Charge Li-ion Cells (AAA, etc) Without a Charger

Candle Power Forums

Help Support Candle Power:

MrAl

Flashlight Enthusiast
Joined
Sep 9, 2001
Messages
3,277
City & State/Province
New Jersey
Hi there,

For people still waiting for a charger, here's a procedure
you can use to charge an Li-ion cell if you have some resistors
laying around and a dc wall wart. If you plan to charge at a
high rate you have to make sure the resistors are capable of
handling the power.

The basic idea is to connect the cell in series with a resistor
and dc wall wart and monitor the voltage across the cell. Anytime
the voltage increases above 4.15 volts you increase the value of the
series resistor to twice it's previous value. It takes time
and patience but it works! I charged my first Li-ion cell this
way.

Let's say you have a 6v power supply and a depleted AAA size Li-ion cell
and you wish to charge at a max current of 0.5C .

For finding the initial resistor value for a given type
cell, make sure the cell is depleted to at least as low as
3 volts.


Starting with a series resistor of 20 ohms, 0.5 watt, in series with the cell
apply power and connect the cell. Check current (by measuring voltage across
the 20 ohm resistor and using Ohm's Law, or using a separate meter). Make
sure the current is about 150ma.
With a voltmeter connected directly across the cell's terminals, check
to see that the voltage is less then 4.15 volts.

If at any point you see current over 150ma or voltage over 4.15 volts then
simply INCREASE the value of the resistor. The current through the cell
AND the voltage across the cell will go down when you do this, and this
is normal.

Once you get the initial current set (with the 20 ohm resistor) you'll be
contantly monitoring the voltage and increasing the resistance as the voltage
gets up to 4.16 volts. You'll never allow the voltage to reach 4.19v, that's
the rule. As soon as it reaches 4.16 you'll disconnect and increase the
resistance to either 1.5 times it's current amount or 2.0 times (if you're
not in a hurry).

Here's how a typical run might go:

Connect a 10 ohm resistor across the cell to make sure it goes down to
3.0 volts. Once it does, proceed. (Skip this step once you've determined
the correct value resistor to start with for this kind of cell).

Let's say we got 20 ohms to start with (as probably with a AAA cell).
Now we connect the voltmeter across the cell and the current meter
in series with the cell, turn on the power supply and connect the
cell.

Now let's say we see 160ma flowing and the voltage is 3.2 volts.
The current is just about right and the voltage is under 4.15 volts
so we allow it to charge for some time.

A little later (maybe 1 hour) we see the voltage rose to 3.95 volts and
the current is 100ma (typical). The current is still ok (unless we are
in a hurry) and the voltage is still under 4.15v so we dont do anything.

A little more later (maybe another hour) we see the voltage rose to
4.15v and the current is less than before. Now the voltage is on the
threshold of 4.15v so we get ready to disconnect. A few minutes later
the voltage rises to 4.16v so we disconnect the cell.

Adding another 20 ohms in series with the 20 ohms that is already
there, we then reconnect the cell. Monitoring the voltage, we
see it's less than 4.15v again so we let it charge a little more.
The current probably be around 50ma so it's ok.

Now sometime later the voltage again goes up to 4.16v so we swap
out the resistor again, this time for a total resistance of 80
ohms, then reconnect the cell and monitor voltage again.

We keep repeating this procedure until the current reaches about
25ma and the cell reaches 4.16v again and we're done.

Disconnect everything and you're ready to use the cell in your application.


Have fun,
Al
 
Or.... with that same 6V supply, use an LM317 in a TO-92 package, use two resistors to set the output to 4.15V and be done with it. [Ignoring for the moment that I think some form of backup protection is prudent]. An LM317 in the TO-92 current limits in the 150-200mA range which is just fine for this application.
 
Hello... Great thing... /ubbthreads/images/graemlins/smile.gif

Can I use 7805 .. and a couple of series Diodes(less current rating) (of .6 V Drop each ???)

