PMIC TP65217BRSLR LDO 3 residual voltage

Hi Fabio,

Can you please go into more detail as to how you are turning the PMIC off, such as what registers you modify and how you are controlling the PWR_EN pin?

Hi Fabio,

By turning the PMIC OFF through i2c, the device switches the Powerpath to the battery and shuts down the regulators. However, if the other supplies are still present then some internal blocks will still remain active, which results in the higher quiescent current.

We don´t see any other supplies in our project present on PMIC OFF state.
I will re-send our schematics using our support here in Brazil. Maybe you can point to me some direction.

Hi Fabio,

I was referring to the power-path supplies such as a source on the AC pin. While this is present the PMIC will still consume some current for internal functions such as monitoring etc.

On our system, we detect the absence of AC and USB supplies starting a shutdown process. At the end, we schedule an RTC power off for PMIC. An issue will be if the operator returns the AC or USB cable during the time between detection of cable remotion and power off command. This action will create an state where battery will discharge with a current above our specification (maximun 14uA).
Is there any sugestion to turn aroud this behaviour?

"Do you find that removing AC again, after the power off command is issued, drops the current consumption back to the expected values? "

Yes. Drops to near of 7uA.

"If the device is expected to stay OFF even after AC has been reapplied, is keeping the AC line removed an option in order to meet your current requirement?"

The user can reapply the AC before the OFF command, like I describe in a previous post. With this, the system will appears OFF but will be discharging the battery above the specification for our product. The result will be a battery without power soon as we expected.
I don´t see any trivial solution without more hardware and software involved. We will use the product manual to guide the user to the process of power-off the equipament. Even if the user does the wrong procedure, we will be able to support this behaviour during the next cycle of battery charge.