BQ2002 Charger for 8 NIMH Batteries

[the1cyrus]

Hello,

I'm trying to design something similar to this: http://www.scumari.nl/www/hardware/charger_with_bq2002.pdf

Datasheet: http://www.ti.com/lit/ds/symlink/bq2002.pdf

However theres a few design requirements that changes things.

1. I need the charger to charge eight 2500 mAH NiMh batteries for a total of 9.6v.
2. A 2watt solar panel will do most of the charging on the batteries but also needs the ability to be charged using a 12v DC source that can provide at least 2 amps for faster charging.

Problems im having is im not sure how to handle the current regulation because when the solar panel is charging the batteries it will have low current but when a DC source is connected it will have at least 2 amps available for charging the batteries. I want to keep things as efficient as possible so using an LM317 is probably not going to be good enough. So what to do about the different charging currents? Is there a way to put a limit on maximum current and still allowing the solar panel to charge at lower currents? For example if i use an lm317 to set a charging current of 2 amps. is it possible to use 100mA while on that setting since a 2watt solar panel can't provide that?


Also in the scrumari pdf linked above, he's using 5v for the entire circuit. In my case i need the availability for 12volts since each NiMh battery is fully charged at around 1.5v and with 8 cells thats 12volts total when fully charged. I may not be able to provide the batteries with exactly 12volts but i want to get as close as possible. So that means i will need maybe a mosfet with low on resistance so there is as little voltage drop as possible. I was thinking something like the IRF9510 or something better. So the change needs to happen where instead of using 5volts to charge the one battery i need 12 volts to charge 8 batteries.

Does anyone have an idea how to do any of this?


So far i have this:




EDIT: Also can someone please explain if the CC pin just goes from normally grounded to 5 volts with a PWM and duty cycle depending on charge state? When grounded it turns on the charging?

[the1cyrus]

Unfortunately I prefer not to use a micro-controller. i just want to get the pcb made and drop in components. I feel like i can learn a lot more this way.

[IanB]

If you put 8 NiMH batteries in series and try to charge them at 2 A or so your batteries will more than likely be going snap crackle and pop in short order.

Yes it is theoretically less than the 1 C rate for 2500 mAh, but with 8 in series you are really pushing the limits. I once tried charging a single AA cell at 2 A just out of curiosity and it really didn't like it. It popped its seal and vented gas.

If you really must charge 8 cells in series then I would recommend keeping the charging current below 500 mA and if possible put a temperature sensor in the pack and use temperature to help detect the end of charging.

[the1cyrus]

If you really must charge 8 cells in series then I would recommend keeping the charging current below 500 mA and if possible put a temperature sensor in the pack and use temperature to help detect the end of charging.

Oh i thought 1C was the standard safe number for charging. So 0.5 Amps is maximum you'd say? How did you come to that conclusion? Also the charger does have a thermistor connected to the TS pin. So if the temperature reaches 40C it will stop fast charging and switch to trickle charging or top off i believe. The digikey part number is BC2302-ND for the thermistor if anyone is interested.

Thinking about it the peak voltage for NiMH cells can be 1.6V, depending on chemistry. Even at 1.5V peak that's still 12V. You are going to need either a boost circuit or higher supply voltage.

Yep I am using a booster to give me 12v from the solar panel. I decided on LT1107 or TPS6734. The tps6734 may not have enough current. So probably the LT1107 is best since it can provide 400 mA so says digikey. Maybe someone can check my circuit below and see if it's correct? (having a hard time reading this datasheet.)

[IanB]

Oh i thought 1C was the standard safe number for charging. So 0.5 Amps is maximum you'd say? How did you come to that conclusion?

It's not quite simple, and there are various ifs and buts.

For a single 2000 mAh AA cell in good condition (like a standard Eneloop), the datasheet says you can fast charge at 1C (2 A). But this is for a single cell, and you must use minus delta-V to detect the end of charge condition and stop the charging. This is, however, the maximum charge rate that can be used, on the margin of what is recommended. As such, it is borderline between what is recommended and what is not recommended.

Now if you take a 2500 mAh cell of the same physical size, it will have higher internal resistance and more fragile internals (it must be so as a result of squeezing more "stuff" into the same volume). A 1C rate on this would be 2.5 A, but now you are putting a higher current through a more fragile cell. Really not recommended.

Next, consider charging cells in series. In this case, you are looking for a termination signal for the pack as a whole, but there is no guarantee that all cells in the pack will reach full charge at the same time. Generally the detectable termination signal will only happen when many or most cells reach full charge, which means that some cells may have reached full charge earlier. These particular cells will now be taking the full 2 A current while already fully charged. They will be getting hot, they will be generating gas and high pressure internally, and will tend to suffer internal wear and degradation. NiMH cells can take a moderate amount of overcharging, but not extreme overcharging.

So all told, the safe charging rate is somewhere below the maximum charging rate, and with more cells in series it is better to be more conservative. My experience of such cells suggests a charging rate of 0.5 A to 1.0 A for an AA cell is in the happy range.

[the1cyrus]

I understand now. thx.

Also will this work? in limiting the current when a 12v adaptor is used to charge the cells? I probably won't use those diodes because of the voltage drop. instead maybe Schottky diodes but the general idea is correct?

[the1cyrus]

Bought some of the parts. Going to test the circuit. If anyone has any answers to my questions let me know.

[the1cyrus]

How did you come up with 12.8v? Maximum voltage per cell should be 1.5 volts for NiMh and with 8 cells thats 12v.

[IanB]

The voltage needed by the charger depends on the specific brand and type code of cell, the charging rate and the charging algorithm used. When fast charging at constant current in the 0.5C to 1C range the cell voltage can reach over 1.6 V before termination. On the other hand, if trickle charging at low current the voltage may not even reach 1.5 V.

The good thing about limiting the peak charging voltage to 1.5 V per cell is that it reduces the risk of overcharging or extreme temperatures. That said, if you want to apply a constant current charge at 400 mA and allow the charger to do end of charge detection, you will need a significantly higher voltage. You need to allow for the drop out voltage of the charger and the internal resistance of the cells. A minimum of 1.7 V per cell would be advisable, possibly even 1.8 V per cell.

Now if you want a simple, cheap charging solution, you might be able to connect the battery directly to a 12 V supply with current limiting (by internal resistance, for example). By the time the pack voltage reaches 12 V the battery should be more or less fully charged and the current will drop almost to zero. It's not a recommended charging solution, but it's cheap and easy and will be fine as long as you limit the charging time to a day or less before using the cells.