BQ76952: CHG FET misbehave when FETs are ON and send the command to enable both FETs (0x0096 ALL_FETS_ON()),

Hi Jinal, 

The behavior you mentioned does seem odd but we are not seeing the same on our side.

Could you please provide the exact command sequence you are sending, are you able to verify that the FET_CTRL_EN bit is set in the FET Options register? If you happen to have the EVM for this device, it would be helpful if you can send the .gg file to check the settings of the device. 

Best, 

-Luis Torres

Hi Luis,

Thank you for quick response.

We are not using EVM, We have our own BMS with BQ76952.

No we have no .gg file.

Yes FET_CTRL_EN is set. 

we set the value of FET Options is 0x1F.

Thank You 

Jinal

Hi Jinal, 

Are you able to go through the exact command sequence for each of your scenarios? 

We need to verify the sequence since we are unable to recreate the same behavior on our board. 

Best, 

-Luis Torres

Hi Luis,

Yes,

Exact seen was when charge FET is ON and discharge FET is OFF that time if we send command for enable both FETs (0x0096(ALL_FETS_ON)).

Also in BQstudio when we send the the command for ALL_FETS_ON, we observed same scenario.

PFA .gg file for your reference.

sendToTI.gg.csv

Thank you

Jinal Ubhatar

Hi Jinal, 

We would still need exact test procedure as this is still not clear. 

By default the CHG and DSG FETs are ON. So if you are starting with CHG FET ON and DSG OFF, how are you disabling the DSG FET? 

Best, 

-Luis T

Hi Luis,

BMIC is in idle mode and the FETs status is

1. Charge FET: ON

2. Discharge FET: OFF

Now, In our case charge FET is already enabled and discharge FET is disabled, now we want to enable discharge FET and charge FET, for that we send the ALL_FETS_ON(0x0096).

We assume that after send the ALL_FETS_ON(0x0096) command discharge FET will be turned ON and no impact on charge FET because charge FET is already enabled before sending the command.

But our observation is once discharge FET is enabled, after 50ms the charge FET is disabled for 200ms and again turned on.

Yellow channel define charge FET status.

Blue channel define discharge FET status.

Note: We are not sending any other commands and doing any activity.

Thanks,

Jinal

Hi Jinal, 

We are still unable to recreate this behavior even with the settings you have sent. Please help by providing feedback to the points below. 

1.) What happens on startup prior to this "idle" mode? You mention: "In our case charge FET is already enabled and discharge FET is disabled" how is this so?

• Note: The FETs on their own will not be defaulted to this state (CHG FET ON & DSG FET OFF) unless commanded to.

2.) Is this behavior observed if a different BQ76952 IC is used? 

3.) Can you please provide the schematic for review?  

Best, 

-Luis Torres

Hi Luis,

1. Yes we enabled charge FET and Disable Discharge FET using command(Default state means after sending command for CHG FET ON & DSG FET OFF after that try to enable ALL_FETS_ON(0x0096), where we observed this behaviour).

2. Yes this is same behaviour in other BQ76952 also.

3. Yes we will provide a schematic shortly.

Thank you 

Jinal Ubhatar

HI Jinal, 

Today we have tested the procedure. 

Start BQStudio -> Load provided .gg file -> FET_EN (0x0022) to allow both DSG and CHG FETs to turn ON -> DSG_PDSG_OFF (0x0093) to disable DSG FET -> ALL_FETS_ON (0x0096). 

Observe the DSG FET pin on the oscilloscope and do not see it turn off at any point after 0x0096 command has been asserted. 

We will wait for schematic for next steps. 

Best, 

-Luis Torres

Hi Jinal, 

I forgot to note that we did notice that the charge pump was disabled on the settings you had provided. Was this intended? 

Best, 

-Luis Torres 

Hi Luis, 

Please observe CHG FET pin because once DSG FET is enabled, after 50ms the CHG FET is disabled for 200ms and again turned on.

Thanks,

Jinal Ubhatar

Hi Jinal, 

That is correct, I have mistakenly noted "DSG FET" was observed but it was in fact CHG FET pin that was observed on the oscilloscope. Apologies for the confusion. 

We will wait for schematic for a better understanding on what may be causing this issue. 

Thank you, 

Best, 

-Luis Torres

Hi Luis,

PFA.

Thanks,

Jinal Ubhatar

Hi Luis,

Yes intentionally disabled the charge pump because we are using Low side FETs.

Thanks,

Jinal Ubhatar

Hi Jinal, 

Thank you for the schematic, there is a known issue with the digital output pins when using the low-side FET topology. 

This may happen for certain commands as covered in Table 5-8. FET Control Commands table. When using low-side we do not recommend the use of the DSG_PDSG_OFF( 0x0093)  &  CHG_PCHG_OFF( 0x0094) commands as this may cause the issue you are seeing. 

Are you using the SET_CFGUPDATE (0x0090) command at all when the FET commands are being used? 

Best, 

-Luis Torres

Hi Luis,

Thanks for the update.

we are not using SET_CFGUPDATE (0x0090) command at all when the FET commands are being used, Is required to use this command?

Thanks,

Jinal Ubhatar

Hi Luis,

When using the low-side you do not recommend the use of the DSG_PDSG_OFF( 0x0093)  &  CHG_PCHG_OFF( 0x0094) commands, then is there any other way to control individual FETs or both FETs?

Thanks,

Jinal Ubhatar 

Hi Luis,

In our setup, we are using a 16s LFP battery.

We have referred to the application note titled "Using low-side FETs with the BQ769x2 Battery Monitor Family" from Texas (SLUAA84A).

In the document, the circuit mentioned for driving the charge FET is attached as an image.

How should we decide the voltage and current rating for driving the P-MOSFET in Part Q3?

Can you please let us know which part number of MOSFET will be the most suitable for Q3?

Please revert with the technical clarification for the selection of the MOSFET.

Thanks,

Jinal Ubhatar