I know that neither keeping battery full or discharged is good for battery life, but I want to know a basic answer: which one of the two bad alternatives is better? Full or empty.

This is given that self discharge bellow 0% is not an issue as I would plug the devise in after it's off for some time.

I have a bad habit of either keeping my phone battery full (by constantly looking for charger) or having it almost empty (by charging when it's dead for 10min and then going after business). So I want to know which of the two habits will wear my phone battery faster? Keeping constantly charged to near 100, or barely charged.


  • \$\begingroup\$ The problem with a "basic" answer is that you're not the one charging the battery; the phone does it. \$\endgroup\$ – Ignacio Vazquez-Abrams Mar 26 '17 at 7:50
  • \$\begingroup\$ For your phone, I think trying to maximize life is counterproductive. Just use it until it gets low and then recharge it. If you are storing lithium rechargeable batteries rather than using them, store them at like 50%. For other chemistries I think it is better to store them fully charged. \$\endgroup\$ – mkeith Mar 26 '17 at 7:53
  • \$\begingroup\$ It depends on the battery. NiCd batteries lasted better when discharged, but that will rapidly kill a lead-acid battery. Likewise a Li-Ion will probably never recover if left for any length of time while discharged. \$\endgroup\$ – Brian Drummond Mar 26 '17 at 11:30
  • \$\begingroup\$ Depends on chemistry. Ignacio mentioned a phone. You didn't. More details please. || LiIon - full is better. Lead acid - full is essential. NimH - full is OK. Empty may be bit I've seen them die. NiCd - they will rapidly self entry . Empty is OK. \$\endgroup\$ – Russell McMahon Mar 26 '17 at 12:24
  • \$\begingroup\$ Russell, I updated the question, It's Lion I'm interested. Can you provide research proff? And make an answer \$\endgroup\$ – Aurimas Mar 26 '17 at 13:17


Since your phone will prevent the battery from over-discharging due to self-discharge and you're keeping it charged up, being on the low-side is the best way to keep your battery in tip-top shape. The lower in voltage you go, the less strain you place on the structure within the cell. Much of the data presented in the link doesn't extend down to the 0-10% range because batteries will virtually last forever down in that range since they're hardly being stressed at that point.

All of the tables and graphs there point to extremely long battery lifetimes if you keep the battery at a very low state of charge (soc). Just storing your battery at room temperature at 100% soc results in 20% loss in a year vs 4% at 40% soc.

Table 3: Estimated recoverable capacity when storing Li-ion for one year at various temperatures:

Temp 40% charge 100% charge
0°C 98% 94%
25°C 96% 80%

Too be clear, there's nothing wrong with keeping a li-ion battery at 0% as long as you don't let it drop below 0% (3.0V or 3.3V depending on the manufacturer). Manufacturer's state that damage starts to happen at 2.0 or 2.5V for lithium ions. Your phone has a protection circuit on it that would prevent your battery from working at all if it got down to this voltage. Since 0% is defined much higher than that, unless you let your phone sit idle for a long time at 0%, your battery will live a very plush life between 0-10% soc.

| improve this answer | |
  • \$\begingroup\$ "Since your phone will prevent the battery from over-discharging due to self-discharge and you're keeping it charged up, being on the low-side is the best way to keep your battery in tip-top shape." - did u mean prevent from discharging beyond 0% ? \$\endgroup\$ – Aurimas Mar 27 '17 at 2:57
  • \$\begingroup\$ @Aurimas Yes. Prevent it from going significantly below 0%. \$\endgroup\$ – horta Mar 27 '17 at 21:01
  • \$\begingroup\$ This is the best answer so far \$\endgroup\$ – Aurimas Mar 27 '17 at 23:09

For storage purposes, a Li-ion battery should be charged to 60-70% capacity. This reduces the damage done to the electrode material over time while still allowing enough charge so that it will hopefully not be completely discharged due to self-discharge when you get around to using it.

For charge cycle purposes, a Li-ion battery should be charged during its constant voltage cycle with a voltage about 150mV less than the voltage specified in the datasheet. This will result in only about 85% to 90% runtime versus full voltage, but an increase charge cycle count of 20% to 40%.

Of course, if the battery is in a device then all this is moot since the device manufacturer decides what "full" and "empty" mean for the battery; you're only responsible for connecting the power supply to the device.

| improve this answer | |
  • \$\begingroup\$ Hi, thanks, but I already know that. I want to know which of the two bad alternatives is better long term? Updated the question to make it more specific (although I asked the same question before also) \$\endgroup\$ – Aurimas Mar 26 '17 at 10:51
  • 1
    \$\begingroup\$ Ignacio - I long said LiIon is best at 70-80% = top of CI ramp, but am now of opinion that lower is better. About 3.6-3.8V range. Cell changes from about 5-10% capacity to say 30-40% somewhere around there. \$\endgroup\$ – Russell McMahon Mar 26 '17 at 12:26
  • \$\begingroup\$ I read that longest shelf life was near 2/3 which most MFG use as a guide for bulk shipments. I recall anything more than 3.9V causes self heat and accelerates aging. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Mar 26 '17 at 15:05
  • \$\begingroup\$ I think international regulations require LiIon batteries to be shipped at 35% (only applies to air). \$\endgroup\$ – mkeith Mar 26 '17 at 20:55

Neither. I've wondered about that myself, so I talked to the battery expert friends I have:

In either state, you risk that the extreme concentration differences in the chemicals that make up the cell lead to a diffusion of elements to regions they shouldn't be in. So, go for something like a 1/3 charge or so to maximize shell life.

However, for bog normal Li batteries, it probably won't matter much, and you should most likely, avoiding the shorting->explosion risk, store them discharged.

, or I constantly overcharge it

No, you don't. You're not the one in charge of the charging cycles, your phone is. Let it do its job, it does protect the battery as far as possible in trade off with fast charging and long usage. It's more likely your battery will deteriorate through normal thermal effects than due to the amount you charge it to.

| improve this answer | |
  • \$\begingroup\$ Hi, thanks, but I already know that. I want to know which of the two bad alternatives is better long term? \$\endgroup\$ – Aurimas Mar 26 '17 at 10:51

According to your comments, if you only have full/empty options, I'd go for full.

That's because Li-Ion batteries usually have control/safety circuits that only work between full and empty. So, if you store an empty battery, self-discharge may get you below this circuit's minimum operating voltage, which would then block battery charging. As a consequence your battery would be unusable, even if the chemistry was still fine.

On a full battery, self-discharge wouldn't be that much of a problem, except that its chemistry wouldn't like it and would potentially damage your battery earlier. That very circuit should stop charging before the chemistry is 100%, effectively setting the "battery full" below that.

After question edit: If you're not storing the battery, I don't really see why you'd keep an empty battery in a system. It would have no purpose. I'd still keep it full, so it'd serve as backup power, or I'd remove the battery from the system (infinite lifespan, since you'd never have to replace the battery). That would also avoid battery malfunction problems and power waste due to occasional recharge to 1%.

| improve this answer | |
  • \$\begingroup\$ Good points about discharge! However the question is still unanswered, I updated the question. \$\endgroup\$ – Aurimas Mar 26 '17 at 14:41
  • \$\begingroup\$ @Aurimas Updated to reflect updated question. \$\endgroup\$ – Ronan Paixão Mar 26 '17 at 17:25
  • \$\begingroup\$ It makes sense, yes, thanks. I still need the answer to the original question, with some reasearch or scientific reasoning \$\endgroup\$ – Aurimas Mar 26 '17 at 20:47

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