 20g is junk. Under 15g is not unknown.

2) Brutal but works

Set multimeter to high current range (10A or 20A usually). Needs both dial setting and probe socket change in most meters.

Use two sharpish probes.

If battery has any light surface corrosion scratch a clean bright spot with probe tip. If it has more than surface corrosion consider binning it. Some Alkaline cells leak electrolyte over time, which is damaging to gear and annoying (at least) to skin.

Press negative probe against battery base. Move slightly to make scratching contact. Press firmly. DO NOT slip so probe jumps off battery and punctures your other hand. Not advised. Ask me how I know.

Press positive probe onto top of battery. Hold for maybe 1 second. Perhaps 2. Experience will show what is needed. This is thrashing the battery, decreasing its life and making it sad. Try not to do this often or for very long.

  • Top AA Alkaline cells new will give 5-10 A. (NiMH AA will approach 10A for a good cell).

Lightly used AA or ones which have had bursts of heavy use and then recovered will typically give a few amps.

Deader again will be 1-3A.

Anything under 1 A you probably want to discard unless you have a micropower application.

Non Alkaline will usually be lower. I buy ONLY Alkaline primary cells as other "quality" cells are usually not vastly cheaper but are of much lower capacity.

Current will fall with time. A very good cell will fall little over 1 to maybe 2 seconds. More used cells will start lower and fall faster. Well used cells may plummet.

I place cells in approximate order of current after testing. The top ones can be grouped and wrapped with a rubber band. The excessively keen may mark the current given on the cell with a marker. Absolute current is not the point - it serves as a measure of usefulness.

3) Gentler - but works reasonably well.

Set meter to 2V range or next above 2V if no 2V range.

Measure battery unloaded voltage.

New unused Alkaline are about 1.65V. Most books don't tell you that.

Unused but sat on the shelf 1 year + Alkaline will be down slightly. Maybe 1.55 - 1.6V

Modestly used cells will be 1.5V+

Used but useful may be 1.3V - 1.5V range

After that it's all downhill. A 1V OC cell is dodo dead. A 1.1V -.2V cell will probably load down to 1V if you look at it harshly. Do this a few times and you will get a feel for it.

4) In between.

Use a heavyish load and measure voltage. Keep a standard resistor for this.
SOLDER the wires on that you use as probes. A twisted connection has too much variability.

Resistor should draw a heavy load for battery type used.
100 mA - 500 mA is probably OK.

Battery testers usually work this way.

5) Is this worth doing?

Yes, it is. As well as returning a few batteries to the fold and making your life more exciting when some fail to perform, it teaches you a new skill that can be helpful in understanding how batteries behave in real life and the possible effect on equipment. The more you know, the more you get to know, and this is one more tool along the path towards knowing everything :-). [The path is rather longer than any can traverse, but learning how to run along it can be fun].

MicroservicesOnDDD
  • 2,557
  • 1
  • 13
  • 48
Russell McMahon
  • 147,325
  • 18
  • 210
  • 386
  • 9
    I use Russell's method #2. Remember to return it to a voltage scale (and move the probe away from the 10A/20A position) afterward or you may get a surprise when you go to test voltage next time. – Spehro Pefhany Aug 20 '14 at 14:39
  • 1
    Method 4 - why not buy or scavenge a battery box and solder on an appropriate resistor. You could even fit banana sockets for the meter leads. – Chris H Aug 20 '14 at 15:57
  • 1
    @ChrisH Your battery box idea has some merit. I usually find myself with a few batteries to check and seek out the nearest multimeter, but a test box at battery central would be useful. I use Alkaline batteries in high power camera flashes for weddings etc when I want to be able to dump a set (usually too hot to touch) and load another asap. Batteries used this way are not reused in the flash but may be useful elsewhere. If I do not use a set of batteries to exhaustion then the remaining capacity varies and being able to test is usefil. – Russell McMahon Aug 21 '14 at 13:43
  • 1
    for #4, what do we do once we measure the Voltage under load? Compare to voltage under load of a new battery? – cloneman Jan 20 '16 at 04:39
  • 1
    @cloneman Similar voltages to in "3" in my answer. - somewhat lower and voltages fade more quickly - especially for more exhausted batteries. Experience will tell you what to expect after a few times. If I have many batteries I tend to test them and put them in groups in say 0.1V steps. Depending how many and what I want them for I may combine groups. For eg Flash use for important event I'd aim for new cells for key moments and little used one for less crucial sections. At eg a wedding I always(try very hard to remember to) swap in a new-new set of cells just before bride enters church. – Russell McMahon Jan 20 '16 at 16:44
9

