• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 1815 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 after booting (bootloop)

Bernhard Suppan
Prodigy 230 points
Part Number: BQ78350-R1
Other Parts Discussed in Thread: BQ76930EVM, BQSTUDIO, BQ76930

I am trying to develop a battery pack with the BQ78350-R1 as companion and the BQ 76930 as AFE.
To test the PCB I connected the companion via SMB with the EV2300 and tried some of the commands in BQ Studio (LED_EN, LED_toggle,..). After some testing I tried the "RESET" command. And it did restart, but it keeps doing it ever since.
When I boot the AFE via the TS1 pin, it applies 2.5V to the REGOUT pin and the companion starts. After 10s the voltage at REGOUT pin drops to 0V and I have to boot the AFE again...
The AFE and the companion are conntected via the I2C bus.
It keeps rebooting even after I disconnect every power source from the PCB and reconnect it.
I assume this command must some how changed something in the EEPROM.

How can I resolve this problem?

  • 0
    TI__Guru 54715 points
    Hi Bernhard,
    In bqStudio the non-volatile parameter memory is called Data Memory. See the bq76930EVM user guide www.ti.com/lit/slvu924 section 6.3, last paragraph. Applying a small charge current or simulating one will keep the gauge from shutting down so you can find and fix the configuration.
  • 0 in reply to WM5295
    Prodigy 230 points
    My PCB is designed to work with 6 cells. Therfore I shortened the pins VC9, VC8, VC7 and VC4, VC3, VC2. To test my PCB, I made a second PCB, which should simulate the Battery pack. It consist of six 330R Resistors, which are connected in series from V_Bat to GND. Each VCx pin of the IC is connected to one of the resistors .
    During the test, voltage is applied to the resistors, therefore the AFE should detect the the Cells and shouldn't shutdown.

    How can I resolve this problem?
  • 0 in reply to Bernhard Suppan
    TI__Guru 54715 points
    Hi Bernhard,
    You will need to look at the sequence to determine the cause of the shutdown. If you have insufficient voltage the AFE may pull into VSHUT and shut down. More normally the situation is the gauge detects some low voltage and shuts down. The current simulation technique mentioned above will allow you to keep the gauge from going to shutdown to see what is wrong.
    If it shuts down anyway, it must be some supply problem. Your setup is very much like an EVM and should work fine. Check that all circuits were implemented and that the power is stable at VC5X and BAT pins. You may need to look at these with an oscilloscope. Since you have resistors instead of a low impedance source, you will see some movement in VC5X, but it should not (must not) approach VSHUT.
    With a simulated charge current, look at your cell voltages in the bqStudio display (or your own interface if using it) All 6 cells must show a voltage above the shutdown voltage parameter or the gauge will turn off the AFE. A common confusion is the AFE cell map. The bq78350-R1 shows cell voltage in logical order, it is the AFE Cell Map which defines which physical cells to use in that logical order. For 6 cells as you have described I expect it should be 0x273
  • 0 in reply to WM5295
    Prodigy 230 points

    Hi,

    after applying some charge current it doesn't shut down anymore. Thank you.
    But I have another Problem. I can't change values in the data memory with bq studio. By clicking " Write to data memory", it says "A read of data written failed comparison". I am using the EV2300. I cant change values and i am not able to write the default values to the IC. All i can do is read the values.
    Can you help me with that too?

  • 0 in reply to Bernhard Suppan
    TI__Guru 54715 points
    Hi Bernhard,
    I suspect either the value you are entering is out of range which is more likely to give an error like Value is beyond maximum value defined for parameter, or the supply voltage is too low. If you go to the Power tab in data memory, the voltage shown for the battery in the dashboard at the left must be above the Valid Update Voltage parameter. Adjust the voltage up and you should be able to write to data memory.
    Default is 3500 mV. If the value is changed too large you may be locked out of changing the part. See the note in section 18.1.1 of the technical reference manual, It is not recommended to change this value.
    If a change to Valid Update Voltage has locked out further changes, you may be able to re-load the default firmware, it is different.
  • 0 in reply to WM5295
    Prodigy 230 points

