5 ms·

Electronics that constantly draw low amounts of power — such as some wall clocks, headlamps, or bike lights — work better with disposable alkaline batteries

Electronics that constantly draw low amounts of power — such as some wall clocks, headlamps, or bike lights — work better with disposable alkaline batteries. Alkalines release power consistently right up until they die, whereas a rechargeable battery’s voltage gets gradually lower and lower over time, slowing down the device or cutting off power prematurely. This is very wrong, and makes me doubt the article's correctness. Every NiMH discharge curve I've ever seen is much flatter than a typical alkaline curve. NiMH batteries drop down to around 1.25V pretty quick when discharging, but they stay around 1.25-1.15V for almost the entire discharge cycle. Alkaline batteries are almost dead by the time they get down to 1.2V. So the reason why NiMHs drop out early on some devices isn't because they're out of charge, or because alkaline batteries have a flatter discharge curve (they don't). It's because some devices stop working at around a 1.2V cell voltage, regardless of how much charge is left in the battery.

zokier 7y ago
1.2V cut-off is ridiculously high even for alkalines. Looking at the charts from e.g. Energizer datasheets[1], it seems like going form 1.2V cut-off to something more reasonable like 0.8V gives you whopping extra 50% service life. Alkalines are definitely not "almost dead" at 1.2V. [1] http://data.energizer.com/pdfs/en91.pdf http://data.energizer.com/pdfs/en91.pdf
- nrclark 7y ago
  that's a pretty interesting datasheet! If you assume a constant-power load like a low-dropout boost regulator, it looks like it depends a lot on the current draw. On that graph, you're right that the 200mA draw goes quite a bit farther after 1.2V. The 100mA draw is around 65-75% done at 1.2V. Presumably that exends to lower-power draws too. Thanks for linking that, it was neat. TIL
  - zokier 7y ago
    > that's a pretty interesting datasheet! If you assume a constant-power load like a low-dropout boost regulator, it looks like it depends a lot on the current draw. One of my unrealized project ideas was to create a battery life estimation calculator, something you could feed in battery configuration, some power supply details (efficiency, cutoff), and consumption figures and get an estimate of battery life. The log-log charts are not all that easy to read and have only limited number of curves available.
    - bookofjoe 7y ago
      https://www.google.com/search?q=battery+life+calculator&oq=battery+life+calculator&aqs=chrome..69i57j0l5.27687j0j8&sourceid=chrome&ie=UTF-8 https://www.google.com/search?q=battery+life+calculator&oq=b...
      - zokier 7y ago
        All of the first-page results do just dumb nominal capacity divided by current calculation, which is pretty useless. This Microchip one seems bit more clever, but it is limited to Microchip microcontrollers only and is a separate installable application, so that is no good either: https://www.microchip.com/SWLibraryWeb/product.aspx?product=nanoWatt%20XLP%20Battery%20Life https://www.microchip.com/SWLibraryWeb/product.aspx?product=...
coin 7y ago
I thought that NiMH will lose about 1% per day, thus making them a poor choice for standby devices.
- felixfurtak 7y ago
  True for older style NiMH, but the newer 'pre-charged' types like Eneloop have very low self-discharge
- pavon 7y ago
  No, after 5 years eneloop AA batteries still retain ~70% of their charge. I have been using them in remotes, bathroom scales, flashlights, and the like for over a decade and they usually last a couple years before I need to recharge them. The only place I have Alkalines any more is my emergency box, and those tend to self-corrode almost as quickly the NiMH self-discharge. I think the name brands (Energizer, Duracell) have similar performance these days as well.
entee 7y ago
From personal experience I couldn't get my wall clock (cheap Ikea one) to run on an eneloop, even after I think I charged it, without the second hand going all weird and failing. Works just fine on a disposable, what am I doing wrong? This article may be incorrect but the conclusion matches my experience.
- askvictor 7y ago
  Rechargeable NiMH cells have a voltage of 1.2v, and stay pretty much at that voltage until they're flat. Alkaline (and other disposable) cells start at 1.5v (actually a little higher), and gradually diminish in voltage over they're life. If the appliance requires more that 1.2v, then a NiMH cell will simply work. In any case, clocks tend to draw a very small amount of power over a longer time, which doesn't tend to suit NiMH cells due to a higher rate of self discharge (eneloops and other similar 'ready to use' rechargeables are much better though)
zelon88 7y ago
Partially true, but you're measuring the cells wrong. Alkaline batteries typically start around 1.5v and are advertised to be around 1.5v. A fully charged NiMH cell will be advertised at 1.2v but at peak charge will actually be around 1.4-1.5v. Even without discharging the NiMH cell at all the peak will wear off within a few hours. Furthermore, NiMH charges can go stale, trapping electrons inside the cells forever. These trapped electrons occupy cell capacity that can no longer be filled with viable electrons. Thus, a 2,000mah cell that has been left with 1,000mah worth of stale charge will likely only ever charge to 1,000mah for the rest of its life. If you measure the curve from the delta peak of a NiMH battery the curve is way more extreme than any alkaline.
