We’re currently working on a project utilizing the MP2632 to charge and manage a single-cell lithium-ion battery (3.7V nominal). Our design has been validated with a functional board using the MP2632B, giving us confidence in the PCB layout and component selection.
However, after transitioning from the MP2632B to the MP2632 (To disable the Automatic Off at Light Load feature). We started experiencing an atypical behaviour: the chip oscillates in a repeated on-off cycle around 2.8 Hz. Although our board is powered by a stable 5V from a bench supply, we observed intermittent dips in the input voltage to the MP2632, which prevents the chip from achieving a full startup.
When swapping back to the MP2632B (same board), we encountered the same on-off cycling issue.
Has anyone encountered similar behaviour with either the MP2632 or MP2632B?
Is it possible that this is a result of the MP2632 chips being compromised during assembly (e.g., from ESD, overvoltage, or thermal stress), which could explain this persistent cycling?
So far, we’ve taken several steps:
Verified that the decoupling capacitors, sense resistors, and NTC resistor divider configurations are accurate.
Assembly only the power management side of the board to see if it was a load problem.
Reflow to make sure no bad pin connections.
Observed that the SW node does not show the expected high-frequency PWM pulses
Any insights from your experiences or recommended diagnostic tests to further investigate this issue would be greatly appreciated!
What else was changed between changing the MP2632B to the MP2632?
Do you have a schematic to show here? I know that you said the board was verified but I suppose a second look to further narrow down the root cause can’t hurt.
Have you checked if the SW node is completely inactive or just low? If it is inactive, then the IC may not be starting up due to some external condition. This would effectively be the determining factor between whether the IC or some other external factor is the cause.
The fact that you don’t see this issue after switching between different ICs leads me to believe the issue somehow stems from your board, supply, or perhaps some soldering issue. To start debugging, I would recommend starting with the following:
You mentioned and that the power supply is stable, but the dips that the waveforms show may be causing the MP2632 to reset. Measure directly at the IC (not just at the supply output) and check if there are transient drops.
Since both the MP2632 and MP2632B see the same issues, it would be worth checking for any micro-cracks or damaged pads at the solder joints.
You mentioned that you had verified the decoupling capacitors, but I would still suggest checking with an oscilloscope to ensure that the input is stable. Adding or removing any other components between swapping ICs may affect stability. Ensure that these capacitors are close to the IC as possible as well.
If checking the previous points are not fruitful, it could be that the system oscillates at 2.8Hz because of some sort of undervoltage or overcurrent protection triggering. Check the STAT and SYS lines to check if the charger is entering a fault state. On this note, ensure the power supply is also providing the proper current levels.
It may also be worth testing these ICs with another board in efforts to further isolate the issue if possible.
I know this is a lot to take at once. Please take each step one at a time and narrow down the possible cause here. Send schematics and SW captures (or any other necessary waveforms) if following these steps don’t narrow down the debugging process.
Keep me updated and I look forward for your response.
Let me know if there are any updates at all. If not, this post will automatically close. Feel free to make a new post and reference this one if in the case if your update comes in after things are closed.
