11
\$\begingroup\$

[Note: Various answers to other questions address this, but I cannot find another question focused on just this. This is my attempt to create a definitive place for answers to this.

[If at all possible, please include references in your answer. If you're answering based on experience and your references are not handy, also giving the approximate era will be helpful since it seems that the advisibility of indefinite charging and acceptable rates may have changed over time as NiMH batteries have developed.]

If it makes a difference, this question is about commonly available modern consumer AA cells, particularly low self-discharge (LSD) cells, since all consumer cells now seem to be LSD.

NiMH batteries are generally charged with a constant current charger that stops charging at some point based on one or more of various criteria (time, voltage, -ΔV, temperature, ΔT/Δt, etc.). Here I am asking about continuous charging with no definite endpoint (e.g., applying charging current for days or weeks at a time).

Note also that I am asking only about constant current charging: this question discusses constant voltage ("float voltage") charging of NiMH batteries.

Various sources suggest that NiMH batteries may be "trickle charged" to maintain their charge state. For example, Linden and Reddy [lin02] state:

Trickle Charge. A number of applications require the use of batteries which are maintained in a fully charged condition. This is done by trickle charging at a rate that will replace the capacity loss due to self-discharge. A trickle charge at a current of between the 0.03 and 0.05C rates is recommended.... Trickle charge may be used following any of the previously discussed charging methods. (§29.5.2 p.29.27 PDF p.889)

The "previously discussed" methods are low-rate 0.1C, quick 0.3C and fast 0.5–1C rates. They go on to give a three-step charge procedure involving a 1C rate charge terminated with ΔT/Δt or -ΔV, a 0.1C topping charge terminated by timer after 0.5–1h, and:

  1. The third step is a maintenance charge of indefinite duration at a current of between the 0.05 and 0.02C rates. The battery should also be protected with a thermal cutoff device to terminate the charge so that the temperature does not exceed 60°C.

However, this source is from 2002, before modern LSD NiMH cells became available, and also capacities of consumer cells (such as AA) appear to have increased since the book was published.

A 2018 Energizer datasheet [ene18] says in its "Recommended Charging Rates" section, after discussing smart fast chargers and slow timer-based chargers, says:

Finally a maintenance (or trickle) charge rate of less than 0.025 C (C/40) is recommended. The use of very small trickle charges is preferred to reduce the negative effects of overcharging. (p.11)

Note that they don't indicate here whether by "trickle" charge they mean a continuous charge or just a low-rate charge. No definition is given in this document, but from their definition in in [ene08] it seems it could be either:

Trickle Charge:
A method of recharging in which a secondary battery is either continuously or intermittently [emphasis mine] connected to a constant current supply that maintains the battery in a fully or near full charged condition. Typical trickle charges are between 0.03C and 0.05C. (p.4)

Further on, they do mention overcharge, but seem to say that only "full" charge currents (presumably C/10 or higher?) are a problem:

Establishing the appropriate degree of overcharge for a battery-powered application is dependent on the usage scenario. Some overcharge of the battery is vital to ensure that all batteries are fully charged and balanced, but maintenance of full charge currents for extended periods once the battery has reached full charge can reduce life. (p.12)

So from all this it's not clear to me if they're saying that this overcharging will have a noticeable negative effect on the battery, but it's worthwhile if your application needs to maintain fully charged batteries, or if at C/40 current the effect is essentially negligible.

I've also looked for an Eneloop datasheet but have been unable to find one that discusses charging in any but the most cursory detail.

A final component of this question is, if some particular low current rate (C/20, C/40, whatever) is acceptable for indefinite charging, is that rate also sufficient to charge a discharged battery given enough time? For example, if an NiMH battery (a particular model or in general) can handle an indefinite C/50 charging current, will that still charge the battery to full capacity if the battery becomes discharged and the current is then applied for 60-75 hours after that? Or will such low charge rates merely overcome the self-discharge and hold the battery "steady" at its currently charged capacity?

References

Related Answers

For reference, here are some answers to other questions that also discuss continuous charging of NiMH cells. Others should feel free to add to this list if the come across ones that I've missed.

