• State TI Thinks Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 5 answers
  • Subscribers 63 subscribers
  • Views 1257 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.

Original question:

BQ76940: BQ76940

BQ76940: BQ76940

user5247366
Prodigy 250 points
Part Number: BQ76940
Other Parts Discussed in Thread: BQ76930, TIDA-01093

Dear WM:

     I have some questions about BQ76940.

     The power of REGSRC pin is form external power,which  supply 13V, but REGOUT pin will output 4.41V, normal is 3.3v ;CAP1 will output 3.88v, normal is 3.3v; Unconnect load and charger，ALERT will output 4.45v, the voltage of the others pins  are normal ,why?

     Thanks for your replay!

  • 0
    TI__Guru 54715 points
    Hi User,
    Check your reference for the measurement to be sure you are referencing VSS of the IC. Be sure the VSS pin is soldered to the board ground and referenced to the battery negative. A normal part will not output high voltages, but supplies are from REGSRC or VC5X which have higher voltages available if damaged. Forcing a voltage into the output pins will raise the voltage and can cause damage.
    The voltages you note exceed the abs max voltage for the pins. If the VSS is not disconnected the part must have been damaged, replace the part but check for abnormal voltages and correct any problems before installing the replacement.
  • 0 in reply to WM5295
    Prodigy 250 points

    Dear WM:

          Thank  you, the problem  is solved.

  • 0
    Prodigy 250 points
    Dear WM:
    I have some questions about BQ76940.
    ① Adjacent cells should not be balanced simultaneously. This may cause cell pins to exceed their absolute maximum conditions and is also not recommended for external balancing implementations.（The Page 29 of BQ76940）
    This may cause cell pins to exceed their absolute maximum conditions, why?
    ② The timelines of the cell groups in the bq76930 and bq76940 are independent, so one group may be balancing while the next group is measuring. At the cell boundary between groups, the adjacent cell may measure a voltage from the balance current in any common path.(why?) Keeping the common path resistance low, using wide traces or returning the current as close to the cell as possible will reduce this effect (see Figure 8). (The Page of TOP10)
    ③ Since the cell inputs pull together during balancing, care must be used to prevent damage to the inputs. If all cells in a group are enabled to balance at the same time, the inputs will pull together and some of the maximum limits will be exceeded, the part is likely to be damaged. Balancing every other cell in a group will double the voltage on the unbalanced cell and is typically OK, check the voltage limits for the cell voltages used in your design.（TOP10第9页）why?
    ④If 76940 connet 13 cells,VC8 and VC9 short,VC13 and VC14 short, Can cell8 (connet to VC8) and cell9 (connet to VC10) they be balanced at the same time?
    ⑤ Test SCD, shunt resistance equal to 1mohm, set SCD value equal to 22mv, calculate current equal to 22A,but test 18A DSG_ON=1, there is 4A error, why? Relate to PCB layout? Straight line distance of SRP and SRN network is 105mm.The error of measuring small current(1.18A) is small.
    Thanks for your reply!
  • 0 in reply to user5247366
    TI__Guru 54715 points
    Hi User,
    1. If cells are at 4 V and cells 1 and 2 are balanced, VC0, VC1, and VC2 will pull to the same voltage, approximately 4V. This will exceed the 3.6V abs max for VC0. The part was not designed for adjacent cell balancing.
    2. The part was not designed with synchronized timelines. See the figure noted and consider IxR drops in the resistance.
    3. The cell voltage of an input moves about 1/2 the cell voltage when the cell balances. See www.ti.com/.../slua775 figure 5. If cells are at 4V and cells 1 and 3 balance, VC1 moves down 2V and VC2 moves up 2V. The cell 2 inputs (VC2 - VC1) experience 8V.
    4. It is not recommended since cells 8 and 9 are now adjacent since VC8 and VC9 are connected together.
    5. Earlier versions of the data sheet showed an incorrect value for the Protect3 register reset value reserved bits. Some users incorrectly set a reserved bit to 1 rather than leaving them unchanged. This shifted the SCD protection threshold similar to what you describe. Please use the latest data sheet, be sure the Protect3 reserved bits are set to 0.
  • 0 in reply to WM5295
    Prodigy 250 points
    Dear WM:
    Question 5 above , Protect3 reserved bits have been set to 0,This shouldn't be the reason. I don't know what caused it.
  • 0 in reply to user5247366
    TI__Guru 54715 points
    Hi User,
    #5. It could be PCB or assembly. The solder resistance will show up in any measurement unless you have the 4 terminal resistors.
    The board layout can affect the measurement, you generally want the traces to the sense filter to be directly to the pads rather than to the larger current trace.
    Sometimes you can find these by probing on the resistor terminals, on the pads vs the resistor terminal, or on the sense input filter resistor to other points.
    Also note in the data sheet the OCSCALEERR of 10%. While error is typically small, combined with another effect it could become large. Try another board if available.
    Hope these comments help you find the cause.
  • 0 in reply to WM5295
    Prodigy 250 points
    Dear WM:
    The document of BQ76940, the programmable range of OV thresholds is approximately 3.15 to 4.7 V, but this is subject to variation due to
    the (GAIN, OFFSET) linear equation used to map the ADC values.The programmable range of OV thresholds is approximately 3.15 to 4.7 V,under 3.15v(for example 2.5v) ,is it OK?
  • 0 in reply to user5247366
    TI__Guru 54715 points
    Hi User,
    The OV threshold register can not be set below 00. This gives the minimum OV threshold value of 0x2008, see the bq76940 data sheet section 3.7.1.2.1. With the gain and offset stored in the part a voltage can be calculated, but with the typical LSB value of 382 uV shown in the data sheet and 0 offset this would calculate to 3.13V. There is no way to set the OV to 2.5V.
  • 0 in reply to WM5295
    Prodigy 250 points
    Dear WM:
    If 76940 connet 13 cells,VC8 and VC9 short,VC13 and VC14 short, the voltage difference is 0 .Test UV, BQ76940 automatically do not detect the voltage between them ?
  • 0 in reply to user5247366
    Prodigy 250 points
    Dear WM:
    An additional question,If there is 20 cells, I would like to use two BQ76930 ,Do you have any recommended documents or schematic design?
    Thanks!
  • 0 in reply to user5247366
    Prodigy 250 points
    Dear WM:
    If acquisition pins of BQ76940 connect to battery, BAT、VC10X、VC5X have power, but REGSRC pin has no power input , Which modules of BQ76940 are working ?
    Thanks!
  • 0 in reply to user5247366
    TI__Guru 54715 points
    Hi User,
    Shorted cells: Yes, if the voltage is below UVMINQUAL the part will not declare a UV. It expects the user has intentionally shorted the inputs. The host will need to realize that a cell which is supposed to be present is missing.
    For 20cells, see TIDA-01093 using 2 bq76930s.