BQ76952: BQ76952: exit deepsleep mode

Before proceed to NORMAL mode, it first completes a full measurement loop and evaluates conditions relative to enabled protections, I am wondering during this period (it seems to take 250 ms and all FETs are OFF), what conditions relative to enabled protections are evaluated?

Hello Luis H.S

In my test, I find that when I send the 0x000E EXIT_DEEPSLEEP() command, the device immediately change the Control Status() [DEEPSLEEP] from 1 to 0, and still all the FETs off, after about 250ms, that is what says "during this period (it seems to take 250 ms and all FETs are OFF)", then it will change the Alarm Raw Status() [WAKE] from 0 to 1, then the device will open all the FETs and then it start to have current flowing across the load.

According to your answer ,when I send the 0x000E EXIT_DEEPSLEEP() command, why the device will not  immediately open all the FETs but keeping all the FETs off?

In my test, sometimes, after I send the 0x000E EXIT_DEEPSLEEP() command, the device immediately change the Control Status() [DEEPSLEEP] from 1 to 0, but before changing the Alarm Raw Status() [WAKE] from 0 to 1, the device sometimes may do a reset, I do not know why it will do a reset?

Hello Yubin,

FET's may take up-to ~250-ms to turn ON. This is normal, every time the FETs turn ON, it takes up-to 250-ms.

Can you specify what you mean by a RESET?

Best Regards,

Luis Hernandez Salomon

In my test，I find that after I send the 0x000E EXIT_DEEPSLEEP() command, before changing the Alarm Raw Status() [WAKE] from 0 to 1, the INITCOMP sometimes may change from 1 to 0 and then back to 1, so I guest it do a reset at certain case, but I did not send any reset command or use the RST_SHUT PIN to do a reset during this period, so I am wondering what are the conditions that may cause the device to do a reset?

Hello Luis H.S

According to your answer, I make a test using the TI’ BQ76952 EVM boards，all the FETs start to open when changing the Alarm Status() [WAKE] from 0 to 1, And all the FETs only take less than 2ms to turn ON, it need much less than 250ms to turn on all the FETs？ and immediately after change the Control Status() [DEEPSLEEP] from 1 to 0, all the FETs off and do not start to open becasue CHG and DSG stay the same voltage as that are in the deepsleep mode.

Hello Yubin,

I am a bit confused on your question. Can you clarify?

Now, the FETs turning ON tales up-to 250-ms. This means that it can take any time up to a typically upper time of 250-ms, this means it could turn-on before 250-ms. The device checks what the FET state needs to be every 250-ms, so depending on where it is on this checking cycle, it could turn-on before 250-ms.

Best Regards,

Luis Hernandez Salomon

Hello Yubin,

Usually in DEEPSLEEP, most of the internal blocks are powered down. So when device exits DEEPSLEEP most blocks power-on again.

As I mentioned, the device evaluates when to turn-on the FETs every 250-ms. So it is likely that this 250-ms cycle starts when the device exits DEEPSLEEP mode. This means that it would take ~250-ms for the FET's to turn-on, then the driver strength is the additional delay, so the 2-ms is just the actual driver turning the FET's ON.

Do you require faster turn-on than 250-ms?

Best Regards,

Luis Hernandez Salomon

Hello Luis H.S

According to your answer, whenever the device exits DEEPSLEEP, it will go straight into NORMAL mode.

Firstly it will power on the device again, and becasue the device evaluates when to turn-on the FETs every 250-ms, it would take ~250-ms for the FET's to start to turn-on, after the additional delay, the FETs are fully open.

Then after about 250ms, it start to make and completes a full measurement loop, and evaluates conditions relative to enabled protections, to ensure that conditions are acceptable to proceed to NORMAL mode, so the BQ76952 RTM writes " This may take ≈250 ms plus the time for the measurement loop to complete." right?

Becasue BQ76952 RTM write "evaluates conditions" and "are acceptable to", so I think "is there any condition that will stop to proceed to NORMAL mode?" If the device was stopped to proceed to NORMAL mode, it means it go into sleep mode or deepsleep mode?

But you said "whenever the device exits DEEPSLEEP, it will go straight into NORMAL mode." 

Hello Yubin,

yubin zhang said:

According to your answer, whenever the device exits DEEPSLEEP, it will go straight into NORMAL mode.

Correct.

yubin zhang said:

Firstly it will power on the device again, and becasue the device evaluates when to turn-on the FETs every 250-ms, it would take ~250-ms for the FET's to start to turn-on, after the additional delay, the FETs are fully open.

Correct. I believe you mean FETs are fully closed (This means FETs are turned ON), open implies the FETs are OFF.

yubin zhang said:

Then after about 250ms, it start to make and completes a full measurement loop, and evaluates conditions relative to enabled protections, to ensure that conditions are acceptable to proceed to NORMAL mode, so the BQ76952 RTM writes " This may take ≈250 ms plus the time for the measurement loop to complete." right?

No, the measurement loop occurs as soon as the device exits DEEPSLEEP, it does not wait for the FETs to turn-on. So the part exits DEEPSLEEP, then takes a measurement loop, which takes much less than 250-ms. Then if conditions allows FET's turn-on after the 250-ms cycle has ended, which happens simultaneously to the measurement loop.

yubin zhang said:

But you said "whenever the device exits DEEPSLEEP, it will go straight into NORMAL mode." 

Correct. When the part exits DEEPSLEEP, it will first enter NORMAL mode before entering SLEEP mode.

Best Regards,

Luis Hernandez Salomon

Hello Luis H.S

Luis H. S. said:

No, the measurement loop occurs as soon as the device exits DEEPSLEEP, it does not wait for the FETs to turn-on. So the part exits DEEPSLEEP, then takes a measurement loop, which takes much less than 250-ms. Then if conditions allows FET's turn-on after the 250-ms cycle has ended, which happens simultaneously to the measurement loop.

According to your answer, the measurement loop which occurs as soon as the device exits DEEPSLEEP is used to decide whether to allow FET's turn-on after the 250-ms cycle has ended, right?

If it is right, at what conditions the device may not allow FET's turn-on after the 250-ms cycle has ended and keep the FET's OFF?