\$\endgroup\$

2 Answers 2

11
\$\begingroup\$

It's not clear to me if it's indicating that even this overcharging will have a noticeable negative effect on the battery (but it's worthwhile if your application needs to maintain fully charged batteries) of if at this current the effect is essentially negligible.

You can charge a NiMH battery indefinitely. Unlike Li-ion batteries, overcharging is not a problem (i.e. their open circuit float voltage will never rise to a level that damages the battery). Any excess energy turns the electrolyte into a gas, which then collapses under autocatalysis, releasing some heat. As long as the heat buildup is not excessive, you can charge it forever without ill effects. LSD cells are much more sensitive than non-LSD cells though, as their gas-recombination abilities are more limited, and trickle-charging might need to be so slow that it's not worth it.

A final component of this question is, if some particular low current rate (C/20, C/40, whatever) is acceptable for indefinite charging, is that rate also sufficient to charge a discharged battery given enough time?

Yes. In fact, many "dumb" chargers merely turn on trickle charging and keep it on for a set period (e.g. 40 hours) before turning it off. It will actually charge it, not just sustain it over its own self-discharge. If you have no temperature monitoring, keep it below C/40. If you want to be really safe, C/100 would work for nearly any NiMH cell (even if it would take a long time to fully charge). If the cell explicitly advertises that it is safe to trickle charge, you can charge it more quickly. Just be aware that most LSD cells have very, very limited gas recombination abilities.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ So how do you know this? Do you have some references? \$\endgroup\$
    – cjs
    Commented Jan 15, 2023 at 5:54
  • \$\begingroup\$ @cjs For gas recombination capabilities, or the specific numbers like C/40? As for determining if it's safe to trickle charge a cell at a practical rate, electronics.stackexchange.com/a/56843/177824 mentions that you really have to ask the manufacturer. \$\endgroup\$
    – forest
    Commented Jan 15, 2023 at 5:56
  • \$\begingroup\$ The specific numbers are the more important part, though the gas recombination abilities would also be of interest. (I'll go back and read §29.9 of [lin02] as well, since that discusses the gas recombination and I'd not really paid too much attention to the chemistry part of that chapter.) \$\endgroup\$
    – cjs
    Commented Jan 15, 2023 at 5:58
  • 2
    \$\begingroup\$ @cjs Unfortunately, there's really no way to determine a specific safe charge rate. It's often based on manufacturer recommendations, and what works for some cells won't work for others. If a cell is designed to be trickle-charged, you can probably charge at C/20. For an LSD cell that isn't explicitly designed for it, it might be C/100 or lower. You'd really have to test to be sure, or find out from the manufacturer. There are probably cells that are not even safe at C/200... If you're designing a charge circuit, you should use dT/dt charge termination instead, and not trickle-charge it. \$\endgroup\$
    – forest
    Commented Jan 15, 2023 at 5:59
-1
\$\begingroup\$

You have not said much about the parameters of your project, but if you are designing something intended to be good, then you probably want a real battery management system. There's no good reason not to use one ... in which case stop reading right here.

On the other hand if you enjoy tinkering with cheap consumer rubbish and trying to make it do something interesting, then it sounds like you want to go any buy a few different solar garden lights and take them apart. Measure the charge cycles those are using. Typically they use an extremely primitive charge circuit, and are designed to be as cheap as possible.

Note that certain types of consumer NiMH batteries are specifically intended for use only in solar garden lights and those are widely available once you know what to look for. I believe these are NOT low self-discharge, and indeed for this particular application self-discharge is quite a desirable feature. Given the lack of good datasheets, the only way is to test a few and put some kind of data logger on the cell temperature, in order to get an idea of what you can get away with. Insert all the normal warnings and disclaimers here regarding experimenting with batteries, chemicals, fire hazard, etc

\$\endgroup\$
1
  • \$\begingroup\$ This is not for a particular project, but a question about the general theory of how charging works so that I a) know more about how those battery management systems work.; and b) am better informed when making decisions for any particular project. Your answer helps with neither. \$\endgroup\$
    – cjs
    Commented Oct 26, 2023 at 17:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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