BQ24103A - Only works with at least 14V VIN. - Battery Management - Chargers Forum - Battery Management

BQ24103A - Only works with at least 14V VIN.

This question is answered

Posted by Carl-Johan Sjoberg

on Apr 25 2012 07:21 AM

Prodigy30 points

BatteryManagement.pdf

Hi,

We've a design with the BQ24103A, please see attached pdf for schematic.

The problem we have is that it sometimes works like a charm and sometimes it enters some linear mode drawing between 30mA to 250mA on the 12V rail.

Especially when the cell voltage is low. The voltage is clamped to the cell voltage.

In order to get it working I need to crank up the VIN voltage to 13-14V and it will start battery detect mode. It doesn't matter if the CELL pin is pulled low or high.

Vbat will be approx. 12.5V.

On the schematic Vchg is 12V. Vbat should be 8.4V, 2 cells mode. LiPoly cells has built-in protection.

All our prototype boards have this symptom. I've scoped all inputs with a 100MHz/2GS oscilloscope and all signals are clean and I've checked the marking on the chip and it's CKO and from the same reel.

I've fiddled around with different inductors and Cout caps but the same result.

I'm a bit confused now so maybe a TI engineer can give hints/suggestion to fix this or/and explanation of the this behavior.

Gerber and PCB from Altium Designer can be sent to a private email.

And a last question, what is the difference between BQ24103 and BQ24103A? Because of 2008 there is no need for a Zener on neither of them.

Kindly Regards,

Carl-Johan Sjoberg

All Replies

Posted by Wang5577

on Apr 27 2012 10:41 AM

Genius12450 points

So far, I don't find the problem on the schematic. But, I don't know your test condition also. Let me summarize your test condtion. And please give more info.

1. Vin=12V, input current is about 30mA~250mA. What is the battery voltage? Is it lower than pre-charge threshold?

2. When Vin=13V~14V, Vbat will be approx. 12.5V. This is not right. Does IC get damage or bad soldering?

bq24103 is pin to pin compatible as bq24103A. The performance should be same. They both don't need a zener after 2008.

Posted by Carl-Johan Sjoberg

on Apr 30 2012 05:42 AM

Prodigy30 points

We've tracked down a couple of issues.

1. Suspecting transients on VCC that breaks the internal LDO. VTSB is either ~9V or 0.7V.

2. Suspecting transients on PG, STAT1 and STAT2 as these pins are not pulled up during initial testing.

I'll swap the BQ chip and pull-up PG,STAT1 and STAT2. Also I'll add a 15V Zener to VIN and VCC.

If we have the batteries connect they will clamp the voltage and after a while the over-voltage protection in the batteries will disconnect the batteries.

First step in testing is to apply +12V and see that we've pulsing on OUT pin. Battery detection.

Second is to check wave form on Cttc.

After that we connect the batteries. Usually around ~7V.

Charging begins and everything is ok.

Next step is to remove the +12V and remove the batteries and do the testing again.

After this step all our BQs broke. So we suspect that transients is the problem.

I'll let you know the outcome of the tests.

Regards,

cjs

Posted by Carl-Johan Sjoberg

on May 08 2012 11:08 AM

Verified Answer

Verified by Carl-Johan Sjoberg

Prodigy30 points

I added a BZX84C15 diode, 15V zener to VIN/VCC, made sure that PG, STAT1 and STAT2 is always pulled up and swapped the 27uH inductor to a 22uH to increase the margin to Isat.

Now it works as expected.

Regards

cjs