Apparently, you can test an AA battery by dropping it and see how much it bounces:

m.Alin
  • 10,638
  • 19
  • 62
  • 89
  • 1
    Have you tried it? You MAY be in for s surprise :-) – Russell McMahon Aug 20 '14 at 14:54
  • 1
    I've done it. Hasn't failed me yet. Having said that, it doesn't involve a multimeter :P – Ignacio Vazquez-Abrams Aug 20 '14 at 15:06
  • 7
    As all those links are to videos, can you give a text summary of how to test an AA battery by dropping it? – Michael Aug 20 '14 at 16:20
  • 2
    Hold the battery facing upwards an inch above a level, rigid surface and then let go. Used batteries will have a lot of bounce and fresh batteries will have very little. – Ignacio Vazquez-Abrams Aug 20 '14 at 17:00
  • 2
    You're going to have to add a text explanation of why this works - some property of the electrolyte? – pjc50 Aug 20 '14 at 18:43
  • 1
    "Using a multimeter"... – Mindwin Remember Monica Aug 20 '14 at 21:14
  • 1
    The best explanation I've heard is that gas generation during the consumption of the battery materials deforms the casing, pushing it away from the solid contents. – Ignacio Vazquez-Abrams Aug 21 '14 at 00:03
  • 1
    It really seems to work. This is evil... – Ricardo Aug 21 '14 at 00:46
  • Isn't it fantastic. I was stunned. I have a large quantity of variably old batteries here (100s). Many are rechargeable - it will be interesting to see if there is any correlation between charge and bounce with them and how it varies if at all with age and capacity and type and ... . I have some very old low capacity NiCd batteries. One of those when dropped was almost alive. It would bounce and spin around the long axis as if powered. Very repeatable. I did NOT get a chance to compare meter test results but I will sometime. Awesome stuff :-). – Russell McMahon Aug 21 '14 at 13:49
  • 3
    @Mindwin place multimeter face down on bench. drop battery on multimeter. – Pete Kirkham Aug 21 '14 at 15:26
  • @PeteKirkham `"Sarcasm: Not Appreciated" - Omega Supreme`. www.youtube.com/watch?v=2vBgKj4106c – Mindwin Remember Monica Aug 22 '14 at 15:00
  • @PeteKirkham Also, the bounciness factor is better evaluated if object drops onto a hard immovable (relative to object size) surface XD – Mindwin Remember Monica Aug 22 '14 at 15:02
  • 1
    From the J. of Materials Chemistry (2015,3, 9395-9400): "The increase in coeff. of restitution (COR) correlates to the formation of a percolation pathway of ZnO within the anode of the cell... the anode gel consists of discrete Zn particles that can move within the gel matrix... once the anode densifies it becomes a stiff ceramic core... and the COR saturates. It is shown that determination of the COR... through a simple bounce test provides an accurate measure of the state of charge of the battery". Full text available (as PDF) at: http://www.osti.gov/pages/biblio/1183288 – MV. Apr 03 '16 at 19:25
  • Forgot to add: the title of the journal article is "The relationship between coefficient of restitution and state of charge of zinc alkaline primary LR6 batteries", the OID is 10.1039/C5TA01576F and for some reason the full text is available at that page from the OSTI of US Dept. of Energy. – MV. Apr 03 '16 at 19:29
  • My best bet as to why this works is that the mass of a battery shifts as it gets used. Electrons flow from the anode of the battery (the flat side on AAs), through the circuit, and to the cathode. When an electron leaves the anode, one of the atoms that makes it up becomes a positive ion and breaks off. When an electron arrives at the cathode, it forms an atom with one of the positive ions near the cathode and attaches onto it. There is a bridge between the cathode and anode that allows the ions to flow. This means that over time, the battery becomes top heavy which is probably why it bounces. – John Leuenhagen Oct 29 '17 at 04:24
6