rini17 7y ago
They should probably have said instead: When NiMH discharges completely in such device, it's prone to failure.
squarefoot 7y ago
It is also possible that the article mistook the cell discharge curve with their self discharge. Self discharge happens due to imperfect chemistry regardless of what is connected to the battery: you charge a cell at 100% then put it in a drawer, and after some time it falls below any usable current by itself. Usually it takes months but I had very poor cells self discharging in less than two weeks, although they could indeed deliver the advertised current in shorter times. As you pointed out, NiMH cells have a much flatter curve than alkaline ones, but they also self discharge quicker than alkalines, so for example in very low power devices such as remote controls or clocks where one would expect long times between recharges, they aren't the best option because their self discharge time is much shorter than their use time. In my kitchen I have an exception: an old electromechanical wall clock in which I swapped the alkaline cell with a NiMH one, but both use and self discharge are dealt with by a small solar panel connected in parallel with the cell, which in this case works as a voltage stabilizer too. So far about 4 years passed without changing or recharging anything, and it gets direct sunlight for a very short time every day.
p1mrx 7y ago
For any 2xAA or 2xAAA device without a center tap, you can use a 3.2 volt LiFePO₄ cell with a "0 volt" spacer. I use LiFePO₄ to run my rear bike light and a wireless thermometer, both 2xAAA devices. The total capacity is reduced because half the space is wasted, but that's not a big deal if you're only recharging a few times a year. I haven't tried this with remote controls, since the batteries last basically forever in those. Another option is to use 3.7 volt lithium ion with an LDO regulator, but I haven't tried that yet.
- bdamm 7y ago
  Fun hack! Do you have a line on cheap LiFPO4 or is it just to get a longer run time between swaps?
  - p1mrx 7y ago
    I prefer not to use disposable batteries for environmental reasons. But with NiMH, a light designed for 3V will be dim, and an LCD will have low contrast. In that sense, the effective run time is zero, so NiMH is a non-starter. I found some AAA-size LiFePO₄ cells, spacers, and a charger on AliExpress a few years ago.
- extesy 7y ago
  LDOs introduce their own inefficiency into the process so it's usually not worth the trouble.
sixstringtheory 7y ago
Slightly related, and maybe you could provide more knowledge here: common knowledge for safety critical devices, such as avalanche beacons, is to only use disposable alkaline batteries. The reasoning I've heard is that due to the flat power curve of non-alkaline batteries, if you put a nearly-dead one in the beacon and check the battery level, it will read as fuller than it really is, and you run the risk of having it die on you during a normal usage period (maybe 12 hours). With an alkaline battery, when it's close to done, it will more accurately register as such on the device.
- Gibbon1 7y ago
  Alkaline batteries start off at 1.5-1.6 volts and the voltage follows the charge reasonably linearly down to about 1.0 volts. So that's correct, alkaline batteries at 1.5V per cell are 'good' full stop. Where a rechargeable at 1.2V might or might not be. Who really knows if it charged all the way[1][2]. [1] Gallant is careful to put his nmh batteries in the the charger each night. Goofus got high and forgot. In the morning Goofus swapped his dead ones for the ones in Gallants charger. [2] Power went out between 11pm and 6am while everyone was asleep.
- tjohns 7y ago
  I have one of the few avalanche beacons that have separate modes for alkaline and NiMH batteries. (Specifically, the Barryvox Pulse.) My understanding is that "alkaline mode" measures the voltage to determine charge level, as one would expect. The "NiMH mode" instead uses a timer-based approach to estimate charge level due to the different discharge curve.
- jhayward 7y ago
  That seems like sound reasoning to me. I've spent many many hours staring at manufacturer provided curves, and countless days running my own and would absolutely use that as a rule.
q_queue 7y ago
All of it is wrong, and it's frustrating that things like this get promulgated with an authoritative voice. It seems to happen especially often about battery technology, for some reason, or else I only notice it more because I know a little bit about battery technology. My other not-favorite thing, tangentially, is how marketing for anything that stores electrical energy has gravitated to "mAH" as the sole unit. I know it's because the uninformed consumer will compare based on the bare number, but every time I see "15000 milliamp hours" I cringe. Off to drive 7 megamicromiles to work.
- Monroe13 7y ago
  https://en.m.wikipedia.org/wiki/Gell-Mann_amnesia_effect https://en.m.wikipedia.org/wiki/Gell-Mann_amnesia_effect

