Memory effect for Li-ion battery

[leo_o2]

Many papers say Li-ion battery has no memory effect.
I want to know if Pulse Charging method will reduce lifetime of Li-ion battery.
Thanks.

 

[tpetar]

You must describe "Pulse Charging" with details. Pulses with what voltage and durations ?

Lithium based batteries dont have memory effect for sure.

 

[leo_o2]

Pulse charging: Charging current is current pulse signal. For example, the cycle time is 20uS. The offtime is 2uS. Current pulse amplitude is equal to Icc. Icc is charging current under constant-current charging mode.

 

[Darktrax]

Li-Ion batteries have the reputation of having little or no "memory effect".
This term comes from the known experience of NiCads when only partially discharged, and then put back on charge.
After repeated cycles, when taken to a full discharge, and a charge is attempted, the charge accepted can seem to be only that amount the cell is usually given, as if it "remembers".

Li-Ion Cells do not do this, but they have their own kind of degradation.
Li-Ion batteries self-discharge with no load, but only just so far. They deliver again OK when re-charged.

Li-Ion , if deep discharged below a critical voltage, about 3V, can be recovered by a specific low current charge regime. Equipment quits unless charging is delivered at this point.
If it gets to 2.7V, there is very little that can be done to recover. Do not attempt to charge a Li-Ion that went below 1.8V.

The Li-Ion type of degradation is not like NiCad. It is regular, steady, even predictable, and depends on the charge state the battery is left in when idle.
Keeping Li-Ion at near 100% by leaving the charger in all the time, degrades capacity about 20% per year.
Keeping the same battery at 10%-15% of full charge, except when you deliberately intend to use it soon, degrades capacity by about 4% per year.
(I quote from memory, but there is lots of authorative information about this)

Regarding pulse charging. I have built and used such where the cycle time was 20mS, and the duty cycle was 95% ON to 5% OFF, but other cycle times like 50uS can be used.
Often, the OFF time is used to measure the cell while deliberately discharging. Good chargers charge each cell individually, and monitor constantly the voltages during the pulse.
The accumulated current*time product is calculated, as is the same product when discharging, and thus a capacity model can be estimated.

 

[leo_o2]

Darktrax,

So, you think there is no harm for Li-ion Battery with Pulse Charging comparing with conventional Constant Current/Voltage charging?

 

[Darktrax]

Pulse charging in most battery types I know of is OK, but you need to look up what best suits Li-Ion.

Pulse charging is not "instead of" constant current charging . When the pulse is ON, the current is still regulated to be constant.
While pulsed charging can be good for some cell chemistry, even to the extent of using brutal biased AC where more is put in than taken out, such as used for some lead acid cells, this is NOT what we mean for Li-Ion

The concept of "constant current", as opposed to "constant voltage" charging still applies, even though the constant current is clearly not really constant, being interrupted by OFF periods.

You have to have clear in your mind what we are doing. The ON period is long. The OFF period is brief, only enough to measure the cell. There is no reverse current during the OFF, and the frequency is relatively low.

The best charge regimes for Li-Ion are extensively published. Do use a search engine.

 

[tpetar]

Is there some special reason why you want pulse charging ?

There is no mistique with Li-Ion charging process, its voltage and current controlled, just see charging graphs and several different methods for terminating charging process. Nothing special, you can use specialized and dedicated IC for this purposes, this simplifyed and make whole design cheap.

Li-Ion , if deep discharged below a critical voltage, about 3V, can be recovered by a specific low current charge regime. Equipment quits unless charging is delivered at this point.
If it gets to 2.7V, there is very little that can be done to recover. Do not attempt to charge a Li-Ion that went below 1.8V.

I didnt have problems to recover several Li-Ion from almost zero voltage 18650 to full operating state with full capacity. Normal voltage range for this bat is from 2,75V to 4,20V. Also the same situation is with mobile phone battery Li-Po without problems from few volts <1-2V to full operating voltage.

Keeping Li-Ion at near 100% by leaving the charger in all the time, degrades capacity about 20% per year.

