Part Number: TPL5010
Please ensure you see and respond to the question at the end of this bug report.
We are using a TPS6197 Boost regulator to power an MSP430 based system. The TPL5010 is used to periodically wake up the system. This has been in production for a couple years but using a different boost regulator that had a slower VCC ramp up.
During periods of low activity the system runs directly off a Lithium coin cell (voltage between 2.0 to 3.0 volts). When higher processing speeds are required the boost regulator is turned on to bring the system voltage (including the voltage feeding the TPL5010) up to 3.3 volts.
We have verified that if the voltage on the VCC pin of the TPL5010 increases by more than 0.7 volts over a period less than 50uS the TPL5010 will generate a reset pulse!
For example, if the battery in our system is at 2.6 volts, then the booster is switched on, the voltage will rapidly rise to 3.3 volts, a jump of 0.7 volts. The TPL5010 will then reset.
The old boost regulator used to take about 500uS to ramp up and we never had this problem, even down to 1.8 volt input.
To test the theory a 100 ohm resistor was added in series with VCC and the power input of the TPL5010 and a 0.1uF capacitor was added to ground directly on the TPL5010 pin. This slowed down the rate of change of the power, but still not enough (plots were taken and attached).
The capacitor was then increased to 1.0uF (creating a time delay ramp of around 100uS) and voila, the system now works perfectly, even with input voltages below 1.8 volts (so jumps of 2.5 volts when the system ramps-up the voltage).
At no point does power on the TPL5010 chip fall below the datasheet value of 1.8 volts. The reset pulse is generated ONLY when VCC on the chip jumps more than 0.7 volts in less than around 50uS.
I suspect that the TPL5010 includes an internal power on reset circuit that is based on a series capactior which is accidentally tripping when VCC rapidly changes (Even when that change is within the operating voltage values of the part).
QUESTION: I do not want to use trial-and-error as a way to work around this apparent power on reset tripping bug. even though I have seen that a 100 ohm / 1.0uF capacitor combination on VCC slows down the transitions enough that the part is happy... I don't know if this is a good permanent solution. Please advise what is the acceptable VCC rate-of-change that will not cause this part to accidentally trip into POR mode.
Thanks
Trace 1 is VCC on the TPL5010, you can see that is is around 2.5 volts then jumps to 3.3V.
Trace 2 is the RST output from the TPL5010 which then trips (even though an appropriate kick pulse was sent just prior to this)
Zoomed out version of the same plot, you can see that the reset is only high for about 10mS (which is the amount of time the MSP430 takes to boot up and reach the point in it's code where it turns on the VCC booster)
Zoomed out even further, you can see that the distance between reset pulses is about 400mS, which is expected given the datasheet values of 320+100mS for the part to read Rext.
.