Troubleshooting Paperwhite 4 battery

[Waylander]

Sorry if this has already been answered elsewhere. I have a PW 4 that almost a year old, and am slightly concerned about the battery. I've noticed that when the power light goes green, it's still on 96-97% or so and takes forever to get to 100%. I charged it to 100% last night, unplugged it and didn't use it for a couple of hours, and when I turned it on it had already dropped 2%. After 1-2 hours reading since last night it's down to 94%. I'm probably worrying about nothing but it this normal behaviour?

Edit: I have the frontlight on 11.

[NullNix]

Originally Posted by Waylander

Sorry if this has already been answered elsewhere. I have a PW 4 that almost a year old, and am slightly concerned about the battery. I've noticed that when the power light goes green, it's still on 96-97% or so and takes forever to get to 100%. I charged it to 100% last night, unplugged it and didn't use it for a couple of hours, and when I turned it on it had already dropped 2%. After 1-2 hours reading since last night it's down to 94%. I'm probably worrying about nothing but it this normal behaviour?

This is routine behaviour for lithium-ion batteries. The charging circuit moves further and further into the trickle-charge regime the fuller the battery gets, because the closer the battery is to full, the more likely that fast charging will cause gas production and then the lovely thing the battery engineers call "venting with flame". I think the green light means "switched to trickle charging, further charging will take ages, you might as well start reading again", but really the meaning of that light is completely opaque and always has been. Why does it flash when you turn the kindle off and when you turn it on? Search me...

The "drop" when turned off is partly the result of the Kindle actually still being on (though not doing much) even when turned "off", but mostly an illusion caused by the battery meter not being entirely accurate. It's based on examination of the output voltage of the battery, and at both ends (empty and full) very much has error bars of a few percent. Also, the battery is naturally going to self-discharge a bit when it's that full: full batteries are unstable systems and very much want to be emptier. (When the battery really starts to wear out, the meter can jump by 50% in an instant as the battery circuit suddenly realises how much less charge the battery holds than it thought it did.)

Also, honestly, a 3% drop in two hours would equate to over a *hundred hours* of reading time. This seems most unlikely: you can expect to get 20--30 hours on 11 before needing a charge. You'll never get a hundred hours of reading on a single charge even on a brand-new Kindle with the light on zero. So if anything you're seeing remarkably *low* battery drop rates (likely because the battery is so full, and it can produce more charge before reducing its output voltage much), and you'll probably see the rate of drain pick up as the battery empties.

(The most extreme Kindle that I know of in this area is the original Oasis. If you turned the light off, a fully-charged charging case could sometimes keep the thing going for *fifty* hours.)

[Waylander]

Originally Posted by NullNix

This is routine behaviour for lithium-ion batteries. The charging circuit moves further and further into the trickle-charge regime the fuller the battery gets, because the closer the battery is to full, the more likely that fast charging will cause gas production and then the lovely thing the battery engineers call "venting with flame". I think the green light means "switched to trickle charging, further charging will take ages, you might as well start reading again", but really the meaning of that light is completely opaque and always has been. Why does it flash when you turn the kindle off and when you turn it on? Search me...

The "drop" when turned off is partly the result of the Kindle actually still being on (though not doing much) even when turned "off", but mostly an illusion caused by the battery meter not being entirely accurate. It's based on examination of the output voltage of the battery, and at both ends (empty and full) very much has error bars of a few percent. Also, the battery is naturally going to self-discharge a bit when it's that full: full batteries are unstable systems and very much want to be emptier. (When the battery really starts to wear out, the meter can jump by 50% in an instant as the battery circuit suddenly realises how much less charge the battery holds than it thought it did.)

Also, honestly, a 3% drop in two hours would equate to over a *hundred hours* of reading time. This seems most unlikely: you can expect to get 20--30 hours on 11 before needing a charge. You'll never get a hundred hours of reading on a single charge even on a brand-new Kindle with the light on zero. So if anything you're seeing remarkably *low* battery drop rates (likely because the battery is so full, and it can produce more charge before reducing its output voltage much), and you'll probably see the rate of drain pick up as the battery empties.

(The most extreme Kindle that I know of in this area is the original Oasis. If you turned the light off, a fully-charged charging case could sometimes keep the thing going for *fifty* hours.)

That's a great help, thankyou.

[Junket]

Yup, normal behaviour as NullNix explained.

IIRC, solid yellow is charging, blinking yellow is charging aborted, green means trickle charge. The light will blink momentarily when pressing the power button to signal that the press registered. If you experience a startup fault you may see a blinking pattern which represents a diagnostic code (specific error message).