Where can i get Best AA & AAA Size Li-Ion cells online?

ViReN
 
Salutation Sensei MrAl, how r u ? /ubbthreads/images/graemlins/smile.gif

Why don't you make a simple LM317 circuit "IN WORD", I'll do my best to translate it into Linear SwitcherCAD schematic into .UUE code as before, so this circuit will stays inside this thread forever ! /ubbthreads/images/graemlins/smile.gif

Vic
 
Hello again,

vicbin:
Hey! Nice to see ya here again! Im not bad, hope you're
very well. I took yours and Doug S's advice and made a
simple LM317 charger circuit, although the initial post
was aimed at people who dont feel like buying any more
parts and have some resistors laying around and dont
mind spending a little time to charge :-)
If you'd like to put this into SwitchCad form please do!
If you try out any circuits please let us know too!

Viren:
Cant really use a 7805 with diodes unless you care
to watch the output to the cell to make sure it doesnt
go over 4.200v or else possibly risk the cell.
The diodes 'drop' isnt really that stable -- it will
decrease with decreasing current and temperature. This
makes it a bad choice for a voltage level that has to
be controlled to within about 1 percent which is what
is required for charging an Li-ion cell normally.
Actually, i havent even checked the spec's on the LM317 :-)

Doug S:
As i was saying, the original post was aimed at people
who just have some resistors laying around and dont
have a charger yet. Sure, an LM317 would be better.

Here's a simple circuit that includes current limit:

Li-ionCharger-LM317-01.gif


The resistor R1 value changes with whatever input voltage
you use and the table is for charging an AAA cell rated at
300mAh although you could increase current limit if
you wanted too with those cells in the other thread.

If you want to charge AA type cells you can halve the
value of R1 or even lower depending on how fast you
want to charge and what the limit is on your particular
cell spec'd by the manufacturer.

Take care,
Al
 
Al, just for good measure, you may want to add a 0.1-1uF cap from the IC input to grd to insure stability under all conditions.
 
As an Amazon Associate we earn from qualifying purchases. Product prices and availability are accurate as of the date/time indicated and are subject to change.
Hi there,

With a rock solid battery voltage on the output i dont
think an input cap would make much difference, but if
you'd like to put one that's fine. It sure wont hurt.
The input/output caps are usually to increase stability
during transients, which the battery doesnt care about.

One thing i forgot to mention is that in order to tell
when the cell is 'done' it's possible to read the voltage
across the input resistor (33 ohms on that schem) and
determine when input current is about 30ma. That's a good
time to stop charging as the output cell at that point
is getting about 25ma and it's considered fully charged.
Too bad there's no super easy way to connect an LED to
tell when it's done, except of course with a transistor
added to the circuit.

Take care,
Al
 
Mr Al.
I just finished building the LM317 li-ion charger and it is now charging a 18650 cell. I measured the charger open voltage to be 4.15V. would this be sufficient in giving the 18650 a full charge? Also, i measured the current to the battery to be 130 mA and maxed out to 150 mA as i increase the Vin. is this normal?
given this charge rate, it would fully charge the batt in about 10 hrs /ubbthreads/images/graemlins/eek.gif
 
Hi again folks,


Newbie:
I was shooting for simplest design so i left out all the 'frills' :-)
The design gets the job done so i figured it was ok. If people
want an LED to indicate charge completion, it will take more parts
but i could always add them on, no prob.

shiftd:
Actually i set my charger to 4.15v for my 18650's now too. Doesnt seem
to bother run time.
Yes, 150ma was the top limit for the AAA cell, but since your cell is
around 1800mAh you can always set the max current higher by lowering the
input resistor (33 ohms on the schem). For 9v input if you were to halve
that value you'd get 300ma output, quarter it you'd get around 600ma
which would charge your cell much faster (ok for the 18650). In fact,
you could probably go with one-eighth that value (5 ohms) and
charge that sucker in 2.5 hours or less, but you'll have to use a heat sink
on the LM317 and use a nice high power resistor like 10 watts.

