Electronics notes/Batteries: Difference between revisions

From Helpful
Jump to navigation Jump to search
(14 intermediate revisions by the same user not shown)
Line 211: Line 211:




{| border="1" cellpadding="5" cellspacing="0"  
{| border="1" cellpadding="5" cellspacing="0" style="font-size:small"  
|- style="background:#eee"  
|-
!   
!   
! Lead acid
! Lead acid
Line 284: Line 284:
|  
|  


|- style="background:#eee"
|-  
!   
!   
! Lead acid
! Lead acid
Line 397: Line 397:




|- style="background:#eee"
|-  
!   
!   
! Lead acid
! Lead acid
Line 509: Line 509:




|- style="background:#eee"
|-  
!   
!   
! Lead acid
! Lead acid
Line 763: Line 763:
:: basically doesn't have the overheating problems that angry pillows can. Also has a longer lifetime
:: basically doesn't have the overheating problems that angry pillows can. Also has a longer lifetime
:: both reasons why cars, and grid power storage, use these.
:: both reasons why cars, and grid power storage, use these.


* Lithium Sulfur
* Lithium Sulfur
Line 774: Line 773:
* Lithium Air
* Lithium Air
** http://en.wikipedia.org/wiki/Lithium_air_battery
** http://en.wikipedia.org/wiki/Lithium_air_battery
* Lithium-thionyl chloride (Li-SOCl<sub>2</sub>), a.k.a. LTC
:: nonrechargeable




Line 1,043: Line 1,046:
Some of the most commonly sold sizes:
Some of the most commonly sold sizes:
* AA  (13.5–14.5mm x 50.5mm)
* AA  (13.5–14.5mm x 50.5mm)
:: a.k.a. R6


* AAA  (10.5mm x 44.5mm)
* AAA  (10.5mm x 44.5mm)
:: a.k.a. R03{{verify}}


* 9V - also known as 6LR61, PP3 (from a series where others are obsoleted), 006P, NEDA 1604 and IEC 6F22 (many of which refer to specific chemistries, but are not always used consistently), E-block (in a few places)
* 9V - also known as 6LR61, PP3 (from a series where others are obsoleted), 006P, NEDA 1604 and IEC 6F22 (many of which refer to specific chemistries, but are not always used consistently), E-block (in a few places)
Line 1,060: Line 1,065:


* The letter-letter-number style are often according to IEC 60086-3
* The letter-letter-number style are often according to IEC 60086-3
** The first letter indicates chemistry, commonly L (Alkaline, ~1.5 V), C (Lithium, 3.something V), or S (Silver, ~1.5 V)
** The first letter indicates chemistry, commonly
*** L (Alkaline, ~1.5 V)
*** C (Lithium, 3.something V), or
*** S (Silver, ~1.5 V)
** If there's an R, it means round
** If there's an R, it means round
** The numbers relate to size.
** The numbers relate to size.


* '''The longer codes''' will more directly mention diameter and thickness, e.g.
* '''The longer codes''' will more directly mention diameter and thickness, e.g.
:: CR2032 (lithium, round, 20mm diameter, 3.2mm thick),  
:: e.g. CR2032 is lithium, round, 20mm diameter, 3.2mm thick; CR2477 is lithium, round, 24mm diameter, 7.7mm thick
:: SR516 (silver, round, 5.8 mm diameter, 1.6 mm thick),
:: e.g. SR516 (silver, round, 5.8 mm diameter, 1.6 mm thick),
 


* '''The shorter codes''' are a little less direct, e.g.  
* '''The shorter codes''' are a little less direct, e.g.  
:: LR44/SR44 are 11.6mm diameter and 5.4mm thick. See also [https://en.wikipedia.org/wiki/Button_cell#Package_size this table]
:: LR44/SR44 are 11.6mm diameter and 5.4mm thick. See also [https://en.wikipedia.org/wiki/Button_cell#Package_size this table]
:: A number have multiple names, like LR44 is basically the same as LR357, and AG13, and sometimes a few others.
:: A number have multiple names, like LR44 is basically the same as LR357, and AG13, and sometimes a few others.
:: This runs into the confusing area of producer-specific or country-specific naming, compatible replacements, and/or same size but different chemistries, which ''might'' be compatible <!--(e.g. silver is 0.05V more than alkaline{{verify}} but most things are designed for it or don't care
:: This runs into the confusing area of manufacturer-specific or country-specific naming, compatible replacements, and/or same size but different chemistries, which ''might'' be compatible <!--(e.g. silver is 0.05V more than alkaline{{verify}} but most things are designed for it or don't care


LR41, AG3 - 7.9mm diameter, 3.6mm thick  
LR41, AG3 - 7.9mm diameter, 3.6mm thick  
Line 1,097: Line 1,106:




B disappeared completely (order of 45V is not something anything else really needeD)
B disappeared completely (order of 45V is not something anything else really needed)


the size of the original A
the size of the original A


CR2  (CR17355, 5046LC, DLCR2)




Line 1,221: Line 1,233:
In a direct sense, battery leaking is due to too much pressure.
In a direct sense, battery leaking is due to too much pressure.


There are multiple reasons for that, so one question is which reasons gets there ''first''.
There are multiple reasons for pressure buildup, so one question is which reasons gets there ''first''.




External heat is bad for batteries, partly because it directly leads to pressure, partly because it helps some of the chemistry go faster.
External heat is bad for batteries, partly because it directly leads to pressure,
partly because it helps some of the chemistry go faster - potentially making it get ''even'' warmer.




More usually the reason that you have a bunch of cells together. If one of them goes empty and another is not, it may reverse-bias that cell.  
A similar issue comes up when you have a bunch of cells together.  
If one of them goes empty and another is not, it may reverse-bias that cell.  
 
This often ends up electrolysing the electrolyte, which creates gas, and pressure.  
This often ends up electrolysing the electrolyte, which creates gas, and pressure.  
If that gas cannot be vented (which seems true for ''some'' parts of a battery, in compact designs),
If that gas cannot be vented (which seems true for ''some'' parts of a battery, in compact designs),
Line 1,234: Line 1,249:




Some things make this more likely:
Some things make this more likely to happen:
* it is a bad idea to mix batteries of different state of charge, so a good idea to replace all at once
* it is a bad idea to mix batteries of different state of charge, so a good idea to replace all at once
: This is less relevant if you charge batteries regularly, or if you use the device so will notice when it stops working, which is usually far before one is ''truly'' flat)
: This is less relevant if you charge batteries regularly, or if you use the device so will notice when it stops working, which is usually far before one is ''truly'' flat)
Line 1,242: Line 1,257:


* it's a bad idea to replace batteries with batteries of different capacity (more likely to be listed in rechargables)
* it's a bad idea to replace batteries with batteries of different capacity (more likely to be listed in rechargables)
: because it's alsmost a given one will be empty before the others
: because it's almost a given one will be empty before the others




...but arguably the most common reason for leaking is forgetting about a device for years, particularly in equipment stored on a shelf.
...but arguably the most common reason for leaking is forgetting about a device for years.




Line 1,425: Line 1,440:
===Battery myths and not-so-myths===
===Battery myths and not-so-myths===


===="Leaking is caused by too low a charge"====
 
===="You should always discharge your phone completely before charging"====
 
<!--
<!--


That is certainly part of a common reason,
The argument is often one of the many [[memory effect]] variations,
but neither the only reason,
but the thing is that the damage model is very different for LiPo.
nor the direct reason.


See the leaking section above.
This effectively comes down to




Also, a cell ''can'' go completely flat without damage,
and in you're lucky, a battery may go the same way.
-->
-->


===="You shouldn't leave your phone on the charger"====
===="Closing apps makes battery life longer"====
 
<!--
<!--
No, until yes.




This is a few different ideas:
Most apps only do things when they are currently on screen,
* you may overcharge


* parasitic load would drop charge enough every few minutes for charge to kick in, which is harsh on the battery
Unless the app specifically chooses to have a job in the background,
it will not use the processor while put in the background.


* full charges uses up charge cycles
And as you've noticed, starting most apps takes a second or two,
that that is usually ''actual work'' being done.




So that idle power is negligible, and the starting up is not,
you're easily using a little more power.
Yet the amount more isn't a lot either, unless you're really doing it constantly.


Phones have charge controllers. They do not allow overcharge.


In fact, pretty much ''every'' lithium charger does this.
The one reason this can be true is that,  
Because they have to. And because there are jellybean ICs that do this for you, because you need to be very little more than a linear regulator to do that (see e.g. TP4056).
while background apps ''should'' also not take a lot of RAM,
once you run out of RAM (and the cheapest phones may have this more easily),
there is more work shifting things around while switching apps,
and that can amount to a little more unnecessary


-->


That said, a lithium battery that is nearly full does not take the last bit of charge as easily.
While you can charge much of it with higher current,
the typical way to treat a lithium battery nicely is to charge the last bit more slowly.


