lithium ion batteries and charging them

[ravener96]

ok, i'm kindof a noob, but not 100%. i have bought myself some 3.7v LiIon batteries off the internet (that i strongly suspect are lying about their amp hour rating of 1150mah, but who knows) and i'm looking into charging circuits for them so my projects can safely charge, but all i can find say they are 5v charging circuits. i could charge "by hand" but that can be boring the n'th time around after some microcontroller ate it all on flashy light and jazz.
also i'm lazy so not having to make my own would be cool.

help me obi wan kenobi, you're my only hope.

[electr_peter]

To properly charge LiIon/LiPo batteries you can use these options:

• dedicated LiIon/LiPO charger for a specific fixed battery capacity - best option for a specific battery
• RC hobby battery charger (like Turnigy Accucel 6) - very flexible, balancing, additional capabilities
• adjustable lab power supply with CC/CV modes - very flexible, expensive, issues with balancing setup

You can build your own charger if you know how. Getting decent and practical connectors to a battery terminals may be an issue/annoyance.

Be aware that lithium batteries are dangerous beasts (bursts into flames relatively easily) if mistreated (meaning excessive discharging, overcharging (too high current or high voltage), balance, etc.).
If you build your own charger or configure lab power supply for charging, set CV limit to 4.2V or less for 3.7V batteries (or 4.1V for 3.6V batteries) and limit current to 0.3-1C.

[Ryano]

..... and limit current to 0.3-1C.

If you didn't already know C-rate is based off the battery capacity. I would keep things at a nice safe 0.5C until you're sure your charging solution is working and you have a chance to verify the actual capacity.  0.5C = 552mA in your case. You should also stop charging when the current being drawn drops to about 1/20th C. 58mA or so for this battery.

[Mr.B]

It is quite literally a 10 minute job to build your own safe and efficient charger.
It may take a little longer than 10 minutes to get the parts together, but all are available on fleabay, seeedstudio.com and others.
Charger chip – CN3063, CN3065 or CN3083 – Datasheet attached below.
Plus 6 other jellybean components.
Build the circuit referenced on page 2, figure 2. “Rx” can be zero ohms for the feedback.
You should calculate “Riset” to deliver about 0.5 C, or 552mA as Ryano mentioned above.

You don't need to go to the trouble of making a PCB for a one off.
Just "dead bug" it and if necessary pot it into a tiny wee enclosure - by tiny, I mean thimble size.

[3roomlab]

i found this guy ... (linked from reading ebike thread from another thread ... )

http://lygte-info.dk/

now to find which seller is real 

[zapta]

Here is an open source single cell charger and regulator

https://www.sparkfun.com/products/11231

[eas]

You can get modules for charging single cell 3.7v nominal Li-ion batteries on ebay for $1 or less.  Some of them even have added protection circuitry to prevent overdischarge and overcharge. Most use TP4056 linear charger ICs.

I have some from this seller. I haven't tried the protection with a load, but the charging works great.

[LaurentR]

Here is an open source single cell charger and regulator

https://www.sparkfun.com/products/11231

Or this:
https://www.adafruit.com/products/1904

She has a few different ones depending on what your power source and application is. Full set of options (and batteries): https://www.adafruit.com/categories/138
The more expensive ones are "in-line" so that the charger is always plugged in the battery and you don't have to unplug the battery to charge it.

[ravener96]

You can get modules for charging single cell 3.7v nominal Li-ion batteries on ebay for $1 or less.  Some of them even have added protection circuitry to prevent overdischarge and overcharge. Most use TP4056 linear charger ICs.

I have some from this seller. I haven't tried the protection with a load, but the charging works great.

it says 5v , is that just the input voltage or should i not use it with a 3.7v battery?

[okashira]

stick with quality cells. Buy some Panasonic NCA chemistry cells from Mountain electronics, illum, fasttech, etc.
examples:
NCR18650PD
NCR18650PF
NCR18650BD
NCR18650BE

are all cells top notch quality should last 10+ years with proper care and are pretty resistant to abuse
they are also much safer then LIPO and can take overcharge, overdischarge, heating, without creating a nasty fire like lipo


They are traditionally charged with a CC/CV power supply with CV set for 4.2 and CC set for, say 1/3 of the cell's capacity per hour.

charging should also stop at 100ma or 50ma in the CV phase.