Viren:
Deleted post?

djpark:
Yes...the design here was intended to be super simple so people
without chargers could charge their cells until they get a better
charger. Of course the LM317 isnt bad anyway :-) but the input
power to the charger cant go to zero or the cell discharges. This means
the circuit always has to have power or the cell should be removed.

Take care guys,
Al
 
LOL... Yep... Actually, I am kinda NUT...

You have mentioned "Too bad there's no super easy way to connect an LED to
tell when it's done, except of course with a transistor
added to the circuit."

and .... I was saying some thing similar... i think djpark has read it.. and must have laughed /ubbthreads/images/graemlins/smile.gif

The thing is... that I would like to know what additional Transistor & (perhaps a resistor to bias the transistor) is required...

The circuit still remains super simple... and will also improve on functionality... if we add the LED... (we will be saved from hassle of measuring Voltage at the battery .. i guess)

Thanks & Regards,
ViReN
 
Hello Viren,

Yes, you're certainly right about that. It would be nice
not to have to wonder what the current is :-)
I'll add a transistor as soon as possible and post an
updated schematic.
You'll have to remember that it will have to be
adjusted for each type of cell depending on current limit
i think.


Take care,
Al
 
Mr Al

thanks for the explanation. I must have done something wrong then, somewhere, someway as my max current is only 150 mA, and i have even shorted the input resistance. Is it because i am using the TO-92 package where the max current is only 150~160 mA?
now it is only delivering 70 mA to the batt, with the input resistance is shorted, no matter what the input V is. The batt V is now 4.07V
 
[ QUOTE ]
shiftd said:
Mr Al

thanks for the explanation. I must have done something wrong then, somewhere, someway as my max current is only 150 mA, and i have even shorted the input resistance. Is it because i am using the TO-92 package where the max current is only 150~160 mA?
now it is only delivering 70 mA to the batt, with the input resistance is shorted, no matter what the input V is. The batt V is now 4.07V

[/ QUOTE ]
That is correct. In the TO-92 package it has a current limit of around 150mA so what you are experiencing is normal. If you need more current, use the TO-220 package which current limits at around 2A. Note that these devices also have thermal current limiting which begins at a die temp of around 160C IIRC. The TO-92 package cannot dissapate much power. When you shorted the input resistor, you forced more power to be dissapated in the LM317 driving it into thermal limit which is why the current dropped. I think that you will find that if your reduce the input voltage [you only need about 2V above the battery voltage] the LM317 will come out of thermal limit.
 
Hi Doug

thanks for the explanation. that explains a lot
so the bottom line is, if the LM317 reaches its thermal limit, the current in & out will decrease?
right now i am using 6V from the DC adaptor. it only gets warm but the current is still ~60 mA. i will try to use the TO-220 package as soon as i received them.
thanks
 
[ QUOTE ]
shiftd said:
Hi Doug

thanks for the explanation. that explains a lot
so the bottom line is, if the LM317 reaches its thermal limit, the current in & out will decrease?

[/ QUOTE ]

Right

[ QUOTE ]
shiftd said:
right now i am using 6V from the DC adaptor. it only gets warm but the current is still ~60 mA. i will try to use the TO-220 package as soon as i received them.
thanks

[/ QUOTE ]
If your supply is truly only putting out 6V, that may be the problem. The LM317 needs 2V min across it to deliver full current. As your battery gets above 4V, this requirement would not be met.
 
Hi Doug
i tried changing the value of the Vin and crank it up to 12V and the current only go about 40 mA (it is now going down as the batt voltage increases). Is this normal?
But i think it would be nice to know that it is now trickle charge the battery /ubbthreads/images/graemlins/smile.gif
 
Hello Mr Al,

can We use a POT ... and may calibrate the things for the first time... so that next time onwards, we will just require to set the POT to required position... and it will give a proper bias to the transistor...

ViReN
 

Latest posts

Back
Top