I’m using the MP2672A in a microcontroller circuit as a battery management IC. The basic circuit is below:
The ACOK and STAT lines are connected to microcontroller (STM32F413RGT6) pins that are setup as inputs with a pull up resistor. The datasheet says the pullup is 30K min, 50K max.
The microcontroller is powered by 3.3V.
When the battery is not charging (U_PWR is not connected) I’m seeing 3.3V on the ACOK line (as expected), but only 1.5V on the STAT line (I’m expecting 3.3V).
Is there a minimum pullup I need to use on the STAT and ACOK lines? Seems like 50K should be sufficient.
When I connect U_PWR, both ACOK and STAT go to ground as expected and the battery is charging.
Just a quick update. I disabled the internal uC pullup and put a pullup of 3.75K to microcontroller VDD (3.3V). I’m now getting about 2.7V on the STAT pin when not charging. The system is working, but there is leakage current into the STAT pin of 160uA ((3.3-2.7)/3.75K), which is way above the MP2672A datasheet leakage : 1uA max. The ACOK pin is behaving correctly with 1uA leakage as expected. What am I missing on the STAT line?
I checked the resistance between the STAT line and ground when the circuit is not powered and I’m getting between 3 and 4 Meg, so there is no short.
The STAT line connection is clean. I tried two different boards with exactly the same results.
Does it make a difference that the STAT line is pulled up to external VDD and not the MP2672A VCC line? It should not since this is what open drain is for.
Hi Igal,
Welcome to the MPS Forums.
I have a few questions that I would like to ask to help you debug this issue.
- I see you mentioned that the system is working, but have you observed the voltage waveform of the STAT pin? If charging is suspended it will blink at 1kHz and that could contribute to the lower voltage you are seeing.
- Please double check that both pull up and pull down is disabled on the MCU to use the external pull-up. Have you tried adjusting the pull-up resistor value further? A stronger pull up such as 2.2k may help pull all the way up to the desired 3.3V
- It can make a difference that the STAT line is pulled up to the external VDD and not the MP2672A VCC line. While the open drain needs the pullup to a specific voltage, different lines could introduce differing amounts of noise and there can also be some issues caused by power sequencing based on when individual rails turn on.
Feel free to reach out with any further questions or updates on this issue.
Best regards,
Zachary
Hi Zachary,
I put an external pullup of 3.75K to the STAT line and disabled the pull up / pull down on the microcontroller. This pulls the STAT line to 2.7V, which is sufficient for the uC to read the pin correctly.
With this configuration, the STAT line produces the correct output as outlined in the table you attached. I did not test the fault condition, but when the battery is charging, STAT is low. When the battery is done charging, STAT goes high. There is a spot in between when the battery is very close to being fully charged where the STAT line goes high and low intermittently, but not at 1KHz. I’m assuming this is because the internal logic of the MP2672A goes from “charging” to “charged” when the battery is close to full.
In any case, I still have a concern that with this configuration the leakage current on the stat pin comes out to 160uA, which is way above the stated 1uA in the datasheet.
From your response I understand that this may be normal behavior. Can you please confirm?