It's good that the charging circuit switches over to trickle before 100%, overcharging is very hard on a battery. Even float (trickle) charging is undesirable. Batteries are most stable around 62% charge. So ideally, you'd only charge them from perhaps 30% to 65%. But we routinely charge to 100% because we usually want the extended run time.

If you want to extend your batteries lifetime keep it away from extreme heat and try to avoid charging extremes. That is don't run it down to zero if you can charge it at 20 or 30% instead. And it helps a bit if you pull the charger at 95% rather than let it go to 100%.

[Waylander]

Originally Posted by Junket

Yup, normal behaviour as NullNix explained.

IIRC, solid yellow is charging, blinking yellow is charging aborted, green means trickle charge. The light will blink momentarily when pressing the power button to signal that the press registered. If you experience a startup fault you may see a blinking pattern which represents a diagnostic code (specific error message).

It's good that the charging circuit switches over to trickle before 100%, overcharging is very hard on a battery. Even float (trickle) charging is undesirable. Batteries are most stable around 62% charge. So ideally, you'd only charge them from perhaps 30% to 65%. But we routinely charge to 100% because we usually want the extended run time.

If you want to extend your batteries lifetime keep it away from extreme heat and try to avoid charging extremes. That is don't run it down to zero if you can charge it at 20 or 30% instead. And it helps a bit if you pull the charger at 95% rather than let it go to 100%.

I've been using ereaders for years and this was really helpful. Just shows you can always learn something new. Thankyou.

[BookCat]

I've been having similar issues with a PW3 which I bought used about about a month ago. But not only does the charge percentage drop soon after being unplugged, it continues dropping.

I have started a thread about this.

[binaryhermit]

Originally Posted by NullNix

venting with flame

Is this video an example of "venting with flame"?

[j.p.s]

Originally Posted by binaryhermit

Is this video an example of "venting with flame"?

Yes, it is.

Accidental examples have been captured at a conference and on an airplane in flight.

[binaryhermit]

And the infamous Galaxy Note 7...

[Junket]

Venting with flame is a rather misleading euphenism. Venting occurs when a battery gets excessively hot and the internal gas pressure in the battery exceeds 10 bar (!). Venting is a safety feature to prevent a more consequential explosion. Thermal runaway is a considerably more serious process resulting in fire or worse. A Li-ion battery is at risk of thermal runaway if it's temperature exceeds 150C. A typical cause is from metallic dust contamination during manufacturing causing an electrical short. But it could be caused by puncture, crushing, over-charging (charger fault) or a short-circuit. Thermal runaway pushes the temperature high enough to ignite lithium.

Lithium battery chemistry has an inherent potential for thermal runaway. The introduction of lithium-ion batteries was a significant improvement in safety over earlier lithium-metal batteries. Early lithium-ion batteries for instance would withstand a nail penetration. That safety margin has been eroded in recent years by the push for higher and higher energy densities. If you impaled a recent lithium-ion battery with a nail it would be a bomb.

Lithium-ion batteries still have a good safety record. A quality Li-ion battery has a failure rate better than 1:10,000,000. The ill-fated Samsung 7 recall was because their failure rate was approaching 1:200,000. Apple and Dell both recalled millions of Li-ion batteries before Samsung but Apple is apparently better at keeping molten laptops out of the news than Samsung.

I use lithium ion batteries, like everyone else. But I wouldn't allow a generic Li-ion battery or charger inside my home. Lithium fires are nasty.









Real-life video of thermal runaway

(Mobileread doesn't support youtube timestamps. Jump to 7:30 in video)

[Richwood]

When charging I have noted a difference between chargers and cables. Some combinations seem to not want to charge above 98%, even if left plugged in overnight, while others reach 100% charge level fairly rapidly on Kindles which give a % charge readout. Just due to manufacturing tolerances I would think that there is some variation in voltage out of USB chargers and cable resistances. USB cables, depending on age, have been designed for anything from 0.5 amp at 5 volts to 5.0 amps at 20 volts or 2.5 watts to 100 watts power ratings.

[Junket]

Given a decent quality i.e. brand name charger, Kindle will draw the current that it needs and automatically switch over to trickle charge when the battery approaches full charge. And will usually charge faster from a wall charger than an USB port.

So your garden variety Samsung or Apple 5W USB wall charger or something similar would be fine. Cables can make a difference, cheaper and generic cables often have such tiny internal conductors that they have limited current capacity and will charge slower. I certainly wouldn't want a counterfeit Apple charger though (e.g. off Ebay). I've measured some of those and some of them could damage your devices and arguably constitute a fire risk as well.

All that said, if you are not time-sensitive with regard to how fast Kindle charges e.g. you charge overnight, then you can extend your battery life by charging your battery slower. Not many people have chargers where you can set the output current. But you could use a 0.3A or 0.5A charger, rather than a 1A+ charger for a less harsh charging cycle.