Some meters have a battery test mode - a voltmeter with a load in parallel. One of mine (a wavetek meterman) does. Mine is ancient but a similar model is designed to draw ~150mA in 1.5V mode, and 5mA in 9V mode. Using this mode you can push down to around 1.2, even 1.1V for remote controls, lower still for a few things (I had a logitech cordless keyboard that stopped working when the 2 AAs got down to 0.65V each, and an LED torch which also must have had a boost converter on the input as it ran off a single 1.5V cell and could extract the last charge from that).

To be honest the the easier approach is: take battery out of demanding device, put in box to be used in undemanding device. It's not just about current but about input voltage - a cheap weather station I have gets fussy below about 2x1.25V - so lasts next to no time on NiMH.

Chris H
  • 2,331
  • 11
  • 18
2

Load the battery with about 6 ohm resistor (that will load the batter by somewhat 250 mA, somewhat like a bike light would draw.). Connect the resistor in parallel to the multimeter. The multimeter should show the voltage enough for a typical device to operate (most of them stop working at around 1.2 V or about). You can wait for a few seconds to be sure.

Such battery can power the device beyond doubt so why to throw it away? If you must be always sure, take a spare new batter with you.

h22
  • 881
  • 6
  • 14
1

I like Russell's #4 answer. What we think of as dead batteries are really batteries with internal resistance built up to limit the available voltage and power.

Russell's #2 answer is a direct method of testing internal resistance. By shorting the terminals and measuring the current, you are really measuring internal resistance because the shorted current is proportional to the internal resistance.

But #4 is better. You are measuring the voltage drop across a resistor, which approximates the way that the battery will be used. The higher the internal resistance, the lower the voltage drop across the external resistor.

I use my batteries twice. I use them first in the camera, which draws a lot of current, and needs fresh batteries. Then I use the batteries in my little transistor radio, which draws very little current and can still use batteries with high internal resistance. (depleted)

Greenonline
  • 2,064
  • 7
  • 23
  • 38
Baruch Atta
  • 192
  • 5
1

I always use the "brutal" method (from Russell's answer). It's most precise because it also tells you the "fitness" of the battery, not only voltage or short-term power.

Plug the probe into the 10/20A slot and test the battery for 1 max 2 seconds. New batteries show 14A+ for me. Used are about 8-10A. If they are 2-4A, you can still use them in remote controls or digital alarm clocks for 1-2 years. Anything below 2 is most likely dead.

Works also for rechargeable batteries but they might get damaged over the time. Testing the voltage never worked for me though.

Community
  • 1
Jonkman
  • 11
  • 1
  • Maybe better limit the "brutal" method to cases where you expect the battery to be not so new (lower current => lower risk)? – Vi0 Aug 20 '14 at 20:03
  • I also have old NiMH batteries that are hard to charge and hard to discharge: short circuit current starts at 300 mA, slowly going up to 1.5A in two-three minutes (if keep shorted)... – Vi0 Aug 20 '14 at 20:05
  • Don't throw away those "dead" cells -- use the "Vampire" light, otherwise known as the "Joule Thief". Many toys and motorized products use cells hard, and there's a lot left when the motor peters out. I even grade my used cells and put equivalent ones in series to squeeze the most juice out of them. But place your battery holder in a plastic bag, because the more you deplete a cell, the higher the pressure, and the more likely it is to leak, causing corrosion. Night lights work for weeks. – MicroservicesOnDDD Jul 10 '21 at 21:45