Phones may do this intentionally, or might even be set up to do this only during the night.
===="Fast charging damages LiPo"====


<!--
It doesn't have to, but it can.




The second point, that flip-flopping near fullness (or trickle/float charge) would be bad is somewhat overstated.
And while dumber devices may do that, it doesn't apply much to phones in that phone (and laptop) chargers will typically allow the charge to fall a few percent before trying to top it off again.




The idea that there is a fixed number of charge cycles that you might be using up
Up to some point, "fast charging" means "we go faster than the simple way to be safe, yet we cared more about the balance to e.g. speed and lifetime"
is an old one, and won't go away because it is only ''partially'' nonsense.


At some point, "fastest charging" means "we ignored all long term because you wanted something faster today"




That said, even if it's not unsafe, some care may make the battery last longer (some recent phones take more care themselves).
The one you get today (and long term if everyone buys like you)
depends on which numbers on the box you care most about
''as well as'' how good the engineering is.




-->
<!--
The fuller it gets, the slower the current is ''if'' you care about long life, and/or giving heating damage,
and ''some'' form of this is even implemented in many of the cheapest chargers.


-->


====Store batteries in the fridge====
So slow charging ''at the end'' is important, but at the same less crucial early on.


<!--
Figuring out a reasonable charging speed at different parts of the process is a thing.
The main idea is that self-discharge will be slower, and there is truth to this.  




The typically mentioned discharge rates for chargeable batteries are at room temperature,
If you care to push the battery harder, but be aware of it,
and you can ''roughly'' assume that every 10-15 degrees C doubles/halves that rate.  
one reasonable answer is to stick a temperature sensor on the battery - it will let us know if we're going too fast,  
and that will adapt to the battery becoming less happy later.


Then again, some battery chemistries are not fond of cold (and even less fond of hot) situations.
Also there's the risk is condensation leading to corroded contacts, so do store it dry.


"time from 0 to 80%" is an industry convention to report,
because it's a decent middle ground to compare things on,
that doesn't chase designers to value the last difficult part.


Also, the self-discharge of non-rechargable (a.k.a. primary) batteries is ''significantly'' lower - shelf life of years rather than months, so for them it's barely relevant.


-->
-->




====On memory effect, voltage depression, and such====
{{stub}}


===="Leaking is caused by too low a charge"====
<!--


"Memory effect" seems to be the "[[Ground loop]]" of batteries - a vague name that groups multiple distinct issues,
The answer to that varies per chemistry.
which has lead to lots of confusion that is hard to clear up.
Some are actually fine with being flat.




'''Memory effect''' points at the idea that "specific charge/discharge patterns limits how much it can charge/discharge the next time"
What even those are not fine with is being charged in ''reverse''.


The thing is that there are multiple reasons you can limit the capacity, like, you know, damaging the battery,
When you have a battery of cells, when one behaves unlike the others, a similar situation happens.
but the suggestion behind memory effect is either that
* you can avoid this happening as quickly with specific charge/discharge patterns
* you can recover some of that capacity with specific charge/discharge patterns


The memory effect claim is often more specific than the above.
We should make a list, but the most common seems to be that
a cell/battery won't easily charge/discharge beyond the level it was last charged/discharged to.
: If true, the implication would be that a a battery should always be discharged fully, before being charged fully
: If not observed, the effective capacity would quickly lessen over time.
: Interestingly, this idea that charge/discharge patterns lessen battery life '''is for a good part, superstitious bunk - but with a core of truth'''.


So leaking doesn't always happen.
Devices with a single cell may be a little happier.


But in terms of devices being unhappy with leaks, you should probably assume it will,
and never long-term store batteries inside devices.


Then there is '''voltage depression''', which means the voltage seems to drop off faster than it did last time.
See the leaking section above.


It's very easy, and not necessarily wrong, to assume that means the rechargable battery doesn't hold as much capacity.
-->


But actually, '''even voltage depression isn't really the right term''',
===="You shouldn't leave your phone on the charger"====
because (depending on details) it is possible to have voltage curve drop faster
<!--
even though the capacity you get out is still much the same.




This is a few different ideas:
* you may overcharge, which is harsh on the battery


'''Need for a better name?'''
* parasitic load would drop the apparent charge just low enough for charge to kick in, which is sort of the same as the previous point


When memory effect and voltage depression point at rather specific things, we may want a broader name.
* full charges uses up charge cycles


Also, it forces a broader view that lets us point out that even if there is value in micro-managing each battery,
you can also make such micro-management pointless if it gets abused anyway.


Smartphones have charge controllers.
They do not allow charging to too high a voltage.


I guess just name it by the symptom we care about most - reduced capacity - and not by specific hypotheses.
In fact, pretty much ''every'' lithium charger does this.
Because they have to.
And because there are [[jellybean ICs]] that do this for you,
because the simplest (if somewhat blunt) solution is little more than a linear regulator (see e.g. TP4056).




Some real causes behind reduced capacity include:
'''That said'''
* cell age
: Different kinds of batteries have noticably different storage lifespans
:: to be fair, there are specific reasons that we should probably end up naming
:: but this seems to come largely from chemical reactions other than that of charge/discharge{{verify}}, which varies with quality{{verify}}
::: and, depending on the chemistry, also with state of charge, temperature, etc., which is why e.g. lithium has specific instructions


* cell use
* a lithium battery that is nearly full does not take the last bit of charge as easily,
: Some age with use more more, others age regardless of use
: so once it is getting fullish, you should start charging more slowly
: You could try to express this in expectable amount of charge-discharge cycles.
: Not everything does that equally much
: whether you do partial or whole charges is often less relevant than that you use it at all
:: e.g. while lithium can be stored longer, it has fewer uses (apparently using them will slowly cause side reactions in the electrolyte which traps lithium, reducing the amount of potential lithium ions that we need. This might be worse with abuse, but happens even if taken care of)


* some are smarter than others.
: If you've ever seen your battery go down to 90ish percent even on the charger, that's a ''feature'' in the sense that it prolongs battery life
: Phones might even intentionally plan to hit the 100% at your alarm time.
: certainly not everything does that.


* state of charge
:: Some vary with the stage of charge it is kept at, others much less so


* cell damage due to overcharging
The idea that there is a fixed number of charge cycles that you might be using up is an old one,  
:: and this varies with chemistry - e.g. lithium is less happy to be left on a trickle charge than many others, which is why a careful charger may ''refuse'' to trickle charge litium, refuse to continue charge until its SoC has fallen a bunch (and in terms of battery life this is preferred behaviour), and why if you ''really'' want to leave with the fullest of battery you can time that cycle (some phones do this, based on your alarm)
and won't go away because it is only ''partially'' nonsense.




* cell damage do to over-discharge
That said, even if it's not unsafe, some care may make the battery last longer (some recent phones take more care themselves).


* cell damage from deep discharge in a ''battery''
:: that is, even if a single cell is resilient to go to 0V, when you combine multiple cells and some go flat, this often amounts to reverse charging other cells, which is damaging


* cell damage due to high temperatures
--------------
** from working in a hot environment
** from overly fast charging
** from overly fast discharging


* means of storage
:: varies with chemistry - in some cases this is negligible, in some cases it is noticable (though often only starts being a problem when the charge is fairly low)


* specific chemical effect
It depends on the type of charging it's doing.  
:: it has been pointed out that sintered plate NiCd, which if discharged to ''precisely'' the same point would show this effect.


This comes from much older common wisdom.
Say, while you would leave e.g. a car battery on a trickle to always keep it charged,
the properties of other batteries, including LiPo, suggest it's best not to.




The dumbest charging might still trickle a LiPo, but this is uncommon, in part because even the cheapest ICs know how to back off.


Misdiagnosis:
And the easiest improvement is is that: back off and do nothing. Let it discharge for a bit before even trying to charge again.
* treating over-charge voltage as indicative
:: if you pick a battery from the charger, it will settle from the charge voltage to its real voltage within minutes or hours. If you look and that voltage as indicating SoC, then it's easy to conclude "see, it lost 20% of its capacity straight away" just because you called it early.
:: note that most devices's battery charge state detection is fairly dumb. Keeping track of energy in and energy out is done on fancier devices, but basic ones just use voltage, meaning this 'just off the charger effect' necessarily means misreporting
:: note that cheap/slow NiCD and NiMH chargers typically still do a trickle charge (which usually overcharge a little), so batteries fresh from the charger will do this a least a little


* "Lithium is damaged by the cold"
And you can assume phones have some of the most complex charging management you have around you.
:: Probably based on noticing it reported as empty more quickly.
:: But temperature makes chemistry go slower, so ''any'' battery will perform worse -- and with lithium's (often more careful) monitoring it may report as empty more quickly. This  "chemical reactions go more slowly" is a temporary effect, though.




