How to charge Li-Ion bare cell

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
I've just converted my ASP TRIAD light to use 18650 Li bare cell. Here is the thread: http://www.candlepowerforums.com/ubbthreads/showflat.php?Cat=&Number=541909&page=0&view=collapsed&sb=5&o=365&fpart=1

I also read an article mentioned that there is two types of Li-Ion cells - with choke anodes (4.1V charge termination) and with graphite anodes (4.2V). So how do I tell what type of those cells are?

The cells are from the Dell notebooks. It has Sony stamp on it US18650GR.

I'd like to know the type before I venture out to getting them recharged using my primitive LM317 method. Thanks.

Henry
 

Doug S

Flashlight Enthusiast
Joined
Jun 20, 2002
Messages
2,712
Location
Chickamauga Georgia
That is a Sony part number for a graphite anode cell. There is no harm and much benefit to setting the termination voltage a bit lower than 4.20V anyway. You will get much more cycle life at 4.15V vs 4.20V. This is especially true for a basic constant voltage charging setup with no form of charge termination.
 

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
Hi Doug,

Thanks for the valuable info. Now I can start to setup the LM317 on the breadboard for experimenting.

At what point should I terminate the charging process if I also have a Amp meter in series? Or just keep an eye on the battery voltage?

The ASP conversion is looking good as I had it on for more than two hours and the brightness seems pretty good and very white. It's about 370mA at 3.71 volts now.

I hope in the future there will be a dimmable circuit that I can put it in.

Henry
 

vcal

Flashlight Enthusiast
Joined
Dec 16, 2000
Messages
3,074
Location
San Gabriel Valley
I know you didn't ask me, but when charging bare cells, I have my charger shut-off the current when the voltage reaches 4.1 volts. This causes charge termination.

Short answer-watch the voltage.
 

PeLu

Flashlight Enthusiast
Joined
Jul 26, 2001
Messages
1,712
Location
Linz, Austria
vcal: this will not fully charge your cells.

I think it was mentioned here zillion times before .-) but there are only few limits:
When cell volatge is too low (usually 3V) start with constant current not more than C/10 (different for different cells).
When cell voltage is above 3V: limit current to 0.7C-1C and 4.1/4.2V.
When the cell is discharged significantly, it will start with constant current until it reaches the 4.1/4.2V. Then current goes down. When you are down at C/10-C/20 terminate charging. No trickle charge (not necessary).

I have cells which reach the 4.1/4.2V limit pretty fast, I gestimate at less tahn 50% capacity. Especially my SAFT choke ones do.

But why care? There are plenty of ICs available which take care of these cells.
 

vcal

Flashlight Enthusiast
Joined
Dec 16, 2000
Messages
3,074
Location
San Gabriel Valley
Pelu-
You've not told me anything that I don't already know, abd BTW, my cells do NOT reach 4.1 volts rapidly. After charging, I have tested them and they can easily put out 1.3-1.5A under load.(!) I do use a very modest rate of charge [250mA}, and can monitor both voltage and amps as the charge progresses. The voltage rise on my cells is VERY gradual.

I've been charging to 4.1v with my Li-ion cells (MANY times) and am thoroughly convinced (proven by ordinary math) that I'm getting the full runtime that I should be getting from my cells.

My runtime -and brightness levels are entirely consistent with the rated capacity of the cell.

If I wasn't getting good service, I simply would not be using this charging protocol for the last several months. /ubbthreads/images/graemlins/grin.gif

-And I DO have numerous other power options to compare it with.
 

Klaus

Flashlight Enthusiast
Joined
Sep 6, 2001
Messages
1,998
Location
Germany
Second Peter,

i.e. do a search for 4054 here on CPF - this LTC4054 at 1.6USD/pcs setup with another one or two cheap parts (cap, resistor) and 5V input (or more) will give you a nice little Li-Ion charger for close to nothing and not much more design work or less than the LM317 IMO.

Klaus
 

PeLu

Flashlight Enthusiast
Joined
Jul 26, 2001
Messages
1,712
Location
Linz, Austria
[ QUOTE ]
vcal said:
BTW, my cells do NOT reach 4.1 volts rapidly.


[/ QUOTE ]
Sorry for using incorrect language (and it was not aimed on you), it should be:
That it could be that your cells are not fully charged. It seems to depend on several factors like the specific cell and they way how the cell is discharged.
But I did not do extensive tests to get numbers up to now.

