[error401]

This is my first post here, I've been lurking a while, learning the vagaries of audiophilia and DIY audio gear. I've got parts for a CMoy in the mail, as well as a mintyBoost DC-DC boost converter to hopefully last me the 21 hours plus 10 hour stopover flight I'll be taking to Australia in November. I don't currently have any NiMH stuff (other than my digicam) and I figured it must be cheaper than $20 to build a charger myself, given that I will only be using it for the mintyBoost, which runs on 2AAs.

A slow charger was out of the question; far too primitive and easy. Besides, 8+ hours is far too long for a charge to take, especially when I often sleep less than 6. A smart charger it is. This circuit is based on the MC33340, and is similar to the charger in the PPA, though it's based more on the datasheet example than tangent's circuit and is designed to be small rather than massive. And single sided so I can etch it here. Unfortunately the IC seems to be difficult to find, but it's about 1/10 to 1/5 as expensive as the (more capable) MAX712. Its (lack of) availability is somewhat puzzling since there don't seem to be any other popular NiMH charging ICs around, and $9 for the MAX seems pretty high. I guess most OEMs just build discrete trickle chargers or use µCs? Anyway...on to circuits!

Update: It seems TI has the bq2002 series that is pretty much equivalent to this IC. If I can't find a decent source for these I'll rework the design for that chip. It's nearly identical, shouldn't be very difficult. Slightly pricier, but DigiKey has a couple hundred of each variety in quantities of 1 for about $2.25. That seems to be it though, nobody else has them.

Note that I have not built this or ran all the numbers. The design seems sound based on the datasheets, but there are probably some glaring errors that will send your batteries up in smoke. I'd appreciate your input *before* I manage to get my hands on a 33340 to actually build it.

Ifast is ~1.25A for an ~2.5h charge into 2x2500mAh cells. This charge current should be fairly safe for any AA cells. It's probably too high for AAAs, and definitely too high for 9Vs. Dividers are set up for 2 cells and would need to be adjusted if using this circuit for more cells.

I've spec'd 2A Schottky's for posterity; only D2 requires this rating (though a 1N4001 or similar would probably be fine), I just made them all the same if anyone's going to build this as digikey has a 10 min on these and they should work fine.

Finally, I plan to build this if I can get my hands on an MC33340. None of the Canadian vendors seem to have this part (I was only able to find it at Newark, and only on the US site). DigiKey non-stocks it, and Mouser has a 600-count minimum. Newark has it for $1.80 each, but they won't ship here and buying one part would be a ridiculous price. If anyone happens to place an order at a place that stocks these and would let me piggyback on a few I'd appreciate it.

Enough talk. Pictures:

Schematic:

Board, w/ silk screen (it's about 1.75" square):

Clean board:

BOM:

Please comment! I know most headamp stuff isn't powered by a pair of AAs, but I'm sure a lot of you have portable sources that are

. Also, check out the mintyBoost if you haven't already. It's about $10 in parts and is a DC-DC boost converter that will charge a USB device from 2AAs quite efficiently (ie. you'll get a couple charges out of 2 cells with an iPod).

 

[intlplby]

how would you adjust this to charge something like 2 9V batteries

 

[error401]

If you want to charge 2x9V, you'd probably be better off with tangent's design and it's power buffers. The charge voltage for a pair of 9V likely needs to be about the 20V maximum of the IC, and thus will need to be brought down to a reasonable range for the IC. There are many ways to do this, none of which are represented on my board or schematic.

Charging anything lower voltage shouldn't be much of a problem. You'll need to calculate an appropriate Ifast (a maximum of C, the cell capacity, is recommended) that doesn't over-tax the LM317 or other components. 9Vs are pretty low capacity, so something like 750mA is probably appropriate.

Select R2 such that I = V/R; V is 1.25V due to the LM317 action (or close, you can mostly ignore R4). Therefore R = 1.25/Ifast. This part should be a high-wattage resistor as it will dissipate a fair amount of power. Appropriate values are 0.75-10 ohms, 5W.

You will also need a higher voltage supply, about 1.5x the fully-charged battery pack voltage is probably appropriate if you use Schottky diodes, maybe slightly more if you're using silicons. Finally, the voltage divider created by R3 and R5 needs to be adjusted for the new voltage. The divider should keep the voltage at pin 1 between 1V and 2V for the range of valid battery voltages. Find out what voltage your battery has fully charged and fully empty and select a divider ratio such that the bottom of the range produces about 1V at pin 1 and the top of the range about 2V.

Oh, and if you use an LED, you might want to twiddle RLED's value a bit. It provides about 9mA at 9V.

 

[joostoo]

Nice to see you mentioned the bq2002. i bought a couple from digikey once but never got around to designing a charger. i just built a trickle charger and kinda forgot about them...

 

[louilouinovo]

deleted