For lifetime, charge slowly, and use lower voltage (suggestion, 4.10v)

[okashira]

It is quite literally a 10 minute job to build your own safe and efficient charger.
It may take a little longer than 10 minutes to get the parts together, but all are available on fleabay, seeedstudio.com and others.
Charger chip – CN3063, CN3065 or CN3083 – Datasheet attached below.
Plus 6 other jellybean components.
Build the circuit referenced on page 2, figure 2. “Rx” can be zero ohms for the feedback.
You should calculate “Riset” to deliver about 0.5 C, or 552mA as Ryano mentioned above.

You don't need to go to the trouble of making a PCB for a one off.
Just "dead bug" it and if necessary pot it into a tiny wee enclosure - by tiny, I mean thimble size.

I don't like any of those. Charging to 4.242V means your cell won't last long (poor charge voltage regulation) and they say the voltage is adjustable, but only UP! 
Know if any more with adjustable VBAT, or at least better regulation?

[zapta]

Also, make sure to watch this before you start dealing with  lithium ion batteries.

https://www.youtube.com/results?search_query=lipo+fire

[electr_peter]

You can get modules for charging single cell 3.7v nominal Li-ion batteries on ebay for $1 or less.  Some of them even have added protection circuitry to prevent overdischarge and overcharge. Most use TP4056 linear charger ICs.

I have some from this seller. I haven't tried the protection with a load, but the charging works great.

it says 5v , is that just the input voltage or should i not use it with a 3.7v battery?

5V is input voltage to the charger, battery gets charged to correct lower voltage.
These type of chargers act like a linear regulators (they drop some voltage to provide correct voltage to battery)  - they should work from 4.5-4.7V to a bit over 5.2V.

[eas]

It is quite literally a 10 minute job to build your own safe and efficient charger.
It may take a little longer than 10 minutes to get the parts together, but all are available on fleabay, seeedstudio.com and others.
Charger chip – CN3063, CN3065 or CN3083 – Datasheet attached below.
Plus 6 other jellybean components.
Build the circuit referenced on page 2, figure 2. “Rx” can be zero ohms for the feedback.
You should calculate “Riset” to deliver about 0.5 C, or 552mA as Ryano mentioned above.

You don't need to go to the trouble of making a PCB for a one off.
Just "dead bug" it and if necessary pot it into a tiny wee enclosure - by tiny, I mean thimble size.

I don't like any of those. Charging to 4.242V means your cell won't last long (poor charge voltage regulation) and they say the voltage is adjustable, but only UP! 
Know if any more with adjustable VBAT, or at least better regulation?

4.242v is just the max voltage within the 1% tolerance. As I recall, I think 1% is pretty typical for these chips.

As for chips that allow the termination voltage to be set though a wide range, they exist, but they are more expensive. I'd started looking for something reasonably inexpensive and easy to work with, but hadn't found anything that stood-out, yet.

[amyk]

Yes, 1% is a common tolerance for li-ion chargers:

http://www.st.com/web/catalog/sense_power/FM142/CL848/SC969/PF103931?sc=internet/analog/product/103931.jsp
http://www.linear.com/docs/1139
http://www.linear.com/product/LTC4054-4.2

You can get 0.7% or even 0.5%, but they'll probably cost more.

Remember that the 1% is an absolute worst-case and the majority of parts will not be anywhere near 1% out; see https://www.eevblog.com/forum/blog/gaussian-resistors/ for example.

[okashira]

I've found this:
http://www.dxsoul.com/product/lm2596s-dc-dc-voltage-step-down-led-driver-power-w-digi-display-6-40v-to-0-36v-901238638#.VK3bt1XF_QM

to be wonderfully accurate (better then 0.25%)

the only problem is there is no way to terminate charge at 100ma, so you either:
-Set it to 4.17V and forget (for a 100% charge, once current gets to 10ma or less it would be about the same as a 4.2 100ma cut charge)
-manually terminate at 100ma
-choose an even lower voltage, such as 4.09 and just let it go for longer cell life