[ QUOTE ]

After charging, I have tested them and they can easily put out 1.3-1.5A under load.(!) I do use a very modest rate of charge [250mA}

[/ QUOTE ]
That may be the main difference. Your method is only efficicient for pretty low max. charging currents. I'm shure you will get a very good lifetime out of your cells.

Thanks for the additional info.
 

MrAl

Flashlight Enthusiast
Joined
Sep 9, 2001
Messages
3,144
Location
New Jersey
Hello there,

One thing i always wanted to try was this:

Using two LM317's, one to limit current and one to
regulate voltage. When the cell first starts charging
the current would be limited by one of the LM317's and
when the voltage gets up near 4.20 volts the second
LM317 would regulate the voltage so the cell doesnt
overcharge.

Some details would include making sure the voltage
regulator cuts off all (or most) current once the cell
gets up to the full 4.20 volts.

Should be the cheapest way to design an Li-ion charger
unless you want to go with one of the chips made for this(?).

Take care,
Al
 

PeLu

Flashlight Enthusiast
Joined
Jul 26, 2001
Messages
1,712
Location
Linz, Austria
The specific chips are probably even less expensive and they take care also about the deep discharge and end of charge situations.
 

MrAl

Flashlight Enthusiast
Joined
Sep 9, 2001
Messages
3,144
Location
New Jersey
Hello again PeLu,

Yeah :)

I was thinking of people who like to use very common
parts sometimes (like myself) to see what they can
whip up :)

To get back to the main concepts involved in charging
Li-ion cells:

1. If the cell is very low it's suppose to be charged
with a low current (maybe 100ma) until it reaches
at least 2.5 volts.
2. At 2.5v or above, the current limit should be at a
maximum of 1 amp.
3. The max voltage should be 4.20 volts.
4. The cell shouldnt be drained down lower then 2.5 volts.

The circuit i use (on a regular basis) to charge the
graphite cells is just that...It's basically a current
limited voltage regulator.
It uses an op amp and an analog OR'ing circuit that
allows either current limit or voltage limit to occur
--depending on the voltage across the cell and the
current demand. It's still just a current limited
voltage supply though :)

So far it has worked quite well over the past two years
or so. Within the past two months i've lowered the
voltage regulation part to 4.15 volts instead of 4.20 volts
to see if i can get more life (cycles) out of a cell.

Take care,
Al
 

PeLu

Flashlight Enthusiast
Joined
Jul 26, 2001
Messages
1,712
Location
Linz, Austria
Just one thing missing:
5. when leaving at 4.2V, switch off when the current goes down to C/10-C/20.
 

PeLu

Flashlight Enthusiast
Joined
Jul 26, 2001
Messages
1,712
Location
Linz, Austria
[ QUOTE ]
vcal said:
my cells do NOT reach 4.1 volts rapidly.
...
I've been charging to 4.1v with my Li-ion cells (MANY times) and am thoroughly convinced (proven by ordinary math) that I'm getting the full runtime that I should be getting from my cells.

[/ QUOTE ]

Just for other people considering using the shut off at terminal voltage method:

I just run accidently over my Li-Ion cells (choke electrode)data sheets and they state that when you shut-off at 4.1V you get about this charge (depending on max charging current, starting with an empty cell):

C: 50%
C/2: 70%
C/5: 85% (and vcal is below that)

These are estimations out of a paper data sheet. And this is aasuming 100% charge efficiency (current not power), so actual numbers are lower.

And I just had one of these cells which was discharged at a very low current (~ C/1000) down to the discharge limit (2.7V for this cell), and after putting it on the charger (and a limit of C) it reached the 4.1V after a few minutes, probably only 20% full or so.

So the shut off at reaching the terminal voltage method may or may not work for your application.

And one more (I forgot to write earlier): It is debateable what a full Li-Ion is. It is actually your decision (as a designer).
You may trade capacity for cycle life. Klaus posted a diagram for how many cycles you may expect for how full you charge your cells.
When I got the first lithium rechargeable's datasheet in the mid 80ies, they stated a cycle life in beween 50 and 1000 IIRC, depending on the used capacity. For a long life, of course, not only charge to a lower voltage, also do not discharge that far.
 
Top