<!--
(Note that that backoff is also why many phones will show you in the ninety-precents even while connected, and this is in many senses correct behaviour if you care more about the minutes of ''total'' lifetine more than the minutes of charge today. 
It would not surprise me if some phones actually reported you at 100% at ~95% of actual charge, soo that it becomes a "for your purposes" rather than a "...technically" metric)


Myths, or overstated
* you should do a full discharge and charge cycle between uses
:: if anything, this leads to more voltage depression. More so in lithium -- in fact you could use them ''way'' longer than the point they will report as empty -- but its protection cuts you off because doing so is damaging.
::


* NiMH has no memory effect, NiCd does
-->
 
====Store batteries in the fridge====
 
<!--
The main idea is that self-discharge will be slower, and there is truth to this.


* Lithium has no memory effect, NiMH does


The typically mentioned discharge rates for chargeable batteries are at room temperature,
and you can ''roughly'' assume that every 10-15 degrees C doubles/halves that rate.


Notes:
Then again, some battery chemistries are not fond of cold (and even less fond of hot) situations.  
* There ''is'' something to be said for conditioning
Also there's the risk is condensation leading to corroded contacts, so do store it dry.
** ...but the common "fully charge and discharge it multiple times" suggestion often does more bad than good. It's largely nonsense, often has negligible effect, and in multi-cell batteries and packs, ''complete'' discharges are more likely to ''damage'' some of the cells due to reverse biasing some.




Also, the self-discharge of non-rechargable (a.k.a. primary) batteries is ''significantly'' lower - shelf life of years rather than months, so for them it's barely relevant.


: They're fine to tell the difference between nearly empty and probably full, but even that idea is fundamentally limited when charging is involved
-->
-->




====On memory effect, voltage depression, and such====
{{stub}}




See also:
"Memory effect" seems to be the "[[Ground loop]]" of batteries - a vague name that groups multiple distinct issues,
* http://www.dansdata.com/gz011.htm
which has lead to lots of confusion that is hard to clear up.


* http://www.resrchintl.com/memory_effects.html


====Lithium - thermal runaway and angry pillows====
'''Memory effect''' points at the idea that "specific charge/discharge patterns limits how much it can charge/discharge the next time"
<!--


Note that 'lithium batteries' is a broad category.
The thing is that there are multiple reasons you can limit the capacity, like, you know, damaging the battery,
While the bulk of what you've handled is probably LiPo, there are others with different quality and uses.
but the suggestion behind memory effect is either that
For example, LiFePo behaves less excitingly, and is currently mostly seen in sturdier setups. And cars.
* you can avoid this happening as quickly with specific charge/discharge patterns
* you can recover some of that capacity with specific charge/discharge patterns


The memory effect claim is often more specific than the above.
We should make a list, but the most common seems to be that
a cell/battery won't easily charge/discharge beyond the level it was last charged/discharged to.
: If true, the implication would be that a a battery should always be discharged fully, before being charged fully
: If not observed, the effective capacity would quickly lessen over time.
: Interestingly, this idea that charge/discharge patterns lessen battery life '''is for a good part, superstitious bunk - but with a core of truth'''.


Lithium in general is partly dangerous just because it can move more current - anything that can move more current (regardless of the chemistry)
tends to be better at making more heat, and heat is bad for most any battery.


{{comment|(Things like NiCd and NIMH absolutely get warm if you pull out energy quickly,
but its higher internal resistance means that most things short of a short won't really do it.
You put energy in too fast it'll also not be happy, but most chargers won't try.
Serious battery packs of this type are often specifically engineered for it)}}


Then there is '''voltage depression''', which means the voltage seems to drop off faster than it did last time.


It's very easy, and not necessarily wrong, to assume that means the rechargable battery doesn't hold as much capacity.


That potentially leads two or three things, that are potentially independent.
But actually, '''even voltage depression isn't really the right term''',
because (depending on details) it is possible to have voltage curve drop faster
even though the capacity you get out is still much the same.




One is that the electrolyte gets warm enough to decompose.
In a direct sense that just means means it becomes gas.
Which has more volume, so swells up.


This needs high temperature, but in theory this can happen just by getting quite warm quite often,
'''Need for a better name?'''
without any other effects.


When memory effect and voltage depression point at rather specific things, we may want a broader name.


Discharge one ''way'' too quickly and it ''will'' become a pillow [https://youtu.be/0tGK1nqXr28?list=RDCMUCtM5z2gkrGRuWd0JQMx76qA&t=737]
Also, it forces a broader view that lets us point out that even if there is value in micro-managing each battery,
Usually that's not done by design, ins
you can also make such micro-management pointless if it gets abused anyway.




I guess just name it by the symptom we care about most - reduced capacity - and not by specific hypotheses.


Another option is thermal runaway, a more general term meaning "got hot so quickly it made itself worse".
The beginnings of thermal runaway still has a change of not running
away very far, because once the decomposition blows it up - which happens faster with more heat -


Some real causes behind reduced capacity include:
* cell age
: Different kinds of batteries have noticably different storage lifespans
:: to be fair, there are specific reasons that we should probably end up naming
:: but this seems to come largely from chemical reactions other than that of charge/discharge{{verify}}, which varies with quality{{verify}}
::: and, depending on the chemistry, also with state of charge, temperature, etc., which is why e.g. lithium has specific instructions


It also depends a little on whether swelling up actually reduces the effectiveness - this is true in some designs,  
* cell use
and makes thermal runaway unlikely.
: Some age with use more more, others age regardless of use
: You could try to express this in expectable amount of charge-discharge cycles.
: whether you do partial or whole charges is often less relevant than that you use it at all
:: e.g. while lithium can be stored longer, it has fewer uses (apparently using them will slowly cause side reactions in the electrolyte which traps lithium, reducing the amount of potential lithium ions that we need. This might be worse with abuse, but happens even if taken care of)




* state of charge
:: Some vary with the stage of charge it is kept at, others much less so


If it happens faster than that, it is at risk of igniting itself.
* cell damage due to overcharging
:: and this varies with chemistry - e.g. lithium is less happy to be left on a trickle charge than many others, which is why a careful charger may ''refuse'' to trickle charge litium, refuse to continue charge until its SoC has fallen a bunch (and in terms of battery life this is preferred behaviour), and why if you ''really'' want to leave with the fullest of battery you can time that cycle (some phones do this, based on your alarm)


It matters that what it decomposed into is fuel and oxygen. Because ignite that, it's self-sustaining. Heat, fuel, and oxygen is all a fire needs, and all three are inside.  You can submerse it in water and it won't stop until it's done.


If it got ignited for other reasons like you throwing it into an exiting fire (striking it with a hammer also seems to work, stabbing it in the right way might but is not a guarantee), but the more interesting
* cell damage do to over-discharge


* cell damage from deep discharge in a ''battery''
:: that is, even if a single cell is resilient to go to 0V, when you combine multiple cells and some go flat, this often amounts to reverse charging other cells, which is damaging


* cell damage due to high temperatures
** from working in a hot environment
** from overly fast charging
** from overly fast discharging


* means of storage
:: varies with chemistry - in some cases this is negligible, in some cases it is noticable (though often only starts being a problem when the charge is fairly low)


 
* specific chemical effect
It also depends a little on whether it manages to rip a hole.
:: it has been pointed out that sintered plate NiCd, which if discharged to ''precisely'' the same point would show this effect.








Misdiagnosis:
* treating over-charge voltage as indicative
:: if you pick a battery from the charger, it will settle from the charge voltage to its real voltage within minutes or hours. If you look and that voltage as indicating SoC, then it's easy to conclude "see, it lost 20% of its capacity straight away" just because you called it early.
:: note that most devices's battery charge state detection is fairly dumb. Keeping track of energy in and energy out is done on fancier devices, but basic ones just use voltage, meaning this 'just off the charger effect' necessarily means misreporting
:: note that cheap/slow NiCD and NiMH chargers typically still do a trickle charge (which usually overcharge a little), so batteries fresh from the charger will do this a least a little


* "Lithium is damaged by the cold"
:: Probably based on noticing it reported as empty more quickly.
:: But temperature makes chemistry go slower, so ''any'' battery will perform worse -- and with lithium's (often more careful) monitoring it may report as empty more quickly. This  "chemical reactions go more slowly" is a temporary effect, though.


"So is a pillow dangerous?"


If it is still in the process of being extremely hot, yes.
<!--


After that, though, it won't spontaneously combust.  
Myths, or overstated
* you should do a full discharge and charge cycle between uses
:: if anything, this leads to more voltage depression. More so in lithium -- in fact you could use them ''way'' longer than the point they will report as empty -- but its protection cuts you off because doing so is damaging.
::


But if you ''help'' it combust, combust it ''will''
* NiMH has no memory effect, NiCd does


So it's dangerous enough to careful, but not dangerous enough to live in fear.
* Lithium has no memory effect, NiMH does




The reason comes from chemistry. The electryte in various Lithium chemistries decomposes into oxygen.  
Notes:
Oxygen is good at making things very on fire, but needs ignition, and fuel.
* There ''is'' something to be said for conditioning
** ...but the common "fully charge and discharge it multiple times" suggestion often does more bad than good. It's largely nonsense, often has negligible effect, and in multi-cell batteries and packs, ''complete'' discharges are more likely to ''damage'' some of the cells due to reverse biasing some.


