MSP430FR2433: Wake-up time from LPM4.5 to active mode

Mauro Trespidi

Prodigy 230 points

Part Number: MSP430FR2433
Other Parts Discussed in Thread: MSP-FET, MSP430FR6989,

Hi,

I am testing an assembly code running in FRAM and composed by:

- a setup section at the end of which the MPU goes into LPM4.5 (with SVS disabled)

- a routine executed at wake-up from an I/O event, running with FLL disabled

- no other interrupt routine

MCLK = 4 MHz.

I would expect a wake-up time of about 1ms (D/S Table 5-3) plus 50us REFO start-up time (D/S Table 5-7).

I have instead a much longer wake-up time, while debugging with MSP-FET (hardware-only breakpoint/s, ULP advisor disabled)

Is there some other contribution to wake-up time that is missing in my computation?

Thanks in advance.

Best regards,

Mauro

over 3 years ago

0 Dennis Lehman over 3 years ago

TI__Guru 71216 points

Hi Mauro,

By chance are you using a 32kHz crystal connected to XT1? What sort of time are you measuring?

0 Mauro Trespidi over 3 years ago

Prodigy 230 points

Hi,

as a further detail, the delay I measure between the wake-up event (I/O low-to-high transition) and the execution after wake-up from LPM4.5 (LED switch-on) is about 1.5s.

Regards,

Mauro

0 Mauro Trespidi over 3 years ago in reply to Dennis Lehman

Prodigy 230 points

Hi Dennis,

I use the internal REFOCLK, no XT1.

Regards,

Mauro

0 Mauro Trespidi over 3 years ago in reply to Mauro Trespidi

Prodigy 230 points

Hi,

I found the solution to my problem in the following post, related to MSP430FR6989:

MSP430FR6989: Abnormal wake from LPM4.5 into LPM4 for one second? EnergyTrace++

It highlights, for that family, that there is a problem related to debugger: it adds a long delay when dealing with low power modes, as described in SLAU367P.pdf, § 38.2.7:

'In ultra-low power debug mode, the LPM entry and exit may be delayed. Low-power mode entry and wakeup from low-power modes is only possible while the debug protocol is in a certain state. Thus, the delay depends on the speed of the selected debug protocol. With Spy-Bi-Wire the delay is longer than when using JTAG.'

This is exactly what I just tested on MSP430FR2433: without debugger (SBW) the delay lowers from 1.5s to 548us!

I checked on MSP430FR4xx and MSP430FR2xx family UG but this behaviour is undocumented. It would be very useful to add such details.

Regards,

Mauro

**Attention** This is a public forum

MSP low-power microcontrollers

MSP low-power microcontroller forum

MSP430FR2433: Wake-up time from LPM4.5 to active mode

MSP430FR6989: Abnormal wake from LPM4.5 into LPM4 for one second? EnergyTrace++