USB Li-ion battery charger

This is a charger for lithium ion batteries which takes its power from the USB port of a computer.

It uses the MCP73861 or MCP73863 Li-ion battery charger chip manufactured by Microchip.

Step 1: The USB power connector

A piece of edge connector hacked off an ancient ethernet board serves as the power connector. To make it, cut off a piece that includes four edge fingers, and then file to make it fit inside the USB connector on the PC.

Comments

[rraallvv]

hi dudes!

just googled "5v 3.7v charger +diy" and was redirected here

I was wondering whether a single resistor in series with a diode would do the trick as follow:

when the batery is completely discharged the resistor will limit the current through the batery, and when the batery is full the current will drop to cero and the voltage in the batery terminals will be about 5V - 0.7V = 4.3V

would it burn/deteriorate the batery?

regards

[Ancient1]

i suggest you google and read about proper chraging of li-ion cells. obviously no one would buy an IC to do what a resistor + diode "can" .
there is little room for error. unless you have a bomb in mind :)

[rraallvv]

Hello Ancient1, thanks, I think at last have got it up and running... Regards!

[Fozzy Vis]

Just a question about the temperature sensing.
The datasheet states that "Applying a voltage equal to VTHREF/3 to the THERM
input disables temperature monitoring."

Am I correct that the resistor values used here are 1.03 and 4.72 (ohm? kOhm?)?
Then how does this work? If I am correct, this would give an Therm on pin 7 of 2.05V, given the reference voltage of 2.5V from ThRef on pin 6.

I just can't figure out what I should use, and what the best values/setup would be if I would want to use the internal thermistor that's inside most battery packs.

Any advice?

[muzica]

Is it possible to just use a Lithium ion battery of 3.6V, directly connecting it with two wires out from the battery and solder with the circuit?

[ters]

nice tip! flatten the bare connecting wires.. never thought of it this way.. =) I will try this method on my next upcoming TQFN proto projects... Thanks!

[sikeren85]

when the battery is full charged.will it automatically stop charging?

[neelandan]

Yes.

[Eonir]

Some commenters say this is not a real instructable, and that this design is just taken straight from the datasheet. That may indeed be the case. One might argue that you have not submitted the actual circuit(for the one shown includes a thermistor).

But I have to say - your method of soldering QFN ICs is awesome. Thank you.

[Gilius]

GOOD GOD! That's a LiPolimer battery! Never EVER charge a LiPo like this. Always check the temperature and adjust charging acordingly.

[tzitzu]

IT'S A  li-ion batery , beside the controler , its specialized for this
more details you can  get watching the controller specs in a (www.)datasheetcatalog(.com)
i actualy  buy few  for charging li-ion - li-pol  bateries because of small dimesions and good power capability , i recomand this charger
and from my side  can get a thanks for posting it
br.

[tzitzu]

Features
• Linear Charge Management Controllers
- Integrated Pass Transistor
- Integrated Current Sense
- Reverse-Blocking Protection
• High-Accuracy Preset Voltage Regulation: + 0.5%
• Four Selectable Voltage Regulation Options:
- 4.1V, 4.2V – MCP73861
- 8.2V, 8.4V – MCP73862
• Programmable Charge Current: 1.2A Maximum
• Programmable Safety Charge Timers
• Preconditioning of Deeply Depleted Cells
• Automatic End-of-Charge Control
• Optional Continuous Cell Temperature Monitoring
• Charge Status Output for Direct LED Drive
• Fault Output for Direct LED Drive
• Automatic Power-Down
• Thermal Regulation
• Temperature Range: -40°C to 85°C
• Packaging: 16-Pin, 4 x 4 QFN
16-Pin SOIC (MCP73861 only)
Applications
• Lithium-Ion/Lithium-Polymer Battery Chargers
• Personal Data Assistants (PDAs)
• Cellular Telephones
• Hand-Held Instruments
• Cradle Chargers
• Digital Cameras
• MP3 Players

[GoatBoy]

Please feel free to explain to us what the differences are between classical "lithium ion" batteries and "lithium polymer", particularly from a charging standpoint.