Fuel




Some abuse may cause heat and puff and look dramatic, but .
: They're fine to tell the difference between nearly empty and probably full, but even that idea is fundamentally limited when charging is involved
-->


Neither will necessarily lead to thermal runaway -- ''though'' the higher density ones in RC may be at some higher risk.


''That said'', these pillows are not happy, so while panic is rarely necessary, getting rid of them is a good idea.




Hitting it with a hammer might actually be worse, because there's a change that there will be a spark.  
See also:
* http://www.dansdata.com/gz011.htm


* http://www.resrchintl.com/memory_effects.html


Thermal runaway, in general, is when heat makes things actively worse.
====Lithium - thermal runaway and angry pillows====
: For example, a LED that is warm lowers resistance, which means more current, more heat, lower resistance, etc. until it's burned. In a LED, this happens almost instantly.
<!--




In lithium batteries the same thing is a series of chemical cause and effect, and is not actually that easy to cause.
Note that 'lithium batteries' is a broad category.
While the bulk of what you've handled is probably LiPo, there are others with different quality and uses.
For example, LiFePo behaves a lot more sturdily and less excitingly, which is why it is used in cars and larger scale energy storage.


Aside from the overly fast discharge (which you can protect from, though cheap devices don't),
the more likely cause is damage and abuse.


But even then, merely-moderate abuse only leads to decomposing of the electrolyte - which leads the cell to puff up (but that needs its own section).
'''Options?'''
If more serious, it can catch fire.


One of the main features of most lithium variants is that it has lower resistance, meaning it can deliver higher currents.


Lithium ion is densely packed with energy, this will not stop until something separates its parts,
Anything that can move more current tends to be better at making more heat (outside it and inside),
be it puffing up, or fire.  
so there is an argument that if you don't need it, it's safer to stick to things like NiMH.


{{comment|(Things like NiCd and NIMH absolutely get warm if you pull out energy quickly,
but its higher internal resistance means that most things won't even try. Even a dead short is less worrying.


Remember that [https://en.wikipedia.org/wiki/Fire_triangle fire triangle] thing on how to stop a fire?
Similarly, put energy in too fast it'll also not be happy, but most chargers won't try.  
At this point, all three are inside (the heat will heat will decompose the electrolyte into oxygen),
Serious battery packs of this type are often specifically engineered for it)}}
and in a cover that should mostly even protect it from water.
You can ''submerse'' it and it will happily/angrily go on.
It will basically not go out until it's done.




This is why
Yes, lithium is also higher density, meaning you can pack in more capacity per weight and per volume, but not by that much compared to NiHM,  
* poor quality lithium
so while you can make this argument for fancy laptops or drones, a lot of low-use devices like wireless keyboards or whatnot,
* not having enough charge/discharge protection on it,
small NiMH versus small lithium barely makes near-zero practical difference.
* not having a temperature sensor,
...can all contribute to making a dramatic little fire.
Actually exploding takes an unusual situation, but is technically possible.






While there were some serious-brand mishaps when lithium batteries were new,
'''What happens to lithium when abused'''
larger brands have learned their lesson (via lawsuits) so these days
these days it's mostly cheap devices - sometimes off-brand phones, but probably largely from things like cheap vapes.


Many battery packs come with protection that should limit discharge speed - but enough of them don't.


Thermal runaway in a phone or vape may be manageable.
High current potentially leads two or three things, that are potentially independent, but worrying if not.
If


* the electrolyte gets warm enough to decompose.
: The electrolyte in various Lithium chemistries decomposes into oxygen. Oxygen is good at making things very on fire, but needs ignition, and fuel.
: So in a direct sense that just means means it becomes gas. Which has more volume, so swells up.
:: This happens most easily under very high temperature (such as that caused by overly-high-current discharge)
:: in theory this can also happen by getting moderately warm quite often - and that can happen without any other effects.
: If you remove the protection and then discharge one as fast as you can, it ''will'' become a pillow [https://youtu.be/0tGK1nqXr28?list=RDCMUCtM5z2gkrGRuWd0JQMx76qA&t=737]


Thermal runaway in a scooter or e-bike is more dramatic.
* thermal runaway, a more general term meaning "got hot so quickly it made itself worse" (e.g. LEDs will also do this, but undramatically)
: The beginnings of thermal runaway still has a chance of not running away very far - because once the heat causes decomposition and blows it up, the capacity to output current lessens quickly and makes thermal runaway much less likely. This happens to vary a bunch with the design of battery.
: The higher density, lower-internal-resistance ones in RC may be at some higher risk.


* self ignition
: this requires a perfect combination of the above two: fast enough to still be hot enough to ignite what is now there
: this because what LiPo internals decompose to is, essentially, fuel (the lithium in XXX form) and oxygen (from the typical).
: Ignite that, and it's actually self-sustaining. [https://en.wikipedia.org/wiki/Fire_triangle Heat, fuel, and oxygen is all a fire needs], and all three are now inside.  You can submerse this package in water and it won't stop until it's done (or enough water can actually make it inside - small punctures won't do that because of overpressure).


For a large part, certification will help here.
But resellers online may make it very hard to recognize you are getting a lowest-bidder battery without certification.


This means it's dangerous to
: throw it into an existing fire
: striking it with a hammer (presumably because this can set up a good short, and create an immediate spark inside?)
: stabbing it in the right way (harder to guarantee bad outcome, but maybe don't try)




A battery gets a little warm during charge (and dischrage) is completely normal,
Aside from the overly fast discharge, another likely cause of becoming a pillow is damage/abuse.
but if it's ''hot'', it is on its way to becoming a pillow, and ''sometimes'' to something more dramatic.  
 


But even then, merely-moderate abuse only leads to decomposing of the electrolyte - which leads the cell to puff up (but that needs its own section).
If more serious, it can catch fire.




Line 1,774: Line 1,822:




Hydrogen fluoride




'''"I found a pillow in my device. Is it dangerous?"'''


If it is still in the process of being extremely hot, yes.


After that, though, it won't spontaneously combust.


But if you ''help'' it combust, combust it ''will''


'''Do all lithium ion batteries turn into angry pillows?'''


All can, many may not.
So these pillows are not happy,
and while panic is rarely necessary, getting rid of them is a good idea.
It's dangerous enough to careful, but not dangerous enough to live in fear.




The cause varies, but the puffing up itself is (mainly) the the electrolyte decomposing {{comment|(in this class of battery ''largely'' into lithium and oxygen{{verify}})}}.


That's usually from heat, external or from high draw.
---




Lithium ion is densely packed with energy, this will not stop until something separates its parts,
be it puffing up, or fire.
This is why
* poor quality lithium
* not having enough charge/discharge protection on it,
* not having a temperature sensor,
...can all contribute to making a dramatic little fire.
Actually exploding takes an unusual situation, but is ''technically'' possible.
While there were some mishaps when lithium batteries were new,
that happened to more serious brand laptops - from companies who have learned their lesson (also via lawsuits)
so these days it's mostly cheap devices that you should watch - sometimes off-brand phones,
but probably largely from specific things like cheap vapes,
or a no-brand scooter or e-bike.
For a large part, certification will help here.
But resellers online may make it very hard to recognize you are getting a lowest-bidder battery without certification.


There may always be a little decomposition going on,
varying with use, ambient temperature, charger dumbness, and other things.


Many cells will die and be disposed of for other reasons, but you could probably get any pack to swell if you really wanted to,
by using it way out of spec.


Yet packs used more intensively see more decomposition,
'''Do all lithium ion batteries turn into angry pillows?'''
as may long steady use, and it can even happen in storage.


All can, many may not.


So you should probably ''not'' assume your device will not want to become larger inside.
The cause varies, but the puffing up itself is (mainly) the the electrolyte decomposing {{comment|(in this class of battery ''largely'' into lithium and oxygen{{verify}})}}.
Or that fancier hardware is immune - I've seen varied Apple devices bulging after a decade.


That's usually from heat, external or from high draw.




"...Angry? Is this dangerous? I've heard people calling them bombs"
There may always be a little decomposition going on,
varying with use, ambient temperature, charger dumbness, and other things.


There are two risks.  
Many cells will die and be disposed of for other reasons, but you could probably get any pack to swell if you really wanted to,
by using it way out of spec.


One is thermal runaway, which means it getting so hot that it catches fire inside,  
Yet packs used more intensively see more decomposition,
and because there is oxygen inside, it is a self contained fire
as may long steady use, and it can even happen in storage.




So you should probably ''not'' assume your device will never want to become larger inside.
Or that fancier hardware is immune - I've seen various Apple devices bulging after a decade.