zokier 7y ago

1.2V cut-off is ridiculously high even for alkalines. Looking at the charts from e.g. Energizer datasheets[1], it seems like going form 1.2V cut-off to something more reasonable like 0.8V gives you whopping extra 50% service life. Alkalines are definitely not "almost dead" at 1.2V. [1] http://data.energizer.com/pdfs/en91.pdf http://data.energizer.com/pdfs/en91.pdf

nrclark 7y ago
that's a pretty interesting datasheet! If you assume a constant-power load like a low-dropout boost regulator, it looks like it depends a lot on the current draw. On that graph, you're right that the 200mA draw goes quite a bit farther after 1.2V. The 100mA draw is around 65-75% done at 1.2V. Presumably that exends to lower-power draws too. Thanks for linking that, it was neat. TIL
- zokier 7y ago
  > that's a pretty interesting datasheet! If you assume a constant-power load like a low-dropout boost regulator, it looks like it depends a lot on the current draw. One of my unrealized project ideas was to create a battery life estimation calculator, something you could feed in battery configuration, some power supply details (efficiency, cutoff), and consumption figures and get an estimate of battery life. The log-log charts are not all that easy to read and have only limited number of curves available.
  - bookofjoe 7y ago
    https://www.google.com/search?q=battery+life+calculator&oq=battery+life+calculator&aqs=chrome..69i57j0l5.27687j0j8&sourceid=chrome&ie=UTF-8 https://www.google.com/search?q=battery+life+calculator&oq=b...
    - zokier 7y ago
      All of the first-page results do just dumb nominal capacity divided by current calculation, which is pretty useless. This Microchip one seems bit more clever, but it is limited to Microchip microcontrollers only and is a separate installable application, so that is no good either: https://www.microchip.com/SWLibraryWeb/product.aspx?product=nanoWatt%20XLP%20Battery%20Life https://www.microchip.com/SWLibraryWeb/product.aspx?product=...

coin 7y ago

I thought that NiMH will lose about 1% per day, thus making them a poor choice for standby devices.

felixfurtak 7y ago
True for older style NiMH, but the newer 'pre-charged' types like Eneloop have very low self-discharge
pavon 7y ago
No, after 5 years eneloop AA batteries still retain ~70% of their charge. I have been using them in remotes, bathroom scales, flashlights, and the like for over a decade and they usually last a couple years before I need to recharge them. The only place I have Alkalines any more is my emergency box, and those tend to self-corrode almost as quickly the NiMH self-discharge. I think the name brands (Energizer, Duracell) have similar performance these days as well.

entee 7y ago

From personal experience I couldn't get my wall clock (cheap Ikea one) to run on an eneloop, even after I think I charged it, without the second hand going all weird and failing. Works just fine on a disposable, what am I doing wrong? This article may be incorrect but the conclusion matches my experience.

askvictor 7y ago
Rechargeable NiMH cells have a voltage of 1.2v, and stay pretty much at that voltage until they're flat. Alkaline (and other disposable) cells start at 1.5v (actually a little higher), and gradually diminish in voltage over they're life. If the appliance requires more that 1.2v, then a NiMH cell will simply work. In any case, clocks tend to draw a very small amount of power over a longer time, which doesn't tend to suit NiMH cells due to a higher rate of self discharge (eneloops and other similar 'ready to use' rechargeables are much better though)

zelon88 7y ago

Partially true, but you're measuring the cells wrong. Alkaline batteries typically start around 1.5v and are advertised to be around 1.5v. A fully charged NiMH cell will be advertised at 1.2v but at peak charge will actually be around 1.4-1.5v. Even without discharging the NiMH cell at all the peak will wear off within a few hours. Furthermore, NiMH charges can go stale, trapping electrons inside the cells forever. These trapped electrons occupy cell capacity that can no longer be filled with viable electrons. Thus, a 2,000mah cell that has been left with 1,000mah worth of stale charge will likely only ever charge to 1,000mah for the rest of its life. If you measure the curve from the delta peak of a NiMH battery the curve is way more extreme than any alkaline.

