This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Part Number: MSP432P4111
Tool/software: TI-RTOS
Hello TIers,
I have a question regaring the power management of the MSP432P411 when used in conjunction with TI-RTOS, but first a little background.
According to the sprui18h document (SimpleLink SDK Power Management), for the old MSP432P401 the default policy managent is the Power_sleepPolicy whereas for the new MSP432P411 series the default policy management is the Power_deepSleepPolicy (pag. 16), similarly to the CC26xx/CC13xx series.
On another section of the same document, it states that for the MSP432P401 devices the clock ticks halt during deep sleep states, hence not being able to wake-up from the Clock module. However, for the CC26xx/CC13xx series the clock ticks do not halt during deep sleep states, hece the device can wake-up from deep sleep and resume task exectution.
So the question is, does the MSP432P411 behave like the MSP432P401 or like the CC26xx/CC13xx in terms of clock operation during the deep sleep states?
Thanks in advance.
Pere
Hello,
I cannot speak to the CC26xx/C13xx but the deep sleep mode of the MSP432 devices means that the CPU is off and consequently Systick is not available. With the MSP432P401 device the wakeup sources from LPM3 are the WDT, RTC, and GPIO. The MSP432P411 includes these same sources along with the communications, timerA, and the clock system (LFXT oscillator fault). (a complete list of interrupt sources can be found here).
If you look at the examples for powerdeepsleep for both the P401 and P411 you will see that they are the same. Please refer to this example for power policy.
Regards,
Chris
Thanks Chris for your reply. However, I still don't see it clear.
As stated in the SimpleLink SDK Power Management (SPRUI18H) document:
- For MSP432P401x devices, the Deep Sleep state is not enabled by default because DEEPSLEEP_0 and DEEPSLEEP_1 disable peripheral modules and cause Clock ticks to be suspended during sleep. For MSP432P401x devices, when you use the Power_deepSleepPolicy(), clocks and peripherals are forced into a clock-gated state during deep sleep. They resume activity when the device wakes up.
- For MSP432P4x1xl devices, peripherals can continue to run during deep sleep states, so Power_deepSleepPolicy() is the default power policy.
Now, my concern is about the TI-RTOS tickless mode to achieve lower energy consumption.
- With the CC26xx/13xx the OS can calculate the next wake-up time from a clock timeout and then go into deep sleep mode and wake-up using the RTC at the exact time.
- With the MSP432P401x the RTC only works in calendar mode at 1 second interval and the WDT in interval mode only offers limited timeout functionality, i.e. 1.95, 15.65, 250ms, so the system needs to wake up periodically to keep track of time.
So my question is: with the new MSP432P4x1xl devices using TI-RTOS is it possible to achieve true tick suspension (as with the CC26xx/CC13xx) or does the CPU still need to wake-up periodically (as with the MSP432P401)? If it is possible, do you mean that you can use another timer to implement the tickless mode? In that case, what would the average current consumption of MSP432P4x1xl devices as compared to the CC26xx/CC13xx series?
Best,
Pere
Hello,
The MSP432 devices use TIMERA0 for the TI-RTOS tick. As you have pointed out, in the case of the P401 devices this clock is not available in LPM3 (deepsleep). The tick suspension (Dynamic mode) is described here.
The current is found in the datasheet. Please note however, that the peripherals are divided into islands so the peripherals that are also on the same island as timerA0 will also contribute.
Regards,
Chris
Hello Chris, thanks for your answer!
It is now clear to me the difference in LPM3 between the "old" and the "new" MSP432 devices and how that enables TI-RTOS tickless mode.
The only pitty is that the information is so buried down in the datasheet (page 135) that it is not easy to find while comparing both devices.
Being such a nice feature with regard to low power operation (it can save several micro-amperes while allowing the system to wake-up with micro-second granularity) I would suggest to add it in the datasheet front-page.
Again, thanks for your clarification.
Regards,
Pere
**Attention** This is a public forum