The oxygen is why it swells, and releasing pure oxygen into a fire is going to be a short amount of intenseness.




Electrically, their capacity goes down, and their internal resistance may go up.  
Electrically, their capacity goes down, and their internal resistance may go up.


You might even get a few uses out of a somewhat swollen battery, and it may get a little warmer when you do,
You might even get a few uses out of a somewhat swollen battery, and it may get a little warmer when you do,
Line 1,839: Line 1,911:




====Lithium thermal runaway====
 
 
Hydrogen fluoride


-->
-->

Revision as of 18:16, 14 April 2024

⚠ This is for beginners and very much by a beginner / hobbyist

It's intended to get an intuitive overview for hobbyist needs. It may get you started, but to be able to do anything remotely clever, follow a proper course or read a good book.

Some basics and reference: Volts, amps, energy, power · batteries · resistors · transistors · fuses · diodes · capacitors · inductors and transformers · ground

Slightly less basic: amplifier notes · varistors · changing voltage · baluns · frequency generation · Transmission lines · skin effect

And some more applied stuff:

IO: Input and output pins · wired local IO · wired local-ish IO · ·  Various wireless · 802.11 (WiFi) · cell phone

Sensors: General sensor notes, voltage and current sensing · Knobs and dials · Pressure sensing · Temperature sensing · humidity sensing · Light sensing · Movement sensing · Capacitive sensing · Touch screen notes

Actuators: General actuator notes, circuit protection · Motors and servos · Solenoids

Noise stuff: Stray signals and noise · sound-related noise names · electronic non-coupled noise names · electronic coupled noise · ground loop · strategies to avoid coupled noise · Sampling, reproduction, and transmission distortions

Audio notes: See avnotes


Platform specific

Arduino and AVR notes · (Ethernet)
Microcontroller and computer platforms ··· ESP series notes · STM32 series notes


Less sorted: Ground · device voltage and impedance (+ audio-specific) · electricity and humans · power supply considerations · Common terms, useful basics, soldering · landline phones · pulse modulation · signal reflection · Project boxes · resource metering · SDR · PLL · vacuum tubes · Multimeter notes Unsorted stuff

Some stuff I've messed with: Avrusb500v2 · GPS · Hilo GPRS · JY-MCU · DMX · Thermal printer ·

See also Category:Electronics.

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

(Knowledgeable help would be much appreciated)


Pedantically, batteries are (parallel and/or serial) combinations of multiple cells, but in practice people use 'batteries' to refer to individual cells as well.


A cell's voltage is fixed, by its chemistry.

The difference can matter to your projects, in that charging batteries can be more involved than charging cells (particularly with lithium).


Batteries in circuits

Internal resistance/impedance

Real-world voltage sources can be modeled as acting like an ideal voltage source in series with a resistor, in that it will only move so much current (and will get warm in the process).

In the case of batteries, it's part of why you can't empty the battery instantly. Which is mostly a good thing.


It's not a literal resistor shoved in there to make your life harder. In batteries this comes mostly from the battery chemistry - and also the manufacturing and sizing.

This effective resistance is called the equivalent series resistance (ESR) (and some variations like internal resistance, internal impedance, series resistance, and series impedance.)


The ESR value is typically below 1Ω, and can be engineered to be much smaller. For example,

AAs may be on the order of 100 mΩ (varies with actual chemistry)
Lithium can more easily be on the order of 10mΩ (verify)
9V batteries may be 1Ω or 2Ω
...which limits their current capacity and means higher voltage drop, e.g. making them a potentially poor choice for certain DIY electronics projects/kits
car batteries are aimed at being on the order of 20mOhm(verify) -- because they are primarily there for the starter motor, which draws dozens of amps (...from 12V and must itself be rather less than an Ohm - you can figure that out via V=IR)


Notes:

  • ESR varies with temperature, because they are chemical reactions.
For example, an AA may be 0.1Ω at 30℃, 0.2Ω at 0℃, and 1Ω at -40℃.
  • ESR varies with age - ESR will rise with age and abuse
  • ESR varies with state of (dis)charge
    • highest when discharged. Often lowest when half charged (verify), somewhat higher than the lowest point when fully charged (verify). How bad this is, and the curve, varies a bunch with battery chemistry and design.


  • Putting cells in series (for higher-voltage batteries) effectively adds their ESRs.
  • ESR is relevant to
    • how much current you can still draw
    • how much the voltage will drop while drawing that much
    • how warm the battery will get while drawing that much
  • (If you need lower ESR than the battery you want, but only for intermittent use, then capacitors (ESR of ~10 mΩ or ~1 mΩ for some supercapacitors) are good cheats, but they pack much less energy per volume, so their increased current capacity is usually only useful as bypass capacitor or other transient loads)


See also:

Battery safety in circuits

In-circuit charging

Quick comparison

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

There are roughly four generally usable battery types that easily scale to most needs: NiCd, NiMH, Lithium-based and Lead-based. Alkaline exists both in unchargable form and in special rechargable form, but the rechargable one is not very common.

Each have their own characteristics, energy density (staying power per volume or weight), behaviour when charged and stored, and so on. Important to note is that capacity depends (in varying degrees) on use and calendar time. It can also depend on whether it is a more restricted commercial battery or a more-risk-acceptable industrial battery


The following table should serve as general comparison.

General notes:

  • Many figures here are (necessarily) approximate (and most of it needs verification)
    • There is generally variation between cheap, 'extra durable', industrial and specialist batteries.
    • figures involving weight and volume can vary by specific design, and figures here may be based on product sheets, best cases, or whatnot.
    • ...and by shape, e.g. cylindrical may last longer than prismatic (flat) used eg. in thin laptops)
  • Some things are specified per cell (while some things you think of as batteries are multiple cells)
  • self-discharge
non-rechargeable batteries tend to have much lower self-discharge than rechargables do
Higher self-discharge usually means you should top off every few months, or store on a trickle charge (...and in some cases specifically not)
some (e.g. nickel-based, lithium-based) have a higher self-discharge in the first day than after that(verify)
  • Overcharge tolerance - overcharging can be avoided for various battery types, with smarter loaders.
NiCd/NiMH are occasionally overcharged because it was/is the common consumer tech and there are many dumb constant-current chargers out there
Lithium is much less resistant, chargers have to be smart to not treat the cells badly, and packs often have basic protection as well.
  • On lithium: Lithium is a group - with some specific variations (some not even non-rechargable), more so if you don't stick to consumer products.
  • On charge speed: for example, if the capacity of a battery is 2Ah, 0.2C refers to a charge current of 0.4A.
e.g. for commercial NiCd/NiMH batteries, overnight charges are usually done at 0.1C, fast charges at 1C, and very fast at 2C or 3C.
  • Memory effect is a vague (and somewhat misleading) name, usually for "reasons your battery doesn't hold as much capacity" - see also the section on it


Lead acid NiCd (aka NiCad)
(Nickel Cadmium) 		NiMH
(Nickel Metal Hydride) 		LiIon
(Lithium-ion) 		LiPo
(Lithium-ion polymer) 		LiFePO4
(Lithium iron phosphate) 		Re-usable alkaline Alkaline (non-chargable) NiZn (Nickel Zinc)
regularly seen in Solar, car, wheelchair, emergency lighting, UPS Standby, Consumer electronics Consumer electronics Portable computing Portable computing Cars, grid storage Consumer electronics Consumer electronics
(Single-)Cell voltage ~2.1V 1.25V 1.25V 3.6V ~3.7V 3.2V 1.5V 1.5V 1.6V
Life
(cycles) (order of magnitude before capacity noticably less) 		300-600 500-2000 300-600(verify) ~1000?
a few years after manufacture? 		~1000?
a few years after manufacture? 		/10 years 3000+
a dozen years 		NA / 1 600(verify)
label cost
(price per Wh)
(varies anyway) 		Lowish Average Average High High Low
Cost/cycle
(relative to others) 		Average Low Average Average/High Average/High High NA (High for its single cycle)
Lead acid NiCd (aka NiCad)
(Nickel Cadmium) 		NiMH
(Nickel Metal Hydride) 		LiIon
(Lithium-ion) 		LiPo
(Lithium-ion polymer) 		LiFePO4
(Lithium iron phosphate) 		Re-usable alkaline Alkaline (non-chargable) NiZn (Nickel Zinc)


Environmental impact (relative to others) Moderate (better if recycled) High Low Low(ish) Low(ish) Lower than basic alkaline?
Storability Good (should be stored full) Long (any state of charge) Average Average (?) Average (store half full?) ?
Self-discharge (rough order, can be better) ~5%/month
(some variants up to 40%/month(verify)) 		15%/month(verify) 15%/month(verify) 7%/month(verify) (half from protection circuit) 10%/month(verify) (half from protection circuit) 3%/month <1%/month
Lead acid NiCd (aka NiCad)
(Nickel Cadmium) 		NiMH
(Nickel Metal Hydride) 		LiIon
(Lithium-ion) 		LiPo
(Lithium-ion polymer) 		Re-usable alkaline LiFePO4
(Lithium iron phosphate) 		Alkaline (non-chargable) NiZn (Nickel Zinc)