[Gilius]

neelandan nailed it. Here's how spontaneous disassembly looks like: http://www.youtube.com/watch?v=Gboge17PNBk
People have lost entire workshops like this.

They're very sensitive to voltage change, overcharging, overdischarging and last but not least temperature. If the LiPo starts to heat up it's bad news. First it'll start to swell and then it'll burst violently. The burst might not even happen on the same charge cycle as the initial swelling so if you find your battery swolen disconnect it and dispose of it.
They're much more sensitive than Li-Ion batteries too. Don't know exactly why (I'm not a chemist)

[GoatBoy]

It was a trick question. From an electrical charging standpoint, there is practically no difference between the two types.

I don't know what you mean by the fuzzy reference to "more sensitive"; they are mechanically more sensitive to abuse, but that's not the fault of the charger, is it?

The real problem with lipoly's, particularly in the RC world, is the fact that those hobbyists would rather throw money at a problem, buying an expensive charger that offers them no protection (because someone else recommended it), than sit down and understand what's going on. Combine that with routine battery pack abuse, and general technical inability (read: blinking 12:00), and you wind up with horror stories and old wives tales that are more of a function of the user than the equipment. These accidents are about as "spontaneous" as getting run over by a slow moving steamroller.

The video you linked was a very staged scenario -- you should read the backstory on it.

[robomaniac]

It was a trick question. From an electrical charging standpoint, there is practically no difference between the two types.

In fact there is!

The lithium Ion has a nominal cell voltage of 3.6V
and the lithium polymer has a nominal cell voltage of 3.7V

I have a Multiplex LN-5014 battery charger and it clearly tells me to choose between Ion and Polymer.

Even if that is 0.1Volt, I don't want to mess with them. Check the video!

J-

[GoatBoy]

You are a perfect example of what I'm talking about.

For starters, it's not the nominal voltage you care about, it's the max charging voltage. Your charger doesn't stop charging at 3.6V or 3.7V "per cell", it stops at 4.1V or 4.2V "per cell".

Second, 3.6V is LESS than 3.7V. The original comment was "never charge a lipo this way". Hrm. Li-Ion is *sometimes* spec'd at 3.6V. Lipoly is 3.7V. 3.6V < 3.7V.

Third, even most lithium ion cells are 4.2V max nowadays. For instance:

http://www.batteryspace.com/prod-specs/18650_2400.pdf

Or how about one of these:

http://www.batteryspace.com/prod-specs/Li-Ion%2014500.pdf

Lastly, I had to put "per cell" in quotes because while I'm sure you feel safe because you have a little magical "switch", the fact is your charger does NOT have a thermal probe. What's worse is that your charger says it will do more than 1 cell in series, but DOES NOT EVEN HAVE INDIVIDUAL CELL MONITORING. So not only are you the perfect example of what I was talking about, so is your charger.

Are you engaging in scare tactics as a way to justify the price you paid for your charger? Why do you keep referencing a basically anonymous video that has NO information in it WHATSOEVER. They give NO indication of:

1. What their power supply settings were, or what kind of battery it was
2. How long it took to make it blow up -- no time information whatsoever. It's clear that there was a time lapse if you watch, but they don't specify how long

Instead of telling people to check a video, why don't you check the facts?

If you want a real example, then go here:

http://www.rcgroups.com/forums/showthread.php?t=151687

For those of you who are reading impaired, let me try to summarize:

It took nearly one HOUR of charging at TWICE the max voltage to get the battery to catch fire. Does that sound "spontaneous" to you?

Please stop engaging in FUD.

[newberrys]

wow you know your stuff.

Can you help me with making a charger for two NiMH AA batteries? I want to charge them of a USB power supply.

http://www.instructables.com/id/Solar_Ni_cad_or_Ni_mh_battery_charger
The Instructable above outlines a solar charger for one AA battery as, '+' to '+' (with a diode inbetween) and '-' to '-'. is this advisable? and are there any side effects to not cutting off the power when it reaches a full charge?