rini17 7y ago

They should probably have said instead: When NiMH discharges completely in such device, it's prone to failure.

squarefoot 7y ago

It is also possible that the article mistook the cell discharge curve with their self discharge. Self discharge happens due to imperfect chemistry regardless of what is connected to the battery: you charge a cell at 100% then put it in a drawer, and after some time it falls below any usable current by itself. Usually it takes months but I had very poor cells self discharging in less than two weeks, although they could indeed deliver the advertised current in shorter times. As you pointed out, NiMH cells have a much flatter curve than alkaline ones, but they also self discharge quicker than alkalines, so for example in very low power devices such as remote controls or clocks where one would expect long times between recharges, they aren't the best option because their self discharge time is much shorter than their use time. In my kitchen I have an exception: an old electromechanical wall clock in which I swapped the alkaline cell with a NiMH one, but both use and self discharge are dealt with by a small solar panel connected in parallel with the cell, which in this case works as a voltage stabilizer too. So far about 4 years passed without changing or recharging anything, and it gets direct sunlight for a very short time every day.

p1mrx 7y ago

For any 2xAA or 2xAAA device without a center tap, you can use a 3.2 volt LiFePO₄ cell with a "0 volt" spacer. I use LiFePO₄ to run my rear bike light and a wireless thermometer, both 2xAAA devices. The total capacity is reduced because half the space is wasted, but that's not a big deal if you're only recharging a few times a year. I haven't tried this with remote controls, since the batteries last basically forever in those. Another option is to use 3.7 volt lithium ion with an LDO regulator, but I haven't tried that yet.

bdamm 7y ago
Fun hack! Do you have a line on cheap LiFPO4 or is it just to get a longer run time between swaps?
- p1mrx 7y ago
  I prefer not to use disposable batteries for environmental reasons. But with NiMH, a light designed for 3V will be dim, and an LCD will have low contrast. In that sense, the effective run time is zero, so NiMH is a non-starter. I found some AAA-size LiFePO₄ cells, spacers, and a charger on AliExpress a few years ago.
extesy 7y ago
LDOs introduce their own inefficiency into the process so it's usually not worth the trouble.

sixstringtheory 7y ago

Slightly related, and maybe you could provide more knowledge here: common knowledge for safety critical devices, such as avalanche beacons, is to only use disposable alkaline batteries. The reasoning I've heard is that due to the flat power curve of non-alkaline batteries, if you put a nearly-dead one in the beacon and check the battery level, it will read as fuller than it really is, and you run the risk of having it die on you during a normal usage period (maybe 12 hours). With an alkaline battery, when it's close to done, it will more accurately register as such on the device.

Gibbon1 7y ago
Alkaline batteries start off at 1.5-1.6 volts and the voltage follows the charge reasonably linearly down to about 1.0 volts. So that's correct, alkaline batteries at 1.5V per cell are 'good' full stop. Where a rechargeable at 1.2V might or might not be. Who really knows if it charged all the way[1][2]. [1] Gallant is careful to put his nmh batteries in the the charger each night. Goofus got high and forgot. In the morning Goofus swapped his dead ones for the ones in Gallants charger. [2] Power went out between 11pm and 6am while everyone was asleep.
tjohns 7y ago
I have one of the few avalanche beacons that have separate modes for alkaline and NiMH batteries. (Specifically, the Barryvox Pulse.) My understanding is that "alkaline mode" measures the voltage to determine charge level, as one would expect. The "NiMH mode" instead uses a timer-based approach to estimate charge level due to the different discharge curve.
jhayward 7y ago
That seems like sound reasoning to me. I've spent many many hours staring at manufacturer provided curves, and countless days running my own and would absolutely use that as a rule.

q_queue 7y ago

All of it is wrong, and it's frustrating that things like this get promulgated with an authoritative voice. It seems to happen especially often about battery technology, for some reason, or else I only notice it more because I know a little bit about battery technology. My other not-favorite thing, tangentially, is how marketing for anything that stores electrical energy has gravitated to "mAH" as the sole unit. I know it's because the uninformed consumer will compare based on the bare number, but every time I see "15000 milliamp hours" I cringe. Off to drive 7 megamicromiles to work.

Monroe13 7y ago
https://en.m.wikipedia.org/wiki/Gell-Mann_amnesia_effect https://en.m.wikipedia.org/wiki/Gell-Mann_amnesia_effect