• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 62 subscribers
  • Views 1994 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.

bq40z60 not responding (no ack) to EV2300/bqStudio SMB commands on custom PCB

Matthew Rubio
Prodigy 30 points

Other Parts Discussed in Thread: BQ40Z60, BQSTUDIO

Hello, people smarter than I!

My bq40z60 is now soldered to a custom PCB. I have followed the Setup Guide in the EVM datasheet (sluub71) and tried everything I could in bqStudio to get it working. My board is modeled very, very closely after the EVM on the communications side. I get the following errors on startup:

- Operation failed because initial read of all data flash failed.

- Command error sending data: No acknowledge from device.

- WriteMACReadBlock communication error:MAC commanf sent does not matchMAC returned in read packet.

[all of which tell me nothing is coming back]

Questions:

- Are there some special considerations for the hardware design of the SMBus lines? (I placed grounds in between each for signal integrity purposes)

- Are there specific commands needed at startup? (I plugged in the load, battery, and charger, and pressed "WAKEUP" but nothing is happening)

- How would I know if I had a dead bq40z60? (I can't conceive of how I might've killed it)

The PCB: For my board, the component sizes were increased for ease of soldering. I have much larger DSG, AC, and CHG FETs as well as mostly 0805 passives instead of all those 0603's (that right?) I saw on the EVM. Could that create issues?

Generally: I'm lost. I appreciate anything y'all can do for me, since my project is due in three days.

  • 0
    TI__Guru 73830 points
    Matthew,
    The external components in the SMBus signal path are placed there for ESD protection. The component sizes should not cause a problem. I would verify that voltage is applied to the ACP and VCC pins to wake up the device. You can monitor the SMBus signals to see whether the gauge acknowledges the address word. The chip hardware will acknowledge the address word and then ass control over to the firmware.
    Tom
  • 0 in reply to ThomasCosby
    Prodigy 30 points
    I confirmed that the VCC (20.5V) and ACP (20.8V) are getting voltage. ACP constantly has this voltage, and I am realizing that my button for WAKEUP does not change this voltage at all when pressed. I am monitoring the clock and data lines one transaction at a time on the SMBus and see no acks happening still. Could the button be the issue?
  • 0 in reply to Matthew Rubio
    TI__Guru 73830 points
    The Wake button is not required. It was added to the EVM as a convenient way to apply voltage to ACP to wake-up the device, if a charger is not present. You have voltage applied to ACP, so the gauge should wake up. The next step would be to determine whether the gauge is awake or in shutdown. Check to make sure that there is voltage on BAT and PBI. You can also monitor the thermistors to see if the gauge is awake. If the gauge is awake and the thermistor is selected in the data flash, then you should be able to see a pulse on the thermistor periodically. If you do not see any pulsing, then you can try measuring the current going out of the VSS pin of the device. You should see at least 100uA current. A convenient place to check this is to insert the current meter in the ground return path the the cells. (1N)
  • 0 in reply to ThomasCosby
    Prodigy 30 points
    I appreciate the help and may come back to this issue next week. I plugged in the battery again while testing my board and blew out both the bq and the 3.3v regulator for the LEDs. I'll be going back to using a simple A/D setup for voltage monitoring. Next week, when my projet is already done, I plan to revisit the issue with another chip/PCB and try the fixes you suggested. Thanks for your help!