TM4C Won't Stay in Hibernation

Hi Amit,

This is a custom board.

Thanks,

Eric

Hi Amit,

I don't think I considered the circuit properly for the hibernation module.  Please see the original schematic for the power section attached.  I just modified the circuitry because I really want the device to operate with VDD3OFF.  The current circuit now uses the HIB signal to control a transistor switch supplying power to VDD.  I have only one supply for both VBAT and VDD,  While VBAT is always on, VDD is now switched (it wasn't originally). 

The result is the same as before.  The device seems to enter hibernation and immediately comes back out.  The HIB signal is asserted for only about 20us before returning high.  This isn't enough time for my transistor circuit to fully disengage due to capacitance.

I did check the control register and the the VDD3ON bit is cleared before requesting the hibernation.schematic.tiff

Any thoughts on what might be happening?

Thanks,

Eric

Hi Amit,

It was for VDD3ON until I modified the circuit board this morning.  I haven't updated the schematic yet, which still shows the original configuration.  I now am using the HIB signal to switch VDD off.  This is the configuration with which I'd like to proceed.  Attached is the screen shot of the control register associated with the current configuration for VDD3OFF.  I'm still only seeing the HIB signal asserted for about 20 us.  Because this pulse is so short, VDD is never actually cutoff.  Is it normal for HIB to assert for only 20us and then reset if the power conditions don't match the configuration ( in this case, is it normal for the Hibernate module to automatically come out of hibernation if VDD isn't removed within 20 us?)  If so, I'll need to modify the circuit to ensure VDD is removed within this time period.

Thanks,
Eric

New Schematic.tiffHi Amit,

Please find an updated schematic attached.  VBAT is connected to the continuous 3.3V.  The VDD connections are tied to the switched 3.3V.

Thanks,

Eric

Hi Amit,

Please see attached.  Dark blue is the HIB and light blue is 3.3VSWITCHED.

By the way, I did reconnect WAKE to VBAT instead of 3.3VSWITCHED and didn't see any improvement.

Thanks,

Eric

Hi Amit,

I'm not sure I understand.  Are you saying this device does not hibernate?

Thanks,

Eric

Hello Eric

I ran the test case on my side and I got the following image. It is different from what you have. The HIBn signal remains low.

I am not sure why it is the case that for your board HIBn comes back up

1. Can you check the RTC Match value and RTC count when going to Hibernate mode?

2. Can you check the interrupt status register to see which bits have been set?

Hi Amit,

I'm using the hibernate module in calendar mode.  Here are the values for the calendar registers and match registers for the seconds, minutes, hours and day of the month before requesting hibernation:

Field                            Calendar                                Match

Sec                              111011                                    111000

Min                               011101                                    100010

Hr                                 111                                            111

DOM                            1010                                         1010

These values look okay to me as I requested a hibernation for 5 minutes.

After coming right back out of hibernation, the only bits set in the interrupt status register are the WC and RTCALT0. 

I'm surprised to see the RTC match flag is set because the all the time fields don't match.  Maybe I don't understand how this mode works.  It was my understanding that, when using calendar mode, all fields in the match register must match the corresponding calendar fields.  It would explain the immediate coming out of hibernation if only one field must match.  But, if this is true, how do you set up the module to come out of hibernation at a specific date/time?

Thanks for your continued support!

Eric

Hello Eric

Attached is a simple RTC wake only with VDD3OFF. I have tested it and it is working for me. The code attached can be used on your device to test if VDD3OFF is working as expected or not. The manner in which the code is written, is such that it goes into hibernate for 10 seconds and on wakeup it will be stuck in a while(1) loop of the default interrupt handler. Attached is the scope snapshot as well where the VDD supply ramps down. Clearly there is no 20 us time where the HIB comes back up.

#include <stdbool.h>
#include <stdint.h>
#include "inc/hw_hibernate.h"
#include "inc/hw_ints.h"
#include "inc/hw_memmap.h"
#include "inc/hw_types.h"
#include "driverlib/gpio.h"
#include "driverlib/interrupt.h"
#include "driverlib/pin_map.h"
#include "driverlib/hibernate.h"
#include "driverlib/sysctl.h"
#include "driverlib/uart.h"

void HibernateIntHandler(void)
{
	uint32_t ui32IntStatus;

	ui32IntStatus = HibernateIntStatus(true);

	HibernateIntClear(ui32IntStatus);
}


int main(void)
{
	uint32_t ui32HibCtl;
	
	SysCtlPeripheralEnable(SYSCTL_PERIPH_HIBERNATE);

	ui32HibCtl = HWREG(HIB_CTL) & HIB_CTL_CLK32EN;

	if(ui32HibCtl != HIB_CTL_CLK32EN)
	{
		HibernateEnableExpClk(16000000);
		HibernateWakeSet(HIBERNATE_WAKE_RTC);

		//
		// 5 Second match
		//
		HibernateRTCSet(0);
		HibernateRTCMatchSet(0,10);

		HibernateRTCEnable();
	}
	else
	{
		HibernateRTCDisable();

		//
		// 5 Second match
		//
		HibernateRTCSet(0);
		HibernateRTCMatchSet(0,10);

		HibernateRTCEnable();

	}


	HibernateIntEnable(HIBERNATE_INT_RTC_MATCH_0);
	IntEnable(INT_HIBERNATE);


	HibernateRequest();
	while(1);
}