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.

BQ34Z100-G1: no communication with bqStudio

Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: BQSTUDIO, , EV2400, BQ34Z100

Hi All,

I'm having trouble communicating with a bq34z100-G1 battery gauge using bqStudio and the EV2400 : the dashboard shows the device in grey and e.g. reading the FW VERSION register gives "SBM Write Word: No acknowledge from device".

The bq34z100-G1 sits on a proprietary board and monitors a 4-cell Li-Ion battery. (I planned to attach the relevant part of the schematic but I can't find how to attach a file....).
I did run a number of checks, all of which seem OK to me:

  • Supply voltage for the bq34z100-G1 (on REGIN, pin 6) is 4.5V
  • REG25 output is at 2.5V
  • VEN pulses once per second during 120ms

=> it looks like the bq34z100-G1 is "alive".

On the SM bus I can see the SCL and SDA signals when BQ Studio polls the device. Initially, I had two LEDs connected to the bus but as you can see in the schematic, there is now nothing more on the bus than 10K pull-ups (to the 4.5V 'REGIN' supply), 100R serial resistors and the EV2400. Looking with my scope, I see that the signal looks pretty good. when I capture a complete start-data-ack-stop sequence there seems not to be an ACK indeed (I'm sure however). The signals are the same at both sides of the 100R serial resistors (R58 & R59 in the schematics).

So the question is, does anybody has any suggestions of what could be wrong ?

bqStudio version: 1.3.54.1 (latest version I could find, downloaded last week)
PC Interface: EV2400 (brand new, first time use)

Thanks for your help !

Jeroen

  • Hi Jeroen,

    To share the schematic and other files, please click on the 'Insert Code, Attach Files and more...' link below this post. This will bring you to an advanced WYSIWYG editor. Click on the attach media button to attach the files to the post. It may be simpler to zip everything together and upload as one archive.

    Please update bqStudio to the latest version at www.ti.com/.../bqstudio by clicking on the 'bqstudiotest' button. As of this writing, it is version 1.3.80.

    What version is the firmware on the EV2400? The current released version is V0.18.

    For a quick guide and links on how to update the firmware of the EV2400 as well as bring up the bq34z100-G1 device on the EVM (should be similiar to your board), please refer to the Quickstart Guide found here: www.ti.com/.../sluubv2

    If the issue persists after updating both bqStudio and the firmware of the EV2400, please submit those files for further troubleshooting. Thanks!

    Sincerely,
    Bryan Kahler
  • Hi Bryan,

    The firmware version of my EV2400 is 0.18, so that seems OK.
    I also just downloaded bqStudio 1.3.80 (I saw it before but I did no choose it since it is mentioned "for Testing Purposes Only").
    Anyway, I still do not have any contact with the bq34Z100-G1.

    Attached is a ZIP fie with the schematics and two screenshots from my scope.

    Thanks for having a look at that !

    Jeroenbq34Z100 Schematic and scope.zip

  • Hi Jeroen,

    Thank you very much for the detailed information and multiple debug steps taken.

    Is the EV2400 connected via a USB hub? If so, please connect the EV2400 directly to a USB port on your computer.

    If this is not the case/does not work, please uninstall the EV2400 driver that windows installed when the device was first plugged in, followed by rebooting the computer and plugging in the EV2400 after a fresh boot.

    If the issue still yet persists, please send a logic analyzer output of the I2C communication that takes place.

    Sincerely,
    Bryan Kahler
  • Hi Bryan,

    Thank for the additional suggestions.

    I <think> I found the device in the device manager that corresponds to the EV2400 although it appeared as a "user interface peripheral".  I uninstalled the driver, unplugged the EV2400 and restarted my PC. After plugging in the EV2400 it immediately reappeared as the same device.

    Attached you find two screenshots from my logic analyzer, one is an overview in which you can see that there is first a group of three words  with 17 ms in between them, followed by another word after 116ms and finally 16ms later just a isolated 'stop' condition.
    The 5 words are all the same as shown in the detailed screen shot: start - 0x0B - N - stop. So there is no ACK from the slave/bq34z100.