[klee27x]

He's right about one thing. After building a DIY charger circuit, a temperature sensor is a must, at least for the first couple of times you use it. Your hand works very well for this, and it would have detected a problem way before any of the batteries in any of those videos exploded. If it gets warm, disconnect it. If it explodes, it is gonna be too hot to even touch. There's one exception - if charged too fast, a cell might get lithium deposited unevenly on one of it's layers.. and this can create a sharp point which can potentially rupture a layer and cause an internal short. The temperature increase could be extremely fast and the cell could potentially rupture "without warning." But to get these uneven deposits, you would have to charge it improperly to the point where it was warming up while charging... That's why you should stop if it gets warm, even though that it may still be several abusive recharge cycles away from even the possibility of spontaneously exploding.

[robomaniac]

HAHAHA! Thanks for the info! J-

[neelandan]

The MCP73861 does have an input to take care of temperature rise (pin 7). However, this charger is fed from a USB port which is limited to 500 mA or so. Since the charging current has already been limited by the circuitry inside the computer it is relatively safe to dispense with the temperature limiting input. However, you might get into trouble if the battery is already hot due to some other cause.

[joeporter]

Nice USB project. I noticed that the MCP73861 chip is available in a 16 pin SOIC package and there are SOIC to DIP adapters (basically circuit boards with pins) from ep board design.

[Sandisk1duo]

or you could buy this:
http://www.dealextreme.com/details.dx/sku.15264

[ReCreate]

you should put a fireproof glass...no bulletproof glass shield in case if the battery goes peanuts

[BjornSmith]

Would you mind to tell me what components you used in this charger? not only the MCP73861 chip that you listed. Thank you in advance

[revellin]

nice, is there an DIP package of the IC???

[neelandan]

Not by microchip. Other semiconductor manufacturers might have DIP versions of their LiIon battery charger chips. Try Maxim or Texas Instruments, they both give out free samples.

[matse9999]

can I connect this charger directly to a cell phone????

[neelandan]

No, but you might be able to connect it directly to the battery of the phone.

[claymation]

would this chip work just as well?
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/4002/t/or
i created an account on that website and was able to get 2 free samples of the chip, just wondering if it would work

[neelandan]

It should work, and be easier to use since the current limiting is "on chip". Just download the data sheet and implement the suggested circuit.

[neelandan]

All batteries with lithium inside them have a polymer separator, and the lithium changes beween two levels of ionisation. Certain combinations of temperature and voltage might cause metallic lithium to form, and then what follows is what the manufacturers like to call "spontaneous disassembly" or, in other words, an explosion. If you do not want your lithium battery to blow up you have to be very careful not to exceed voltages and temperatures specified by the manufacturer of that particular unit. Some cells can be charged to a maximum of 4.2 volts while others are limited to 4.1 volts.

[LasVegas]

A truce has been made. Mr. Neelandan and I are joining forces as Collaborators on the follow-up project. If others are interested in helping, feel free to ask to join in the collaboration. Trolls need not apply. :)

[LasVegas]

Hmmm.... Apparently, the "truce" was a ruse to steal the rest of my idea and publish it as 5V stabilised supply for USB hub. Since it's already been published in another form, I might as well post my schematic here for others to use...

[LasVegas]

So now I know the purpose of Instructable Colaborations! So others can steal ideas and claim them as their own! BTW: That chip running without the required Thermistor my very well destroy the Li-Ion battery attached to it. The REAL INSTRUCTABLE

[morcheeba]

This is no ripoff - it uses a totally different circuit and is well illustrated. You can't lay claim to every USB Li-Ion battery charger design.

[LasVegas]

I'm not claiming it; He is. By posting my drawing in his following instructable. That is where he got the idea. Then he posted the schematic directly out of the chip's spec sheet as his own.

[spinach_dip]

i do something like this for leaded parts. I just glue the ICs upside-down on the PC board and run wire wrap wire (soldered) between the leads. The blank copper board can be used as a ground plane. It's called (among other things) dead bug prototyping. Nicely pulled off, though you might want to pot or encase the finished board. BTW why cut a hole? wouldn't gluing it on top work too?

[neelandan]

The level difference between the top of the ic and the board surface makes it difficult to solder a wire if the ic is just laid on top of the board.

[ladyada]

the max1811 is, perhaps, easier to solder