SparkFun Forums

[qwertymodo]

I'm wanting to build myself a car charger for my Galaxy Nexus spare battery using the MCP73833 charger IC. I have a couple of questions, since I've never built a battery charger before, and the whole potential for catching fire or exploding thing has me rather cautious in proceeding. First of all, I want to run the charger at or near its 1A max charge rate, so I'm using a 1.5A regulator in order to ensure that the charger can get enough current. Also, the 12V DC jack has a 2A quick-blow fuse in it. Should I replace the fuse with a 1.5A equivalent, or will the existing one be fine? Also, reading through the MCP73833 datasheet, it says the voltage regulation can be set to 4.2V, 4.35V, 4.4V, or 4.5V, but the battery is 3.7V (specifically, it's this battery), how should I select the output voltage (it's factory set, so I have to order the correct part number for the voltage I choose)? Here's the datasheet for the chip. I've attached my schematic, pretty much just pulled straight from the reference design, with a regulator added on the input to handle the 12V-to-5V conversion (since the MCP73833 only supports up to 6V input), so could anybody with experience working with Li-Ion charging take a look at it and offer feedback? Battery charging is one of those things I've wanted to tackle for awhile now, and with fully integrated chips like the MCP73833, it seems to be simple enough, I just want to make sure to get it right the first time so I don't have any... energetic disassembly.

Thanks.

[waltr]

That battery has a charge Voltage of 4.2V so use the chip set to 4,2V.
http://batteryuniversity.com/learn/arti ... _batteries

A 2Amp fuze is correct since its for that 'what if' case of a short.

[qwertymodo]

Where are you getting the information that it charges at 4.2V? It's very clearly marked 3.7V. Also, my question about the fuse was whether or not the point-of-failure should be set lower (i.e. 1.5A), considering I'm only charging at 1A, and the regulator is capable of 1.5A max.

[Mee_n_Mac]

LiPos, like any battery, have a discharge curve, a voltage output vs time and current drawn. LiPos are fully charged at 4.2+v and then as current is drawn, produce lower output voltage. The "nominal" voltage is listed as 3.7v. You didn't think the 1.5v batteries were always 1.5v did you ?

[qwertymodo]

Ok, that makes sense. I'm just used to seeing Li-Ion cells listed as 3.7V/cell, I hadn't seen the 4.2V mentioned before. One issue I've just thought of with this design is that I was planning on using a linear regulator, but using 12V input, that comes out to a little over 7W dissipated through the regulator, and the DPAK package I wanted to use is rated at 100C/W. So... 700C might be just a *little* problematic

So I've been reading into it more, and the suggestion I came across was to use a switching regulator instead. Will a switching regulator work for this application, or will it be too noisy for the charger?

[waltr]

Yea, you must read the charging details of the Battery to learn about the 4.2V as per the link I left you.

A switching regulator (most of the newer wall warts are switchers) should work fine. If unsure just add a good size electrolytic cap.

[jremington]

Don't forget to add some sort of spike protection. Voltage spikes of hundreds of volts have been observed in automobile electrical systems, which (although rare) will eventually fry a voltage regulator. Here is a simple example that includes a noise filter: http://webpages.charter.net/dawill/tmor ... Filter.png That 1.5KE18 diode is a 1500 watt suppressor zener, available from Digikey for a few cents.