Ruggedness Average (avoid full discharge) High Average (avoid full or fast discharge) Average (if not abused; has care circuitry) ?
Overcharge tolerance High (but too-fast charging causes damage) Average Low Low Low Average NA
Undercharge tolerance Low


Heat sensitivity Average Average/low High(ish) High High(ish) Lower than other lithium(verify) Average
Safety Better than LiIon (e.g. doesn't decompose at high temperatures)


Lead acid NiCd (aka NiCad)
(Nickel Cadmium) 		NiMH
(Nickel Metal Hydride) 		LiIon
(Lithium-ion) 		LiPo
(Lithium-ion polymer) 		Re-usable alkaline Alkaline (non-chargable) NiZn (Nickel Zinc)
Charge speed (slow,fast,max?)
in fraction of capacity 		0.1C - 1C - 3C ? 0.1C - 1C - 2C ? NA


Discharge speed4 (continuous, burst)
Internal resistance
(order of magnitude)
Internal resistance change


Capacity per weightest.
Wh/kg and
kg/Ah12 at 12V 		Heavy
30-50Wh/kg
0.34kg/Ah 		Heavy
35-80Wh/kg
0.34kg/Ah 		Average
60-120Wh/kg
0.17kg/Ah 		Light
100-160Wh/kg
.06kg/Ah 		Light
130-200Wh/kg
 Average/light
90–160 Wh/kg 		Average
80Wh/kg2 		Average
60Wh/kg 		Light
100Wh/kg
Capacity per volumeest.
Wh/l 		80-90 80-150(verify) 60-300?(verify) ~300(verify) 100(verify)
Voltage depression Low High (exercise helps) Average (exercise helps) Low? (but wears over time) Lower than LiIon High NA
Maintenance
(best every x months) 		3-6, but little 1-2 2-3, but more details unnecessary unnecessary NA


Lead acid NiCd (aka NiCad)
(Nickel Cadmium) 		NiMH
(Nickel Metal Hydride) 		LiIon
(Lithium-ion) 		LiPo
(Lithium-ion polymer) 		LiFePO4
(Lithium iron phosphate) 		Re-usable alkaline Alkaline (non-chargable) NiZn (Nickel Zinc)

est. - Necessarily an estimate. The product may tell you more (but advertisement may lie).

2 - Given to decrease over time whether used or not. After about three years after manufacture it may not be very useful for the same application, lasting perhaps half as long. Manufacturers of batteries, gadgets, or laptops do not like to admit this, partly because it means that if you need a new battery and the model is not available, you're screwed.

3 - Measured to when capacity dips below about 80%. Varies a lot because of the differences between battery design, as well as the difference between perfect use (listed) and practical use (may be a third of listed value). 'Over 1000 cycles' tends to be an ideal claim true for large batteries. The three numbers represent energy inefficient design, usual design, and energy efficient design. (eg. prismatic cell phone NiMH, AA-cell NiMH and D-cell NiMH) and are approximate. Note that NiMH 'twice as long as NiCd' claims are based on the larger capacity, not the amount of cycles.

4 - Discharge speed also relates to how the chemistry reacts. For example, for Lead batteries you may quickly get a decent voltage on only the physical surface, which slowly distributes and may mean you won't be able to start your car later. Also keep in mind that even with high ratings, voltage will drop noticeably by a few C (related to ESR)

12 - Quite approximate for a few reasons, amongst which the fact that you would need to approximate 12V with various cell voltages and various practical cell types. Probably needs some verification too(verify).


Sources:

Charging, discharging

Battery charging

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

At its core, battery charging putting voltage on the battery.

It's not entirely unlike charging a capacitor, with some consideration the battery's chemistry: it's a good idea to choose an appropriate voltage, limit the current, slow down/stop when it's full, and more.

If you know what you are doing, you can do it yourself with very little circuitry, but if you want safety, long life, and/or automatic charging, it takes more care and circuitry.


Simple chargers are constant chargers - they usually have a low, fixed current and a table showing after how much time you need to pull out a battery of a particular capacity. These can damage batteries when you allow them to overcharge (for a considerable time).


The cheap sort are often fixed-current, which means the charge time depends on the battery capacity.

Somewhat nicer chargers are slowish-and-timed, often combining the charger's known current with an assumption about capacity, cutting out some time later when the battery's likely full enough. If the battery has larger capacity than expected, they'll just stop short.

More advanced versions actually measure things, and are often sold as being microprocessor-based. They assume only charge characteristics of a battery, and use measurements to stop charging when it seems to detect the battery doesn't like any more charge. If you (want to) pay no attention to battery charging, these are good for your batteries' life, they may charge your batteries more fully without overcharging or stopping early. (Note: These occasionally to misjudge batteries that are not broken in)


Aspects to charging include:

  • avoiding more current than the cell can comfortably charge with (varies with its chemistry)
    • Charge speed is often chosen proportional to cell capacity
    • There is no standard to 'slow charging' or 'fast charging', and the terms seem biased to NiMH/NiCd, but:
      • slow charge tends to refer to no more than 0.2C, which for NiCd/NiMH means more than half a day from flat to full
      • moderately fast charge is something like 0.3 to 1.0C, perhaps 1 to 4 hours
      • fast charge may be up to 2C or 3C, meaning charge times well under an hour -- given that the battery can take it and there is a good cutoff method.
  • You should stop charging at some point or you'll damage the cells.
    • Just applying a constant current without a cutoff is a bad idea, mostly because once a battery is full, significant applied current will cause the cells to heat up, which eventually builds pressure and releases its electrolyte, or rupture.
    • Simple chargers are often either timed for a reasonably modern average battery capacity and a moderately fast charge - or just slow-charge without cutoff, which will heat the batteries once full, but not unreasonably so.
  • a method of telling when/what full actually is. Options include:
    • In theory, the delivered cell voltage can be used, but this varies per chemistry, and also on cell health, needs to settle a bit after charging, you need to be conservative for batteries with a fairly constant voltage though most of their charge/discharge curve (e.g. NiCd/NiMH) which may lead to terminating a charge before the battery is full. Not the handiest option.
    • Negative delta-V detection refers to the fact that battery voltage of NiCd and NiMH falls a little when it can take no more charge, and zero delta v detection refers to the change levelling off. A charger that can detect such a peak can use it to stop. However, NiMH has a less pronounced negative delta-V than NiCd, and this is not so useful for some other chemistries.
      • Fast chargers with negative-delta-V detection tend to start with a slow charge, because there may be peaks when starting to fast-charge an almost empty battery.
    • a temperature sensor in the battery pack. Good as protection, not as an early signal, as serious temperature increase is not good for the battery, and temperature rises a little in normal charging, and to do this thoroughly you arguably need a a sensor on each cell, making this less of an option on household batteries, though interesting on e.g. laptop batteries.
  • possible wish for cell-specific behaviour, such as
    • start charging slowly when battery was near empty (often better chemically)
    • fast charge for much of the time
    • slowish charge when nearing full
    • trickle/float charge, to top off completely - also useful in storage


Further charger features:

  • switch to a trickle charge (on the order of 0.04C. Varies - NiCd can often take C/10, some NiMH is suggested at at C/40)
Batteries slowly discharge because of their internal resistance, and a trickle charge keeps the battery full. In Lithium Ion batteries the trickle charge is so low that it is almost impractical to do, so not commonly seen (verify), and whether this is a good idea (varying per chemistry) is debated
  • balancers
try to charge individual cells equally
  • occasional cycling: discharge-and-charge.
  • do a top-off charge, particularly after a relatively conservatively terminated but fast charge, to make sure that we didn't stop because of something heat related. Regularly interval-based to avoid heat. May add a few percents of power.
  • in-circuit charging
That is: you do not have to disconnect the battery from the device to charge it, and you may be the device can be used while charging (possibly with uninterrupted power)
not hard in theory, as the charging supply can just power the circuit while it charges. May have a few basic implications to the circuit (e.g. a regulator if it is sensitive to having a few volts more on it)


Charger chips and charger circuits often allow some basic cleverness. They

  • often include a regulator to manage/limit charging voltage:
    • linear regulator is simple but a little wasteful
    • pulsed/switching voltage control - control voltage through pulse width modulator, in general or near the end of a charge. More efficient with power than linear regulator, but needs to be smoothed, can cause interference/noise
    • buck regulator - switching regulator with step-down
    • shunt regulator - effectively shorts out the power source (or at least leads it past the battery) at a certain voltage level. Can help avoid reverse biasing too (so handy for solar)
    • Note that you preferably want a regulator with low drop-out, so that you don't need
  • my do a constant charge and cut off after some fixed time, or don't cut off at all
    • Preferably slow chargers, as this makes damage from heating less likely) More general-purpose (e.g. NiCd+NiMH) chargers are often slow chargers for this reason
  • react to voltage and/or negative delta V characteristics (basic intelligence is becoming more common in chargers)
  • can be a smart element between charge source, battery, and (possibly regulated) voltage going to the load/circuit


Some of the more common ways of applying the power while charging:

  • constant voltage, such as attaching an adapter (preferably regulated?)
    • LiIon is constant voltage -- plus a bunch of protection circuitry
    • works well enough for lead acid(verify)
  • constant current - vary the voltage to maintain a constant current flow
    • common for NiCd, NiMH
  • pulsed current - duty cycle of mostly charge, and a pause with a little discharge
    • arguably can be chemically good
    • pauses also allow for open-circuit voltage measurement
  • Trickle charge - to balance the battery self-discharge current
  • Float charge - battery, load, and charger are all in parallel, at a constant voltage below the battery's limit. Mostly used with lead acid batteries, commonly seen in emergency backup lighting and such.
  • Random charging - describes non-constant, unpredictable power application, such as in solar charging, regenerative braking, car alternators. Generally needs current and/or voltage limiting to stay within battery allowance.


