Battery Charger using Max713 IC - help needed

[Jack.Straw]

max713

Hello everyone. I've been building these small battery power amplifiers and have decided I want to make them rechargeable. I decided on the **broken link removed** battery charging IC for my charging circuit and got them to send me a few samples. The battery packs I will be charging will consist of 6 AA NiMH 1.2v 2100mAhr Rayovac "Hybrid" batteries. I plan to find the power supplies (wall-warts) at garage sales and such, so I need to know what to watch out for. I was thinking 12v would work and be common, but i usually see lower mA ratings than what i need. Am I correct in my assumption that if I want to charge these 2100mA batteries at a C/2 rate (the slowest recommended rate) I need a wall-wart rated for at least 1050mA? If i wanted a 1C rate, the rate recommended as the max fast-charge rate by Rayovac, I would need a wall-wart rated for 2100mA?

Thanks for your time,
-Jack

 

[banjo]

max713 charger

I took a look at the datasheet and it does appear that you would need at least 1.05A of current to charge this series string of batteries. Of the approximately 12W of max power entering the charger, about 9W is put into the batteries and 3W is dropped as heat in the regulator and the batteries.

I think chargers for this many high capacity batteries use a switching circuit to reduce these heat losses.

If you attempt the wall-wart method, make sure to include lots of air flow and a heat sink on the prototype until you get some experience on the heat that must be dissipated.

 

[Jack.Straw]

max 713

Banjo, thanks for your reply. What do you mean "attempt the wall-wart method"? What would be the alternative? Also, do you know of a charging IC that uses a switching circuit like you mentioned?

Finally, judging by your name, i thought you might be interested seeing the banjo i just finished building:

 

[banjo]

how to reduce current in fast aa battery charger

That is a very awesome looking banjo. The angling of the frets and the bridge must have some advantage for you. Seems like it would take some extra practice to learn how to play it. Probably had alot of math involved in getting all the angles right.

The wall-warts made more than 5 or so years ago all had just a 60 HZ step-down transformer, a bridge rectifier and a filter capacitor. Given there size limitation and limited cooling, the output current was not very high. The cheaper ones may have been rated at 1A, but tended to get very warm if loaded that heavily. In later years some the these wall-warts now contain switching power supplies and can provide much more current without lots of heat. Garage sales probably will have the older ones though. Perhaps a better alternative is to look for old laptop AC adapters. These have higher current outputs and are often cheap. Check some of the surplus resellers like Electronic Goldmine or AllElectronics. Here in Texas, the Goodwill stores centralize their computer offerings in one store. There they throw all these ac adapters into a big box and your sort them and buy them as needed.

My many concern remains the heat dissipation problem. I am concerned that during the fast charge, the PNP transistor will get very warm. The heat should be the highest during the initial charging portion when the battery voltage is the lowest. Perhaps the easiest way to investigate this is to breadboard the circuit and use a car battery as the power source. All you need is a car cigarette lighter plug.
If you find that the heatsinking and cooling of the PNP is sufficient, then move on to the search for suitable AC supplies. If the PNP is overheating, then perhaps a switching arrangement is required. The end of the MAX713 datasheet does show an example of a switching configuration. However, there does not appear to be any design equations for it. Switching power supplies are not really trivial. Contact Maxim/Dallas Semiconductor and see if they have some detailed app. notes for doing this design.

 

[Jack.Straw]

max713 charger circuit

Banjo, thanks for the reply. I took your advice and started looking at laptop adapters. I found a source on ebay for a **broken link removed**. The seller says there is no manual way of witching between voltages, but that it decides automatically? I can't really tell if it's a switching power supply. It's 2A, so I couldn't quite do a 1C rate, but could do C/2. Does a slower rate of charge (lower mA) reduce the amount of heat created? I already purchased small heat sinks made to fit on the PNP, but it sounds like maybe that might not be large enough? When i'm testing, how can i tell if the PNP is getting too hot?

Also, when calculating my values with that power supply, what value do you think i should use for "minimum wall-cube voltage"? I was thinking a 1k resistor would be fine, but i'm still such a newb when it comes to this stuff.

R1 = (minimum wall-cube voltage - 5V)/5mA

Thanks again for all your help!
-Jack

 

[banjo]

texas battery charger ic

Yes, slowing the rate of charge decreases the PNP temperature and the battery pack temperature.

When fast charging, they usually place a termistor or other temperature sensor inside the battery pack. This allows the charger to stop fast charging if the batteries get too hot. The MAX713 part will also stop fast charge if the batteries are too cold.

The AC adapter that you linked from Ebay will most likely do BOTH 12V and 5V. It appears to have a standard molex four pin disk drive plug. That standard requires both voltages although some drives only use the 5V section. I have a similar adapter for my USB to IDE drive interface. However, mine is only rated at 1.5A.

First test for the PNP getting too hot is a simple touch test. Run the charger for a minute or two on a discharged battery. Turn it off and quickly touch the heatsink. If it burns your finger, its too hot. Assuming its not buring hot, run it while longer and attempt to hold the heatsink. The heatsink should get hot and may even be too hot to hold. How hot is too hot is tough to determine. The junction of the transistor is rated at probably 125 degrees C. However, the case is far removed from the junction. Check the transistor manufacturer's datasheet for recommended case temperatures.

The minimum wall-cube voltage depends on the type of wall-cube you use. The ones with a 60HZ stepdown transformer will see the output voltage drop as the low increases. The one with a switching power supply will have a very constant regulated output voltage which is mostly independent of the load current. I would probably start with 12V as the minimum which makes R1 closer to 1.5K. If during testing you find that the minimum is lower, you can always recalculate and replace R1 with a better value.

You may want to check the datasheet for your batteries for the highest battery voltage during charging. According to the MAX datasheet, this voltage can be as high as 1.9V per cell for some batteries. The wall-cube voltage must be greater than (max battery voltage * number of cells) + 1.5V. For the 1.9V case and six cells, you would need 12.9V minimum. I suspect that fast charging would be aborted if this minimum is not mantained.
(I know the rated voltage of the cell is 1.2V, however this voltage is higher during charging to basically force current back into the battery.)