Workaround for PMM6

From Texas Instruments Embedded Processors Wiki

Jump to: navigation, search
Translate this page to   

The PMM6 Erratum

Module: PMM, brownout startup problem

At low temp (less than -20C) the device is likely to not work correctly, after brownout condition occurred. A brownout condition is defined as drop on the DVDD supply to DVSS level and shorter than 2 seconds.

Workaround: None

How PMM6 affects an application

This effectively states that device startup and BOR recovery cannot be guaranteed under temperatures of -20 Celsius. Following the startup or BOR condition, a value of ~.8V can be observed on the VCORE pin if the unit has failed. The MSP430 will also not execute any code.

All units are affected, but the failure is probabilistic and the temperature at which devices fail below -20C varies between devices. The same stimulus could or could not result in a failing condition. This behavior cannot be screened at production test.

The PMM6 Wokraround

The only known workaround includes an external SVS with a programmable delay whose RSTn line is tied to an IC load switch that changed the Vdd of the MSP430 between Vcc and GND. If the external voltage dips below the operating range of the device (1.8 V) Vcc is grounded. If the voltage returns to a value > 1.8V, there is a two-second delay before Vdd of the MSP430 is once again connected to Vcc. This circuit has been tested and proven to prevent the failure on 18 units.

18 failing units were identified using a temperature chamber by cooling the units down to -40 Celsius and applying power. After using the workaround to supply Vcc to the MSP430, all 18 units passed the test and the processors started up as expected.

PMM6 workaround.png

One needs to keep 100nA or more of current flowing in the resistor divider to the sense pin in order to maintain the sensing accuracy.

Leave a Comment
Personal tools
Namespaces
Variants
Actions
Navigation
Print/export
Toolbox