BQ21040: CHG Pin Doesn't go Hi-Z at EOC

Andrew Hawkins

Part Number: BQ21040

I was struggling with a different charger IC, so I switched over to the BQ21040DBVR. Right now it is Deadbugged and mod wired in place of the original charger IC.

The target is to charge a Sparkfun PRT-13851 battery with a USB connector as the VIN and monitor the charge status with an ATTINY88.

I have logged the charge and it looks like it is charging the battery fine but when the charge current gets below 10% of the Fast Charge current, the CHG does not go from LOW to Hi-Z as it should. However, if I unplug the charger and plug it back in, CHG goes into Hi-Z state.

This is a similar problem that I was having with the previous IC so there is likely a common problem that I have no idea how to fix.

Here is a simplified version of the schematic that shows the value choices and the interface with the ATTINY88.

[url=https://postimg.cc/w39kCfxB][img]i.postimg.cc/.../url]

Since the CHG pin has a light pull-up to VIN, I feel like I need to divide the signal down to 3.3V levels so I don't fry the ATTINY88 pin with 5V. I simulated the resistor configuration on EasyEDA, and it should put the correct voltages for VINH and VINL.

Is this the correct way to interface with a processor?

Is there something else that I am missing?

Thanks!

9 months ago

0 Andrew Hawkins 9 months ago

Prodigy 10 points

I am having real issues trying to get my schematic posted.

0 Michael Emanuel35 9 months ago in reply to Andrew Hawkins

TI__Mastermind 35985 points

Hello,

Please post your schematic so I can better assist you. It is likely that when you power cycle the adapter that the battery is not below the recharge threshold VRCH so charge does not start again. Can you try and drain the battery voltage below the recharge threshold, reapply the input and see the result?

Regards,

Mike Emanuel

0 Andrew Hawkins 9 months ago in reply to Michael Emanuel35

Prodigy 10 points

0 Andrew Hawkins 9 months ago in reply to Andrew Hawkins

Prodigy 10 points

Wow! I was actually able to just paste it this time.

This is the simplified schematic because, in the grand scheme, the processor and the 100K pull up to 3.3V is on a separate board that is connected to the board with the BQ21040 through an RJ50 cable.

0 Michael Emanuel35 9 months ago in reply to Andrew Hawkins

TI__Mastermind 35985 points

Hello,

My schematic review is as follows:

1. This is incorrect setup for the CHG/ pin. It needs to be pulled up to a single rail with no voltage divider. Use appropriate resistor pullup.

2. The minimum bypass capacitor for VIN is 1 uF.

3. I confirm the charge current is 385 mA. Can you please include some scope capture or measurement where you are not terminating at 38.6 mA?

Regards,

Mike Emanuel

0 Andrew Hawkins 9 months ago in reply to Michael Emanuel35

Prodigy 10 points

Hi Mike,

I am working on verifying my measurements right now using different DMMs. Once I can verify that my measurements are good, I can get a log of charging current and voltage.

I will also change the VIN cap to 1uF or even 2.2uF to reuse the same cap as the Output.

Thanks!

-Andrew

0 Michael Emanuel35 9 months ago in reply to Andrew Hawkins

TI__Mastermind 35985 points

Hi Andrew,

Looking forward to your results.

Regards,

Mike Emanuel

Power management

Power management forum

BQ21040: CHG Pin Doesn't go Hi-Z at EOC