I think this is wrong. All laptop/notebook manufacturers suggest to keep and use device with connected chargers/adapters, to avoid battery cycling and prolong battery life (Li-Ion battery life 500-1000-1500 cycles or in time 5-10 year). Also I have many SMS alarms with non-stop connected charger to batteries mobile phone Li-Po 700-1200mAh and for other non-stop operating devices Li-Ion 18650 with larger capacities and I didnt notice capacity degradation. I recently make capacity tests with voltage logger under constant current, and I didnt notice some difference. Just to mention Li-Ion is charged to 4,20V after that charging process is turned off, and charging process is turned on again when voltage is around 75%. I use cheap 0,3eur MCP73832 Single Cell Li-Ion Li-Po Charger Controller IC. Off course there will be problems if you keep 4,20V constant on battery, but that is described if we look into Li-Ion charging graph, because of that we use dedicated IC or specilized circuit for this.




Its good to hear something new.

:wink:

 

[Darktrax]

Is there some special reason why you want pulse charging ?

Not for me, but for leo_o2.

Regarding why I think there is a strategy to prolong Li-Ion battery life, esp the 20%/year degradation being charge dependent..

From [1] Battery University A very complete description of Li-Ion charge characteristics.
from [2] Wikibooks --LINK "Battery Power/Lithium Ion Batteries". See "Disadvantages" and "Guidelines for prolonging Li-ion battery life"
and from [3] Wikipedia --> LINK "Lithium-ion Battery"

Normally, the charge monitor will prevent (say) a notebook computer from letting the battery get to a full discharged state.
Outside of that safety space, a fully dischaged Li-Ion cell, left for more than a week, could have a problem.

Until sale, 40% charge with storage safety current cutoff is I think the norm, which stops being the case as soon as the cell is used.

The "battery enters sleep mode forever" phenomenon, if you don't leave it with some charge in it? I am not sure how much of what we experience is about Li-Ion battery basic nature, and how much is forced by the specific charger behavior.

I stripped down my notebook's previous "dead" battery and found the cells perfectly OK, but the controller circuit would never play again. I have been told that some notebook batteries deliver a preset "timed" life, and then quit. I don't know how true this is. I do have to replace battery packs more than I would like.

I am impressed that your experience with Li-Ion cells recovery is so good. Coupled with my discovery of good cells in a "dead" battery, it is starting to make me cynical about what begins to look like a battery planned obsolescence industry.

 

[tpetar]

Some of Li-Ion batteries are with protection circuit, after "dead state" if you remove this part battery works again, but of course this is called "protected battery" with reason.

 

[leo_o2]

Is there some special reason why you want pulse charging ?
:wink:

I am IC designer. In conventional charger design, it uses Constant Voltage charging after constant current charging.
However, CV charging is slower according to its smaller average charging current.
Pulse charging can achieve larger average charging current. So it will reduce charging time.
But I worry if pulse charging will reduce Li-ion battery lifetime.

Another method: detecting ESR of battery. And charging battery with CC always and estimate if battery is full.
But it is complicate and costly.

 

[tpetar]

I am IC designer. In conventional charger design, it uses Constant Voltage charging after constant current charging.
However, CV charging is slower according to its smaller average charging current.
Pulse charging can achieve larger average charging current. So it will reduce charging time.
But I worry if pulse charging will reduce Li-ion battery lifetime.

Another method: detecting ESR of battery. And charging battery with CC always and estimate if battery is full.
But it is complicate and costly.

They call constant voltage, but its not complete appropriate term. Voltage is adjusted to charging Li-Ion charging method, in start voltage can go up to 4,50V and slowly lowering to 4,20V. My suggestion is to skip experimenting with this, to avoid unwanted bad and dangerous flammable siturations. Keep charging process in checked and well-known range. If you want higher current for charging this is not done with pulsed charging, just look this dedicated IC for Li-Ion/Li-Po charging, charging current is adjusted with putting adequate resistor between pins 5 and 2 according to wanted current. Max is around 500mA and formula for current define is Ireg=1000V/kΩ=Current in mA.



Dont try to charge Li-Ion with higher currents if you dont have fire extinguisher, helmet with protective visor and another home.



:wink:

 

[leo_o2]

Understand,tpetar.
However, it is my job to design a better (faster and safe) charger IC.
I want to surpass the conventional scheme.