    I think your are right. The value of the cell voltage is 400mV and the reigster value is 3500mV as you mentioned.
    But I coulnd't check if this resolves my issue because of another issue:
    I connected a array of resistor to simulate a battery pack (six 330R in series) and applied a charge voltage (24V).
    I assume the cell Volatge in the bqStudio is measured via the AFE. In bq STudio the average cell voltage is 400mV but measred with a multimeter it is 4V
    What i also noticed is that the temperature is wrong (317°, only connected TS2, TS1 has a 10k resistor, TS3 not available because i am using a BQ76930 as AFE) and also the charge current (527mA) I assume it should be (24V /(6 * 330R))

    The voltage at the VCX pins of the AFE are the correct volatges. I can control the charge FET via the bqstudio tool. And the Cell voltages don't change, when i change the Power source voltage.

    I am trying to find the cause, but if you have some more hints for me, that woudl be great.

  • 0 in reply to Bernhard Suppan
    TI__Guru 54715 points
    Hi Bernhard,
    This may be a terminology confusion, and you said you are developing you own board, but look at figure 2 of the bq76930EVM user guide www.ti.com/lit/slvu925. The resistors would be on the battery side (they are included on the EVM if selected), the left side of the board in the figure and the power supply must be across the resistors to simulate the cell voltages and properly power the bq76930. If the power supply is the charger voltage connected to the pack terminals of the system where the electronic load is shown in figure 2, the resistors may not have power. The register values you show above seem to indicate either the voltage on the resistors is very small or it is displaying an offset value. It is not clear how the part would get power if the voltages were so low. The voltage in the left column is the battery voltage or the sum of the cells. The voltage of 407 seems consistent with 9 cells at about 45mV. While it is a different product, lower cell count, and simpler function, you want your setup to look like figure 1 of www.ti.com/lit/slvuan2.
    If you do add a power supply and want to use a supply as a charger also, be sure your supplies can tolerate being connected together, some supplies which I use are damaged by reverse current and must be used with blocking diodes when used to simulate cells and a charger in the same setup.
    Another concern with the registers above is your 6 cells: 24 V and 6 resistors is consistent with this. However the register display above shows 9 cells with values. Please check the AFE Cell Map and be sure only the appropriate cells are selected. As mentioned with the update voltage before you will need to get power to the part to recognize the threshold before it will allow you to change the data memory.
    The Ext Avg Cell Voltage in column 3 (shows as 0V) is measured at the BAT pin of the bq78350-R1, scaled and divided by the number of cells. This only works if the VEN is enabled in the configuration. Above it shows 0.
    The temperature is 0 by default and displays as -273.2 degrees as shown in the figure above. I'm not sure where it would get 310.7 for TS3 which is not supported in the bq76930, or where it would have found the 317 deg. It would seem the part is not powered correctly and you may have bad data. The resistor on TS1 should always give some nominal value and the TS2 should represent the temperature. Be sure to configure the temperature properly to select TS2.
    The current shown in column 1 is not the current through the simulated cell but the current in the sense resistor. Once you get the board to come on and display voltages properly, you should be able to set configuration to turn on the FETs and provide a current such as figure 17 of the bq76930EVM user guide. A few mV on SRP to simulate charge current may be needed to keep the gauge awake to correct configuration (mentioned Feb 1 above) before you are able to turn on FETs.

    Hopefully these notes and the user guide figures give guidance to get proper power on the board, the part should then read adequate voltage so you can fix the configuration bits and get normal operation.
  • 0 in reply to WM5295
    Prodigy 230 points

    Thank you for your help, but I solved it myself.
    I programmed the companion with firmware again and it worked.
    I think I got myself locked out of the companion because I configured the Cell map wrong and the companion therfore measured only 400mV (46mV if the Cell is not connected)
    But how did I configure the Cell Map wrong? In the technical Reference it says 1 = Cell connected, 0 = Cell not connected and in bqStudio I assumed green means 1 and red 0. But it is the opposite. After I configuered it with red = 1 and green = 0 it worked.