Full discharge is not recommended because of a few reasons. One of them is that most batteries are made of individual cells, the behaviour of which varies. If one cell is empty before another, you start charging it in reverse while discharging the others, contributing to voltage depression. This means that discharging NiCd/NiMH should stop at about 0.8V to be safe.

Battery chemistries are made for moderate temperatures; batteries don't deal well with either high or low temperatures.


The charge should be switched to a trickle charge at this point or the battery will begin to vent and release its electrolyte. My old battery was rated at C=1300mAH and my old charger was rated 400mA (30% of C) so the charger should have been switched off after about 4 hours, provided that they were almost flat to begin with. However there is no way of knowing if C was actually 1300maH or if it had decreased a bit, and once the a battery starts to deteriorate, I suspect this becomes a vicious cycle and the battery deteriorates rapidly due to more and more overcharging. The manufacturer suggests these cells should be good for 500 to 1000 cycles if properly treated!


It is not easy for circuitry to detect the charge left in a battery, or the overall usable charge of a battery. Perhaps the most reliable method is to measure the power delivery over a practical discharge, something that laptops may do.


See also / unsorted / to read:

C rating

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

A C rating basically says the battery will deal with a particular discharge rate well enough, and it relates it to the battery's capacity.

This seems somewhat arbitrary, until you realize that's also a good indication of how low the ESR is.


For example, e.g. a 10Ah at 1C means 10A (for 1 hour).

C ratings are useful to do some quick estimations for load and/or time time.

Say, 2C means you could drown it in half an hour, 10C means you can drown it within six minutes.


C is also used to indicate amount of load - it'll tell you whether your some-amount-of-amp load is going to work at all, and whether discharging at that current is going to make your battery unhappy very quickly.

Say, some of the remote-control-vehicle crowd can care about a lot of power at once, so will prefer packs that are rated at a few C (and these packs may also make distinction between continuous and burts which, if you get your assumptions wrong, will not make a happy battery).


C rating also relates to charge current. In theory this is roughly the same as the discharge current, in practice you would often use a lower rate unless you're in a hurry - mostly to not get the pack as hot.

Depth of discharge

Chemistry-specific notes

NiCd and NiMH notes

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

A NiCd slow charge tends to aim for 0.1C - 0.2C charge to be safe, which tends to mean the order of 10-16 hours per Amp-hour .

Fast charge, 1-3hr, is possible but is likely to reduce lifetime somewhat faster, particularly if the battery wasn't rated for that current.


NiMH deals better with faster charges, probably 4-8 hours per Ah. Slower is still somewhat better, partly because excessive heat (>40 Celcius) will more easily cause damage (NiCd can stand heat somewhat better than NiMH).

Note that batteries heating up a little near the end of charging is normal - in that it is to be expected, not that it's good.


Note that while NiMH/NiCD batteries will charge in chargers not directly intended for them, different care and guards apply to each, so you can avoid a few types of mild life-reducing effects (mild damage) by using a charger specifically made for the battery type.


Avoid storing batteries entirely empty, and if you suspect a stored battery may be empty, charge it before use (avoids possible reverse charging because of a complete discharge).

For long-term storage, consider cells exhibit self-discharge. This is lessened when stored at somewhat colder temperatures, but apparently increases at particularly cold temperatures(verify).

Each chemistry has its own range of self-discharge rates - it differs between chemistries, but it also differs between designs within a chemistry.

NiCd is apparently more resistant to empty storage than NiMH(verify), but in both cases, avoid it if possible.


You may observe that in first charges the capacity isn't large - that's the reason to break batteries in. Ideally, new batteries (usually empty) should broken in: they should be fully charged and discharged three to five times, and ideally slowly charged at that (slower for NiMH).


Lithium notes

Lead acid notes

Variations
  • flooded lead-acid, a.k.a. vented
    • cheaper, simpler
    • should be used upright, charged upright
    • should be maintained by adding distilled water


  • VRLA (valve-regulated lead-acid) - informaly SLA, Sealed Lead Acid
    • allows faster charging than flooded type
    • do not require occasional addition of water to the cells
    • have a safety pressure relief valve (do not spill their electrolyte if inverted)
    • Gel battery
      • safer for indoor use than flooded (negligible emissions)
      • gel instead of liquid - orientation does not matter; little spillage concern
    • Absorbent glass mat (AGM) battery
      • better specific power, worse specific energy (?)
Charge state estimation

Battery voltage is influenced primarily by:

  • state of charge
  • current currently drawn
  • temperature (on an order of magnitude that you'll notice the difference between freezing winter and scorching summer, but you can ignore if things stay room temperature-ish)


The discharge curve is mostly straight - down to ~20% is straight enough that open-terminal voltage alone gives a decent estimation of how charged the battery is.

Measure this some time after the last noticable charge/discharge, when the chemistry has settled. That, or watch the voltage slowly settle.


With a little more modelling you can do it known currents too.

  • Open-terminal (note this is relatively meaningless)
    • Full is usually below something like 2.13V per cell (times six cells is ~12.8V)
  • under ~C/100 load
    • Full: perhaps 2.11V per cell (times six cells is ~12.7V)
    • 20% charge left: perhaps 2.02V (times six cells is ~12.1V)
  • under ~C/10 load
    • Full: perhaps 2.06V per cell (times six cells is ~12.4V)
    • 20% charge left: perhaps 1.92V (times six cells is ~11.5V)
  • under ~C/5 load
    • Full: perhaps 2.01V per cell (times six cells is ~12.1V)
    • 20% charge left: perhaps 1.81V (times six cells is ~10.9V)
  • under ~C/3 load
    • Full: perhaps 1.90V per cell (times six cells is ~11.4V)
    • 20% charge left: perhaps 1.74V (times six cells is ~10.4V)

There are various charts out there - these figures are based on one or two of them.


Under charge, things are a little more complex - above ~70%, the voltage starts rising considerably, rather deviating from a straight line.


See also:

Discharge protection; sulfation

Sulfation refers to crystallization of lead sulfate on the negative plates.


Most of the lead sulfate will easily convert back when charged, but some will not, and over time will coat the battery's plate, which isolates them, and isn't reversible.


So for battery longevity, this is best avoided. Which mostly means "don't let it get too empty":

  • Below ~1.8V (10.8V for six cells) damage is pretty certain
  • Charge controllers asked to discharge will often cut out well above that (perhaps ~25% charge, ~2.0V, 12.0V for six cells)


Sulfation at all actually starts happening below ~2.06V (12.4V for six cells), which is at ~70% of capacity, so for the longest lifespan, try to keep it above 50% and preferably even above 75% if you can. This is the main reason behind advice like 'half-discharge cycles are better.'


Self discharge is approx 2% per month(verify), so when storing batteries, a quick recharge once every year helps lifespan.

Charging (and corrosion)
✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

Basic chemistry means that applying approx 2.15V (per cell) will charge a lead acid battery.

Lead batteries do not like overvoltage, charging too fast may heat them more than they like, and overcharging will start the generation of oxygen, which corrodes the battery.

As such charging lead-acid involves some voltage regulation.


How high is too high, voltage-wise, depends noticeably on the type of battery. Gelled, AGM or flooded differ somewhat in this and other details, such as what they should be trickled at. When in doubt, follow the battery or its datasheet. For example

  • some variants can be fast-charged at 2.4V per cell (15V for 6 cells) for most of the cycle.
  • Still, chargers often have to assume the lowest common denominator, so most stick to 2.3V per cell (13.8V for 6 cells)
  • ...or even just a 2.23V per cell (13.38V for 6 cells) trickle-like rate and to be nicer to the gelled variants


If a charger tries to be smarter, it may

start with a slow charge if the battery was connected entirely flat
faster for the main charge
eventually start tapering off for top-off/trickle charge

You could charge the battery at at its trickle-charge voltage. Since this takes forever, this only really makes sense for things like emergency lighting.