    Two additional observations:

    1) the bq34z100 is supplied with 4.5V, as are the pull-ups on the SMBus. As soon as I connect the EVM2400, the voltage on the SMbus drops from 4.5V to 3.6V. So there may be a problem with the bq34z100 not seeing the 3.6V as a '1' ? (although not very likely...)

    2) In my first post, I saw a 120ms pulse at a 1Hz frequency on output VEN. Today there is no activity on this pin. The only change that occurred since that first post is the updating of bqStudio. I did not touch the hardware (except from unplugging the battery...).

    Hopefully this shines some more light on the situation....

    Jeroen

  • Hi Jeroen,

    Thank you for reinstalling the driver.

    You're spot on. If VEN is not pulsing, the device is most likely in 'single cell mode' and using the internal divider. However, it may also be in ROM mode.

    Please scope the TS pin to see if there is a pulse there. If no pulse is detected, the device is in ROM mode. If in ROM mode please either reprogram the device the execute command 0x08 to execute the firmware to bring the device into 'firmware' mode.

    Please let me know if the pulse is not seen on the TS pin after performing these two options above.

    Sincerely,
    Bryan Kahler
  • Hi Bryan,

    After hooking up my scope to both the VEN and TS pins this morning, I discovered that there actually IS a pulse on VEN. But not every second as before but only once every 20 seconds (about 125ms wide).
    On the TS pin I also see a 140ms pulse every 20s, about 125ms after the VEN pulse ends (Note that I had to add a 10K resistor between TS en REG25 since I do not have a thermistor in my schematic!).

    So the device seems to be alive. You talked about reprogramming or execute the 0x80 program but remember that the initial reason why I posted the forum is that I have problems with the communication between the device and bqStudio. This is still the case, the symbols at the left side of the screen are still grey.

    Thanks,

    Jeroen

  • Hi Jeroen,

    It's alive!

    Which bqz file are you using when the device is not detected? Please ensure it matches the firmware that is programmed on the device.

    Please completely remove bqStudio and reinstall fresh with the test version and then try running again.

    Try cycling the power on the device and removing the EV2400 and plugging it back into the computer.

    If the issue still persists after the above steps, please send a screenshot of bqStudio, including the left side of the screen and include any errors you've seen in the bottom left hand corner when trying to communicate with the device.

    Sincerely,
    Bryan Kahler
  • Hi Bryan,

    Yes, its alive ! But not very communicative :-(

    The bqz file is 0100_0_16-bq34z100G1.bqz as listed on top of the targets to select when I start bqStudio. I don't know whether this matches the firmware that is programmed on the device. I never made it to communication with the device so it must be what ships with it.
    Until now I used version 1.3.80.1 of bqStudio. As requested I have uninstalled it and installed the latest testversion: 1.3.86.1 but I still get the usual warning "The detected device is not compatible with this application! ...." and if clicking OK the left side of the dashboard becomes grey.

    Cycling the device and removing/plugging the EV2400 did not help either.

    Attached/below is the screenshot after clicking on the FW_VERSION button. As you can see, the error on the status line says "SBM Write Word :No acknowledge from device".

    I have a space device so I could eventuelly replace the current device on the PCB. But since it seems "alive" I'm not sure whether it is useful.

    Hope this helps....

    Thanks,

    Jeroen

  • Hi Jeroen,

    Thank you for the screenshot.

    Yes, the 0100_0_16-bq34z100G1.bqz should match the firmware on the shipped device.

    Could you please try repeating the steps in this app note: www.ti.com/.../sluubv2

    If the issue continues to persist after stepping through that document, could you please send an analog scope shot of the I2C lines showing rise/fall times.

    Sincerely,
    Bryan Kahler
  • Hi Bryan,

    The communication issue is solved !
    By reading the app note, I realised that I used Port 1 of the EV2400 (SMB) instead of Port 2 (I2C). Changing that did the trick.
    bqStudio now shows the device, although the reported battery voltage is 757mV instead of 16.7V but that's another issue. I did not look into that yet.

    Thanks for being so patient !

    Jeroen