• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 4 answers
  • Subscribers 63 subscribers
  • Views 1053 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.

BQ78350-R1: Shutdown not performed after ManufacturerAccess command executed

Emanuel Wahlqvist
Prodigy 120 points
Part Number: BQ78350-R1
Other Parts Discussed in Thread: BQ76940, BQ78350

Hi,

We have an issue with shutdown that seems to be related to temperature. We have an MCU that uses the SMBus to read data values from the BMS and also controls when to shut down. The procedure we use to shut down is:

- MCU sends MAC Shutdown

- MCU reads the operation status register until the SDM bit is set, then it seizes all communication (so as not to wake the BMS again)

- BMS is supposed to shut down, hence disable the 2.5V LDO from the AFE (bq76940)

However, when we do our climatic tests the BMS does not shut down after setting the SDM bit, instead i clears the bit and continues to operate. When this occurs, the UTD protection has been activated (we have the threshold set to -20°C). 

This procedure work well when in room temperature, even with simulated temperature sensors at -20°C and UTD present.

I cannot find any reference to why the BMS does not shut down, also I read in sluubd3c section 17.2.10:

"Once SHUTDOWN mode is enabled, it is not possible to clear it without entering SHUTDOWN mode".

Which does not seem to be correct, the SHUTDOWN mode is never entered (what I can detect atleast).

Can you please shed some light on this?

Best regards

Emanuel W

  • 0
    TI__Guru 54715 points
    Hi Emanuel,
    Nothing is suspected in the parts, both should work to -40C, the lower recommended operating temperature condition. The 2 expected explanations of the SDM bit clearing would be a reset of the gauge, or the system shutting down and re-starting. Shutdown should show up in the lifetime data shutdown counter if you have it enabled.
    Current can prevent gauge shutdown, but as the section 17.2.10 indicates, it is not expected to clear the condition.
    Tests below 0C can be difficult due to condensation on the circuit boards. You might monitor the REGOUT and TS1 signals to see if something is causing a reset or re-start after the shutdown in your test.
  • 0 in reply to WM5295
    Prodigy 120 points

    We have now measured the REGOUT and TS1 signals at room temperature and at -0, with a potentiometer simulating a -20 value of the NTC.  I do not se anything suspicious in these signals but maybe I am missing something?

    In the pictures the blue signals is TS1 and the magenta is REGOUT.

    First picture: -0 degrees and shutdown does not occur

    Second picture: Room temperature, shutdown occurs and REGOUT falls 

  • 0 in reply to Emanuel Wahlqvist
    TI__Guru 54715 points

    Hi Emanuel,

    It is suspicious that there is a pulse on TS1 when REGOUT falls.  There should not be a temperature sample when the part turns off.   With the long default shutdown delays it can be hard to get a frame of reference for when shutdown should be occurring.  

    Look at your boot circuit to see if it is triggering.  It may be almost triggering at room temp and triggering at cold. You may need to hold off your boot circuit at shutdown.

    In this example the FET off time and ship Delay were reduced so the waveforms fit on the screen.  It is at room temp.

  • 0 in reply to WM5295
    Prodigy 120 points

    Hi,

    It indeed seems like our startup circuit is making something funny at that point. See attached images where a shutdown at 0 degrees succeeds and a shutdown at -20 degrees fails. We will measure and probably perform some modifications to the circuit and come back with the results.

    //Emanuel W

    Freeze test.zip

  • 0 in reply to Emanuel Wahlqvist
    Prodigy 120 points

    Hi,

    We have now revised our startup circuit and the problem seems solved. We have a circuit that makes the SMBus clock generate a pulse if the EV2300 is attached, and the SMBus lines where pulled up to a 3V3 feed that falls much slower than the 2.5V REGOUT. Hence, when the BMS correctly shuts off REGOUT, the 3V3 feed made the startup circuit generate a pulse via SMBus clock.

    Pulling the SMBus lines to REGOUT insteda of 3V3 solved the issue.

    I do have another related question:

    In a separate project also using the BQ78350/BQ76940 they have had problems starting the boards. In order to get the AFE to start properly they needed to mount three 220K resistors, one each in parallel with the TS1, 2 and 3 inputs. Without these, some boards would fail to start completely, on others the AFE would just partially start (for instance, bank 1 and 2 would start, bank 3 would not). This would be detected by the cell-measurements from the non-starting block not being done, just reporting 0. Have you seen any similar behavior somewhere else?

    Best regards

    //Emanuel W

  • 0 in reply to Emanuel Wahlqvist
    TI__Guru 54715 points
    Hi Emanuel,
    Glad to hear you found a solution.
    The TSn pins are inputs and should have a pull down resistor if the thermistor is not connected or not used. These should be supported on a test fixture if the board is separated from the thermistors for test. Residual charge on the TSn pins can prevent a group/bank from starting. If the group does not start the device will have an XREADY fault which will not clear and different cells will read 0 depending on which group did not start. The data sheet does not show floating TSn pins, the device FAQ does mention the pull downs, and there are likely several e2e posts describing the behavior, but it is not obvious.
  • 0 in reply to WM5295
    Prodigy 120 points

    Hi,

    Ok, so for any TS pin that does not have an NTC a pull-down is required. Good to know, thanks! From what I understand from my colleagues though, there where 10K NTCs connected on all three TS pins and still the device would not always start. Do you have any idea why?

    Best regards

    Emanuel W

  • 0 in reply to Emanuel Wahlqvist
    TI__Guru 54715 points
    Hi Emanuel,
    I don't know why it would not start. The know causes for not starting would be:
    Missing power to upper group
    Shorted CAPn capacitor
    TSn pull down missing allowing residual voltage or voltage held on TSn pin
    TSn pulled to VSS rather than the local reference
    If something interferes with internal communication XREADY can set but if the part subsequently starts the voltages will be present and the status can be cleared. A damaged part might have XREADY but one would expect it to be permanent.
  • 0 in reply to WM5295
    Prodigy 120 points

    Hi,

    Ok but thank you for the information. We will keep these resistors in place and adjust the temperature coefficients to compensate for it. 

    As the original issue has been addressed I will now mark this issue as solved.

    Best regards

    //Emanuel W