BQ76930 for parallel cells

So you mean using two bq76920? actually, the idea was to use one bq76930 to save the cost and effort for an extra device and the level shifter for i2c. the schematics so far are mainly in my head, not yet started drawing them...

so do you think the bq76930 will not work when the input VC5b is connected to the same voltage level as VC0? from the datasheet, I could not find anything that opposes this.

Thanks,

Michael

In the bq76930, the VC5b cannot be used as GND. It can only be used as a power pin.

A team member,  who was involved to  define this device, will elaborate of the  options available to you using the bq76930.

VC5X on the bq76930 is a power pin between the upper and lower cell groups.  Note that in figure 8-2 it connects to the cell point "A" between the 2 groups.  Note also the requirement for VBAT in the recommended operating conditions.  When you have cells in series the part will operate OK.  When you want to parallel the cells, VC5X and/or point A can not be both at the top of one cell group and at the bottom of both.  This will short one stack of cells and even if there is no damage the bq76930 will not work.

VC5B should be OK as long as it follows the same reference as VC5X

When using 2 bq76920 you can have VSS of one part referenced to either the top of the group or VSS of the other, the challenge then is to have isolation in the interface signals so that they can sometimes operate at 0V and sometimes at ~20V.  Another challenge will be dealing with the differential currents each time the cells are connected in parallel since the voltages will never exactly match until connected.

Thank you both for your informative reply. I am not sure I understood you right:

so you say that VC5X can not be at a higher voltage then VC5b. This means, that in case of parallel batterys, VC5X must be at GND Voltage, because VC5b is tied to GND. But VC5x is the supply pin for the lower block, and it must be at a minimum of 6V for the device to work properly. This means, the bq76930 will not work in this design.

two times the bq76920 would work, if I can translate the I2C bus to the high-level IC (probably I will find some cheap capacitive way) and if I can handle the differential currents (I will be using a PTC anyways).

So I will probably go for the 2x bq76920.

Just one more simple question on the bq76920: In the figure on the first page of the datasheet, Vss  is connected at SRP voltage Level. I would like to use the ground level behind the shunt resistor, at SRN level, which may differ by a few mV. Will the current measurements still work in both directions and will the voltage measurement for the lowest cell still work?

The part will operate with VSS on the pack side of the sense resistor, but from the specifications it is not recommended.  Discharge currents are typically higher than charge currents, and may include short circuit currents.  With VSS at the battery side of the sense resistor, the sense pins of the IC will see positive voltages, and they have a larger positive range to abs max (3.6V) than negative range (-0.3V)  The thresholds are within the abs max range of course, but the voltage on the sense resistor will depend on the current capability of the cells, the system interconnect and the attached system/short.  The filter will remove fast peaks from the sense pins but you may need to clamp the voltage to a safe limit.

aah, one more question:

is it possible to use BQ76930 with only 4 cells connected? I know it does not make much sense, but we might need to do it because of different assembling options on one PCB.

Another way to look at the MINIMUM cell requirement, 3 cells/group; As such, the bq76920 will mandate a  min  of 3 cell connection. the bq76930 will mandate 6 cells. the bq76940 will need 9 cells. The cell connections are detailed in pp42  of the d/s.