• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 419 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.

BQ76PL455EVM:

Nick Hartley
Prodigy 125 points
Part Number: BQ76PL455EVM

I'm using the BQ76.....EVM with associated software running on a Windows 7 laptop to measure voltages on 3 x 16 Lithium cell LGchem batteries. The EVM hardware (board address 0) works fine on the first battery. It also recognises the BQ boards on batteries 2 & 3 (addresses 1 & 2), but when I try to access data from them, I get  "Board communicating at address 1 (or 2) can no longer be found. Check communicating connections (I have). Repower all boards. Close & restart this tool".   The BQ76.. boards on batteries 2 & 3 are also from LGchem.  The BQ76..... manual seems to imply that you can only use additional BQ76....EVM boards in a stack.  It seems that the LG boards are waking up sufficiently to be identified by the EVM, but are then going to sleep again before data can be obtained.  The BQ data sheet says that a sleeping board, if 'above' the base board (address 0) has to be woken up by a 1ms x 100 kHz burst.  I don't see this burst at any time on the differential comms 'daisy chain' !   Can anyone throw any light on this problem?

  • 0
    TI__Genius 11480 points
    Hi Nick,

    If I am understanding correctly, you are using one EVM to communicate to stacked devices built by LGChem?

    How long is the cabling you are using for the daisy chain, and is it a high capacitance cable? There is a limit to how much capacitance the Pl455 can drive (~140pF) while maintaining good communication integrity.

    Are you also able to probe/ see the circuitry used for the LG chem board daisy chain circuitry? If it does not match that of the EVM, you may have some issues cropping up from the source/load ends of the cable not being the same.
  • 0 in reply to David Wiest Jr
    Prodigy 125 points
    Hi David, I'm using a 4 wire ribbon cable between EVM and BQ board. Length 45 cm, conductor spacing 1.3mm. With a Fluke Scopemeter 123 I measure approx 10 pf between adjacent conductors. (i.e. reading changes from 0.01 to 0.02 nF). Between batteries the same, only length 12 cm. I will check the LG chem diff comms circuit and get back to you. Thanks for your quick response. Nick H
  • 0 in reply to Nick Hartley
    Prodigy 125 points
    Hi again, Yes, I'm using a single EVM with 16-cell battery inputs. It is connected by daisy chain to a stack of 2 more 16-cell batteries each with LGchem BQ76PL455 boards which are multilayer SMT board with ground plane top & bottom. The differential comms circuit has 2.2 nF coupling capacitors on each line followed by common mode chokes (can't measure these), then series 10 ohm resistors, zener diodes to ground across the appropriate BQ chip COMMS+/- & FAULT+/- pins. All as recommended in the data sheet it would seem.
    Regards, Nick H
  • 0 in reply to Nick Hartley
    Prodigy 125 points
    I found out that the LGchem boards were set up to shut down if no communications were received within 1 sec of wakeup. Hence by the time I was polling for cell voltages, the board had switched off. Two simple commands overcame this problem: F1 0C 20 55 2B to reset followed immediately by F1 28 00 4F F3 to disable the comms break shut down. Now able to read the upper BQ boards. Regards, Nick Hartley