bq2002T - ending fast charge too quickly

Can you look at the IC and make sure it is a bq2002T?  Actually, the T version does not have -dV termination.  Possible fast charge terminations are explained on page 1 and 4 of the datasheet.

Could you post your schematic?

Why do you think that the pack is not full and terminating properly?

Hi

You obviously have the TM pin set to 0V.  This is the setting required for charging at a rate of 2C (i.e. 4.2A for your 2100mAH battery).  At this rate a flat battery will be fully charged in 30 minutes, at which time it will terminate due to dV/dt.  The bq2002C incorporates the additional backup that the charger will terminate anyway after 40 minutes if TM = 0V.

If you charge at a different rate TM shoul be set accordingly.  For example, if you are charging at 1A (i.e. approx C/2) it will take 2 hours to charge a flat battery.  TM should be set to mid-rail.  The charge will then be terminated after 160 minutes, or 2.5 hours, regardless of dV/dt.

A secondary function of the TM pin is to set the trickle charge mode.  Once a NiMH of NiCd battery is fully charged it should be trickle charged at about C/32 to overcome self-discharge.  Thus a 2100mAH battery should be trickle-charged at about 65mA.  The bq2002 satisfies this by pulsing the charger at a defined duty cycle, determined by the TM pin.  For example, if TM is mid-rail the charger will be activated for 73ms every second.  On the assumption that the charge current is 1A as above, this gives an average trickle charge current of 73mA.

Alan, Chris, good morning.

I haven't my schematic, I'm on holiday, but it's simple. I made your dev. Board DV2002L2/TL2 on my own. Only I've added a uC (USB) to read the CC signal and I've an ADC to read the battery voltage.

This is the simple scenario. I've a BQ2002T aand TM=0 and ten cells. The VBAT start with (i.e.) 1.4V (14V fot total pack) and during fast charge the voltage rise.

I'm using a 1,86A for fast charge. In 1h and 20' (80' of TM =0) I should charge totally the pack. Also if I'm not using 2.1A this is not the point.

The problem is: after a 30/40 min (the temperature is ok) the voltage is about 1.51V and your device terminate the fast charge.

It goes in top-off. I leaved it in top-off and after that i did a discharge fase with high power resistor.

The pack seems to be charge at 60/70%.

Is the pack a 2100mAh, fake? Could be. Or theT/t termination is not so accurate for some NiMH Pack?

Could I use another BQ2002x (PVD,-AV)? 

Thanks for your answer.

Can you look at the Vbat pin with a scope?  Does it ever reach 2V?  Also, can you measure your charge current?

1.4V per cell is quite high to begin with, so maybe your pack is very charged initially.  It should be full at around 1.5V per cell.

The only way the 2002T will terminate is through a dT/dt, max temp, max voltage, or charge time.

What is your input voltage?  It could be that as your pack charges, your Vin is too low to support the charge current.

Here is an app note that shows the various bq2002 options: http://www.ti.com/lit/ug/sluu229/sluu229.pdf 

Yes Chris I can do that. I did that. The VBAT pin didn't reach 2V absolutely! The Vin was 22V and the charge was a 1.86V (1.27V of LM317 with a R sense of 0.68 Ohm);

 I've a power supply unit with 2 display for voltage and current supplied....

It's all ok, it's a normal charging. ;)

1.4V of voltage was only an example.

I'm on Holiday. When I will come back in my office I will do this test: I will charge the battery pack at 0.1C and discharge it at 0.2c like the manifacturer recommend and I test if this pack has that capacity or not.

I've some doubts that a pack charged with 1C current could reach 100% of nominal capacity. Do you know something about that?

After that I will do others test to find a solution and don't throw away all this work. 

I suppose to test the Bq2002F (at 1C with PVD) that I bought for error for a comparison with the previous tests.

Talk to you soon.

Thank you for replying with the resolution of this issue--that your pack had a different capacity when charged differently from the manufacturers recommendations.

The voltage divider range is only a recommendation and a starting point for the values.  If it is less than 200k, then the current through the divider when the charger is off could become significant relative to the battery's self discharge.  Greater than 1 Meg and the bias current into the Vbat pin could produce offsets that affect the voltage measured.

If you have a second divider on the battery voltage, this is independent from the first Vbat divider.  But it too will have a minimum and maximum resistance based on how much current you can allow to flow through it when the charger is off and what the input current to your analog input pin is.