• State TI Thinks Resolved
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 65 subscribers
  • Views 224 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ79616-Q1: Board locking up and requiring power cycle to recover

Andrew Fillingim
Prodigy 30 points
Part Number: BQ79616-Q1
Other Parts Discussed in Thread: BQ79616

Tool/software:

Hi,

We are dealing with an issue where the BQ79616 chip is locking up. In this state it is drawing a large amount of current ~60 mA and will not respond to a HW_Reset COMM tone. Below are some measurements we have taken and tests we have tried to recover the board. From what we can tell, CVDD has a high current draw on it and this is preventing communications, but we cannot figure out what is causing the current draw.

  1. Could you help provide some insight into what state the BQ79616 may be in?
  2. Is there anything internal to the chip that may be drawing this amount of current?
  3. Are there any measurements or tests that would be useful to do?
  4. Any chance we can setup a call to discuss this issue?

PCB Power Supply Measurements

All measurements using BAT_A0 as ground.

Temperature of Q301 and C302 in power regulation circuit: 71 degC

 

Name

Measured Voltage [V]

BAT_A16

66.5

Q301-2 Input

48.3

 Between R317-R318

54.3

NPN_Base Q301-1

4.25

LDOIN

3.846

 

BQ79616 Measurements while in lock up state

All measurements using BAT_A0 as ground.

Name

Pin

Measured Voltage

LDOIN

47

3.846

TX

52

3.063

TSREF

51

0.3008

RX

53

3.056

SRN

63

0.0169

SRP

64

0.0234

BAT

1

66.5

NPN_Base

48

4.25

AVDD

38

0.006

NEG5V

44

0.022

REFHP

37

0.006

CVDD

45

3.064

 

COMML+ TVS Diode

1.502

 

COMML- TVS Diode

1.502

 

COMMH- TVS Diode

1.498

 

COMM H+ TVS Diode

1.498

 

Tests Done to try to recover BQ79616 chip that is in the locked up state

  1. HW_RESET
    1. We set CONTROL1[HW_RESET] = 1 of the previous chip
    2. We confirmed on a scope that HW_RESET was sent over COMM
    3. The board did not recover
  2. Unplugging GPIO
    1. We are using 8 thermistors
    2. We disconnected all thermistors
    3. The board did not recover
  3. Unplugging COMM
    1. We removed the COMM line connectors
    2. The board did not recover
  4. Recovery while taking measurements
    1. On rare occasions, measuring the voltage between the ground side of the power supply capacitor and BAT_A0 (Ground) with a multimeter caused the board to recover
    2. On rare occasions, measuring the voltage between LDOIN and CVDD caused the board to recover with a multimeter.
    3. On rare occasions, a 1 MOhm resistor between LDOIN and CVDD caused the board to recover.
  5. Removing BAT_A0-BAT_A16 connector
    1. This always resolved the issue, but is a nonviable solution for production.

Scope Traces while in lock up state

HW RESET COMM Tone to chip

LDOIN while in lock up state

Thanks,
Andrew

  • 0
    TI__Expert 7000 points

    Hello Andrew,

    Are there any events that are happening that are causing this issue to occur?
    Seeing as CVDD has a high current draw were there any events on CVDD or the COM pins?

    How reproducible is this?

    Best,

         Quentin

  • 0 in reply to Quentin Silic
    Prodigy 30 points

    Hey Quinton,

    This event happens fairly randomly. We cannot tell what the source it because we lose comms when this happens, so we cannot read the fault registers.

    We have been able to reproduce this sometimes by opening the HV contactor/shutting down while AC charging, but this event also has occurred on battery modules that have not been installed in a vehicle as well.

    Do you have any more detailed info on what may cause a high draw on CVDD?

    Thanks,

    Andrew

  • 0 in reply to Andrew Fillingim
    TI__Expert 7000 points

    Hello Andrew,

    So you can sometimes reproduce when closing the HV contactors during AC charging?
    Do you have ring architecture? If yes have you tried a HW_RST in the reverse direction?
    What are the values of your TVS diodes?
    Can you please capture the voltage on the references and COM pins during HV contactor closing?
    Does this happen on the same IC each time or is it always different?

    Best,

         Quentin

  • 0 in reply to Quentin Silic
    Prodigy 30 points

    Answers are below.

    1. Could you provide some more detail on what could draw this amount of current from CVDD?
    2. What reference pins are you wanting traces of?
    3. Where on the COMM circuit would you like a trace? At the pin itself?

    Answers

    1. Only can sometimes reproduce when "opening" contactors during AC charging, never closing. Again, this event has also occurred when the battery is not connected to a vehicle.
    2. We do not have a ring architecture setup. We do know HW_Reset does not reset the device in the set direction and have confirmed the tone does get sent to the device.
    3. We are using these TVS diodes:
    4. Yes we can capture traces and will send them we when get them.
    5. This event often reoccurs on the same IC, but we were able to get an IC further up the stack to have this event first and even get a different stack all together to lock up first.
  • 0 in reply to Andrew Fillingim
    TI__Expert 7000 points

    Andrew,


    1. With the information we have now we can't say for certain what is causing CVDD to have a high current draw. It could be that something on the COM pins is getting damaged and a transistor is put into the wrong state or slightly damaged which is causing the current consumption. 

    2. LDOIN, CVDD, DVDD, AVDD and COM pins especially.

    3. Yes I would like it at the pin side to the IC GND. 

    Best,

         Quentin 

  • 0 in reply to Quentin Silic
    TI__Expert 7000 points

    Andrew,

    We can continue this conversation via email as this will be a little faster to share information. 
    Please email me at q-silic@ti.com. 

    Best,

         Quentin