See also:

Battery sizes

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

Some of the most commonly sold sizes:

  • AA (13.5–14.5mm x 50.5mm)
a.k.a. R6
  • AAA (10.5mm x 44.5mm)
a.k.a. R03(verify)
  • 9V - also known as 6LR61, PP3 (from a series where others are obsoleted), 006P, NEDA 1604 and IEC 6F22 (many of which refer to specific chemistries, but are not always used consistently), E-block (in a few places)
(which actually contains six to eight 1.2V to 1.5V cells, possibly AAAA, or other shapes)


  • 18650, 18500, 18350, 16340, 14500 and such are size references and typically imply Lithium cells
These are millimeters indicating diameter and length. For example, the 18650 has a 18mm diameter, and is 6.5cm long.
(I'm not sure whether the 0 means anything(verify) - might be millimeters but in cells this large it seems to always be 0)
Note that:
14500 is basically AA size (...but not in voltage, if lithium)


Small batteries, called things like button cell, coin cell, watch battery, of which there are many variants.

  • The letter-letter-number style are often according to IEC 60086-3
    • The first letter indicates chemistry, commonly
      • L (Alkaline, ~1.5 V)
      • C (Lithium, 3.something V), or
      • S (Silver, ~1.5 V)
    • If there's an R, it means round
    • The numbers relate to size.
  • The longer codes will more directly mention diameter and thickness, e.g.
e.g. CR2032 is lithium, round, 20mm diameter, 3.2mm thick; CR2477 is lithium, round, 24mm diameter, 7.7mm thick
e.g. SR516 (silver, round, 5.8 mm diameter, 1.6 mm thick),


  • The shorter codes are a little less direct, e.g.
LR44/SR44 are 11.6mm diameter and 5.4mm thick. See also this table
A number have multiple names, like LR44 is basically the same as LR357, and AG13, and sometimes a few others.
This runs into the confusing area of manufacturer-specific or country-specific naming, compatible replacements, and/or same size but different chemistries, which might be compatible




http://en.wikipedia.org/wiki/List_of_battery_sizes

Sensing cell/battery state

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

Just using voltage is better than nothing, but also never great.


Lead-acid may be best to read off by voltage, but even then only when not in use, and for anything battery that was better engineered to keep its voltage until it's empty (which is great in other ways), reading it off this way will go worse.


Battery indicators almost must be specific to the characteristics of the a specific battery type, which can work well for a car, or special batteries, or built-in batteries.

...but if you've ever wondered why a portable device doesn't like to work with rechargeable batteries, well, that's because it wasn't designed with them in mind (non-rechargeable are ~1.5V full and rechargeable ~1.2V, and four of those means the difference between 6V and less than 5V).


They may also consider the load characteristics, because most things will indicate their state while they are being used.

measuring when under mild load actually gives you better information -- but only when you have more contextual information (e.g. current capacity)
devices that know they don't load a lot can ignore that without being noticeably wrong


Particularly where voltage is a rough estimate at best and you want to do better (basically true for all Lithium), you might want some or all of:

  • context (how much you are currenty loading, what a charger is currently doing - and note that lithium charging has distinct stages),
  • ideally some memory (how much voltage depression it has suffered) is, and more.
  • some metadata (what chemistry, what energy capacity, what current capacity / equivalent series resistance) and
note that some of those change over time so you might care to estimate that too

Things like laptops can do this because such variables are better known and more controlled, but in many other situations the estimation is a lot rougher.



Practicalities:

Measure when drawing current. Open-terminal voltage tends to tell you less than voltage when drawing a little current, though the difference can also be informative.


You may be interested in a relatively small voltage range. For example, for a 12V lead-acid battery, most of the interesting stuff happens in the 10V-13V range.


Voltage sensing tends to mean doing some voltage shifting/dividing/crafting, plus some zeners or a comparator if you want an indicator, or an ADC if you want a number on a display somewhere.


TODO: read and sort


Battery leaking

What causes leakage?

What is it that leaks out, and how bad is it?

How do I clean it?

Battery myths and not-so-myths

"You should always discharge your phone completely before charging"

"Closing apps makes battery life longer"

"Fast charging damages LiPo"

"Leaking is caused by too low a charge"

"You shouldn't leave your phone on the charger"

Store batteries in the fridge

On memory effect, voltage depression, and such

✎ This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.


"Memory effect" seems to be the "Ground loop" of batteries - a vague name that groups multiple distinct issues, which has lead to lots of confusion that is hard to clear up.


Memory effect points at the idea that "specific charge/discharge patterns limits how much it can charge/discharge the next time"

The thing is that there are multiple reasons you can limit the capacity, like, you know, damaging the battery, but the suggestion behind memory effect is either that

  • you can avoid this happening as quickly with specific charge/discharge patterns
  • you can recover some of that capacity with specific charge/discharge patterns

The memory effect claim is often more specific than the above. We should make a list, but the most common seems to be that a cell/battery won't easily charge/discharge beyond the level it was last charged/discharged to.

If true, the implication would be that a a battery should always be discharged fully, before being charged fully
If not observed, the effective capacity would quickly lessen over time.
Interestingly, this idea that charge/discharge patterns lessen battery life is for a good part, superstitious bunk - but with a core of truth.


Then there is voltage depression, which means the voltage seems to drop off faster than it did last time.

It's very easy, and not necessarily wrong, to assume that means the rechargable battery doesn't hold as much capacity.

But actually, even voltage depression isn't really the right term, because (depending on details) it is possible to have voltage curve drop faster even though the capacity you get out is still much the same.


Need for a better name?

When memory effect and voltage depression point at rather specific things, we may want a broader name.

Also, it forces a broader view that lets us point out that even if there is value in micro-managing each battery, you can also make such micro-management pointless if it gets abused anyway.


I guess just name it by the symptom we care about most - reduced capacity - and not by specific hypotheses.


Some real causes behind reduced capacity include:

  • cell age
Different kinds of batteries have noticably different storage lifespans
to be fair, there are specific reasons that we should probably end up naming
but this seems to come largely from chemical reactions other than that of charge/discharge(verify), which varies with quality(verify)
and, depending on the chemistry, also with state of charge, temperature, etc., which is why e.g. lithium has specific instructions
  • cell use
Some age with use more more, others age regardless of use
You could try to express this in expectable amount of charge-discharge cycles.
whether you do partial or whole charges is often less relevant than that you use it at all
e.g. while lithium can be stored longer, it has fewer uses (apparently using them will slowly cause side reactions in the electrolyte which traps lithium, reducing the amount of potential lithium ions that we need. This might be worse with abuse, but happens even if taken care of)


  • state of charge
Some vary with the stage of charge it is kept at, others much less so
  • cell damage due to overcharging
and this varies with chemistry - e.g. lithium is less happy to be left on a trickle charge than many others, which is why a careful charger may refuse to trickle charge litium, refuse to continue charge until its SoC has fallen a bunch (and in terms of battery life this is preferred behaviour), and why if you really want to leave with the fullest of battery you can time that cycle (some phones do this, based on your alarm)


  • cell damage do to over-discharge
  • cell damage from deep discharge in a battery
that is, even if a single cell is resilient to go to 0V, when you combine multiple cells and some go flat, this often amounts to reverse charging other cells, which is damaging
  • cell damage due to high temperatures
    • from working in a hot environment
    • from overly fast charging
    • from overly fast discharging
  • means of storage
varies with chemistry - in some cases this is negligible, in some cases it is noticable (though often only starts being a problem when the charge is fairly low)
  • specific chemical effect
it has been pointed out that sintered plate NiCd, which if discharged to precisely the same point would show this effect.



Misdiagnosis:

  • treating over-charge voltage as indicative
if you pick a battery from the charger, it will settle from the charge voltage to its real voltage within minutes or hours. If you look and that voltage as indicating SoC, then it's easy to conclude "see, it lost 20% of its capacity straight away" just because you called it early.
note that most devices's battery charge state detection is fairly dumb. Keeping track of energy in and energy out is done on fancier devices, but basic ones just use voltage, meaning this 'just off the charger effect' necessarily means misreporting
note that cheap/slow NiCD and NiMH chargers typically still do a trickle charge (which usually overcharge a little), so batteries fresh from the charger will do this a least a little
  • "Lithium is damaged by the cold"
Probably based on noticing it reported as empty more quickly.
But temperature makes chemistry go slower, so any battery will perform worse -- and with lithium's (often more careful) monitoring it may report as empty more quickly. This "chemical reactions go more slowly" is a temporary effect, though.




See also:

Lithium - thermal runaway and angry pillows

See also

http://en.wikipedia.org/wiki/List_of_battery_sizes

Retrieved from "https://helpful.knobs-dials.com/index.php?title=Electronics_notes/Batteries&oldid=95208"