BQ76942: how to enter sleep mode without micro

Isee

Part Number: BQ76942
Other Parts Discussed in Thread: BQ40Z50

Customerm is using BOTHOFF functionality on the DFETOFF pin in order to disable power on the battery connector when the battery is unplugged from the device. But BQ76942 doesn't enter sleep mode when unplugging the battery.

Reading in the manual it seems a microcontroller is needed for that? That is not an option due to no uC being available.

Do you have any idea for that?

They thought about introducing a "fake failure" which would trigger the logic to turnoff the FETS? For example they could use a second thermistor input and design a small circuit that gives a pulse upon disconnection that gets interpreted as FET under temperature and triggers the protection.

Any other/better idea?

over 1 year ago

0 Max Verboncoeur over 1 year ago

TI__Mastermind 25980 points

Hi Isee,

This thread explains the conditions to transition to SLEEP mode: (+) BQ76942: Sleep mode - Power management forum - Power management - TI E2E support forums

In short, a uC is not able to transition the device into sleep mode directly. Instead, SLEEP mode is entered when the device detects that the magnitude of the current is below a user defined threshold. If the load/charger is removed from the PACK terminals, then the device should automatically enter SLEEP mode shortly after.

Regards,

Max Verboncoeur

0 Cyril Meister over 1 year ago in reply to Max Verboncoeur

Prodigy 80 points

Hi Max,

Thank you for your feedback. Unfortunately this doesn't quite address the issue I'm facing. I found a thread on the TI forum discussing the exact same problem I have, it's already 3 years old but unfortunately the resolution isn't posted. Please check: https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/925566/bq76952-sleep-mode-and-bothoff-functionality. We also wanted to use the predischarge feature. Thanks for your support.

Regards,

Cyril Meister

0 Max Verboncoeur over 1 year ago in reply to Cyril Meister

TI__Mastermind 25980 points

Hi Cyril,

See below, from Section 5.2.3.2.2 of the TRM.

The device will initially be blocked from SLEEP mode, however sending a DSG_PDSG_OFF() or ALL_FETS_OFF() subcommand should allow the device to enter SLEEP mode.

Could you try one of these subcommands and see if the device is able to enter SLEEP mode?

Regards,

Max Verboncoeur

0 Cyril Meister over 1 year ago in reply to Max Verboncoeur

Prodigy 80 points

Hi Max,

Thanks, yes, these subcommands work. However in our application there is no microcontroller on the BMS PCB. So this means I can't send commands once the battery pack gets physically disconnected from the device. I want to achieve the following sequence:

Battery pack gets disconnected --> BOTHOFF signal gets asserted --> all FETS turn off --> BQ76942 enters SLEEP.

As I can't send a command once it's disconnected, the only workaround I see would be to introduce a temporary protection fault condition that happens after the BOTHOFF signal gets asserted and therefore releases the BQ76942 from this blocked condition.

Regards,

Cyril Meister

0 Max Verboncoeur over 1 year ago in reply to Cyril Meister

TI__Mastermind 25980 points

Hi Cyril,

Is there anything preventing the customer from using SHUTDOWN mode while the battery is disconnected instead of SLEEP mode?

The customer could drive RST_SHUT high for >1s to trigger the device to enter SHUTDOWN mode without needing a host command. The device could then be woken from SHUTDOWN with a falling edge on the TS2 pin.

This FAQ discusses SHUTDOWN mode in more detail: [FAQ] BQ76952: How does SHUTDOWN mode work? - Power management forum - Power management - TI E2E support forums.

Regards,

Max Verboncoeur

0 Cyril Meister over 1 year ago in reply to Max Verboncoeur

Prodigy 80 points

Hi Max,

SHUTDOWN mode has been considered in the past but SLEEP mode together with BOTHOFF functionality would be much preferred. The motivation is actually to emulate the functionality of the "PRES or SHUTDN" pin there is on the BQ40Z50. It would be possible to achieve similar behavior also with SHUTDOWN mode but then there is the inherent 1s delay until the FETs turn off after disconnecting the battery pack.

Regards,

Cyril Meister

0 Max Verboncoeur over 1 year ago in reply to Cyril Meister

TI__Mastermind 25980 points

Hi Cyril,

You may be able to configure one of the multi-function pins as a thermistor, and have it normally connected to VSS via a fixed 10k resistor to simulate room temperature. When you need to trigger a protection fault to allow the device to enter SLEEP mode, you can short it directly to VSS to trigger an OTD/OTC fault.

Regards,

Max Verboncoeur

0 Cyril Meister over 1 year ago in reply to Max Verboncoeur

Prodigy 80 points

Hi Max,

Ok sounds good, I'll try that. Thanks!

Regards,

Cyril Meister

0 Cyril Meister over 1 year ago in reply to Max Verboncoeur

Prodigy 80 points

Hi Max,

I just tried if to achieve the desired behavior with the FET overtemperature fault. From the behavior I can see that it doesn't work if the FET overtemperature is only temporary. When I assert BOTHOFF and then FET overtemperature, the BQ76942 does go to sleep, however as soon as I remove the FET overtemperature condition, it wakes up again and doesn't go back to sleep. So the only way would be to keep the FET overtemperature condition constant as long as the battery pack is not connected to charger or device. However I worry if there could be any undesired side-effects when having the BQ76942 constantly in FET overtemperature condition while the battery pack is not in use.

Best regards,

Cyril

0 Max Verboncoeur over 1 year ago in reply to Cyril Meister

TI__Mastermind 25980 points

Hi Cyril,

The device will wake from SLEEP mode if a protection fault recovers, so it will have to remain asserted to keep the device in SLEEP mode. It should be fine to keep it in the fault state since the fault is only simulated, and the device will just keep the FETs off the whole time.

Another option is to use BOTHOFF to turn the FETs off while asserting RST_SHUT to trigger SHUTDOWN mode. This shouldn't need to have a simulated fault trigger to work. It is worth noting that the device settings would be reset to their default/OTP values after exiting SHUTDOWN mode.

Regards,

Max Verboncoeur

0 Cyril Meister over 1 year ago in reply to Max Verboncoeur

Prodigy 80 points

Hi Max,

Thanks a lot. As you say you don't see a risk with this, we'll try to work with the option of keeping the fault state asserted constantly until BOTHOFF gets de-asserted again.

Regards,

Cyril Meister

0 Max Verboncoeur over 1 year ago in reply to Cyril Meister

TI__Mastermind 25980 points

Hi Cyril,

Glad I could help!

Regards,

Max Verboncoeur