Hi I have this battery pack, 20x1,2v NiMH cells. I have replaced the original 1300mAh cells with new 1600 mAh cells, now the charger doesn't start charging anymore and signal an error.
2 guesses:
1) the IC had some data in volatile memory which was lost when I disconnected the cells
2) the IC regognizes the different electrical characteristics of the new cells as too different from what it expects and interpret it as cells failure
I need to understand what this IC is and how it works in order to attempt a fix, I've searched for days for "25 BRPI" "25 8RPI" "25BRP1" and all possible interpretations on Texas Instruments website and all internet but no joy.
Can someone help? Thanks!

 

Only one i know of from past experience is the BQ24401.

Several listed on their web site...

http://www.ti.com/lsds/ti/power-management/battery-charger-ic-products.page#p338=NiMH

 

It might be a custom part or custom version of a part and it is marked with a custom part ID.

Have you tried going to the TI website and checking the datasheets of their charge controllers?

 

Hi
thanks, yes I tried, ther's nothing with this id on texas instruments website. Bit strange they put TI logo on a custom part tough...
Any other suggestion?
Do you know a cheap charger I can use to charge this battery pack? If I can't convince the original charger to charge this battery may be i can use a simpler charger

 

I think you will want the attention of the battery charging experts. In that case the best way to attract their attention is to start a fresh thread in the Projects forum with a title indicating that you need a charger for a battery pack of 20x1.2v NiMH cells.

 

Can you possibly draw the circuit out roughly,?
as to how the chip is connected to the cells, voltage pins, thermistor, maybe it only uses 8 pins and was put into a 10pin package for convenience.

 

Only one i know of from past experience is the BQ24401.

Looks like a Bequerel part number which I think TI probably own - either they've killed off the name or I can't spell it right.

 

Hi
thanks, unfortunately it's impossible to draw the circuit from the pcb, most tracks are on the same side of the components and are covered by paint. It's certainly possible that 2 pins are not used, but some tracks may connect to pins from the inner side so can't be sure. The termistor is those 2 black wires on the bottom right, so it seems to be connecting to a pin on the bottom right of the ic.

 

From another forum they suggested this: http://www.ti.com.cn/cn/lit/ds/symlink/ads1015.pdf
And I think it's possibly right, since TI doesn't seem to make any battery management IC for 20 cells in series, it may well be just an AD converter sending voltage and temperature data to the charger via i2c.

 

Looking at your pcb, there are two sets of identical fets, resistors, caps, I'm wondering if the cells are charged in two batches of 10, with a series current sense resistor in each, and one thermistor?

 

If the voltage of the new batteries exceeds the protection limit of the chip, then it will go into a protection mode.

Also, if there is lack of current flow because of bad soldering, the current sensing circuitry will not detect an appropriate response when the chip applies current to the batteries, it will throw an error and shutdown charging.

Finally, if your batteries are discharged to a very low level and the charger is supposed to "auto-detect" how many cells are in series, it may be tricked into detecting a smaller string of batteries than you actually have and going into protection mode (like paragraph 1 (above)) - because the logic makes it think it is over changed.

Best solution, your new batteries should each be changed/discharged to the nominal voltage of the cell before connecting them. Then assemble and charge as a group.

 

Looking at your pcb, there are two sets of identical fets, resistors, caps, I'm wondering if the cells are charged in two batches of 10, with a series current sense resistor in each, and one thermistor?

yes it can well be! indeed there is a pack/2 yello wire which is connected in the middle of the pack!
@GopherT the voltage should be the same really, what is different is the ampere/h old cells were 1300mAh new ones 1600.
I think the only solution is to tap into the i2c interface and try to understand what the ADC sends to the charger.
It's an ADC for sure, looks exactly the ame as this one https://www.adafruit.com/product/1083
But this ADC doesn't seem to have any volatile memory, so that's not a possible explaination for the battery stopping to work after replacing the cells.

 

If it's only the current rating that has changed, then it's the Current Sensing that needs to be changed.

 

but how can it sense the current? the battery pack is connected to the host AND to the pcb in parallel, the current doesn't flow trough the pcb.

 

But this ADC doesn't seem to have any volatile memory, so that's not a possible explaination for the battery stopping to work after replacing the cells.

or may be the charger is programmed to kill the battery by reprogrammin the ADC once it decides end of life is reached.

 

yes it can well be! indeed there is a pack/2 yello wire which is connected in the middle of the pack!
@GopherT the voltage should be the same really, what is different is the ampere/h old cells were 1300mAh new ones 1600.
I think the only solution is to tap into the i2c interface and try to understand what the ADC sends to the charger.
It's an ADC for sure, looks exactly the ame as this one https://www.adafruit.com/product/1083
But this ADC doesn't seem to have any volatile memory, so that's not a possible explaination for the battery stopping to work after replacing the cells.

Yes, the same once the new battery pack is within the same voltage range as the old one. If an over voltage alarm is tripped because a "fresh-from-the-factory battery can be "overcharged" by some definitions. Likewise, a poor solder joint on the battery or to the board will also cause an error.

[sarcasm]Probably not worth looking at battery-related options though and best to focus on the IC (since the IC was the major change to your circuit since the last time it was known to work).

Also, I agree with you that it must be an ADC since it looks like a 10-pin SIOC package on a device labelled "ADC". What else would it be?[/sarcasm]

 

or may be the charger is programmed to kill the battery by reprogrammin the ADC once it decides end of life is reached.

You might want to check what fraction of ADCs have volatile memory to validate your theory.

 

but how can it sense the current? the battery pack is connected to the host AND to the pcb in parallel, the current doesn't flow trough the pcb.

Which is why i asked you to draw the layout of the wiring, or better still take better pictures, we can't work with dumb information!!

 

tanks for the new replies
Gopher the IC is marked "25 BRPI" so it took a suggestion from another forum to discover that it's an ADC, I agree the battery was the only thing I changed but what can I do on the batteries? Solder joints on the pcb are too microscopic to check, on the new batteries they are fine. From what I understood of the ADC datasheet there is no volatile memory in this ADC. If I'm wrong correct me.
Dodgydave I'll try to draw something in the next few days to give you a better idea.
Thanks for helping.

 

or may be the charger is programmed to kill the battery by reprogrammin the ADC once it decides end of life is reached.

My E-cigarette batteries are from laptop packs rescued from recycling - I've just replaced a set that I've been using over 3 years.

People don't bin them because they like paying for new ones - they do it because one of the chips stops the thing working.