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.

TXS0108E: Abnormal output voltage on port B...

Part Number: TXS0108E
Other Parts Discussed in Thread: LSF0108

Dear Sirs,

My customer use TXS0108E for 1.8V to 3.3V level shift for SMBUS.

However, they found while the input signal is "LOW" at port A, the output signal on port B is not "0V" but around 0.4V.

Is there anything I can fix for this issue?

Thanks.

  • Hi Peter,

    There are a couple of reasons why this is occurring. If the input signal at port A does not go to 0V then the output port B will also not go to 0V.

    This happens because when the line is pulled low by an external driver, the TXS0108E will sink current, IA, through that external driver. If that driver has some resistance to ground, then the current, IA, will create a voltage causing you to see that .4V. I can help you find a solution here by possibly placing an open drain buffer or to use some other device beside the TXS. If you provide me with a schematic, I can help you find parts.

    Here is an app note about the operation of the TXS devices. www.ti.com/.../scea044.pdf

    Another possible reason is if you are using external pullup resistors around the TXS device, they will be in parallel with the internal pullup resistor in the TXS and may reduce pullup resistance causing more current, IB, to be driven through that same resistance causing a high voltage like .4 V instead of the 0V you expected. The solution here would be to not use external pull up resistors to the TXS device, and if they are necessary, please provide me with a schematic so I can help pick another device for translation.

    Here is an app note on effects of pull up resistors on TXS devices. www.ti.com/.../scea044.pdf



    Here is a training series that talks about using our LSF devices for translation. These are very similar to the TXS family, except the TXS family has internal pull up resistors and one shots. training.ti.com/TLM-LSF-Down

    Differences between the auto directional translation familes. training.ti.com/TLM-LSF-Down

    Thanks!
    - Karan

  • Hi Karen,

    Thanks for your reply!

    Attached is customer schematics, pls help to give some advice.

    Thanks.

  • Hello Peter,

    This is not the full schematic. I cannot tell what the SMbus is connected to. If there are any external pull up resistors this would be a problem.

    You said the output was .4V can you let me know what the input was to the device? This appnote discusses what is expected of the output when the input is not at 0V. www.ti.com/.../scea056.pdf

    If the SMBus input is not reaching 0V then that means it is probably a weak driver and the TXS device will give an output that is not 0V. Does the fact that the output is .4V impact any devices on the SMBus? If not, then you can just continue using the TXS device with this condition. If you want to mitigate the .4V then you could use the LSF device with larger pullup resistors.

    Thanks!
    -Karan
  • Dear Karen,

    Attached is the complete schematics.

    I will study the appnote, too.

    Thanks.

  • Hi Peter,

    I have removed your schematic from your post as this is a public forum. However, I do have a copy that I will review for your issue.

    Thanks!
    - Karan
  • Hi Peter,

    After reviewing your schematic, I see that you have pullup resistors on both the input and output of your TXS device as shown here.

    As I mentioned earlier, these external pullups are in parallel with the internal pullups of the TXS device. If you remove these pullups, your issue should go away.

    Read this appnote on the effects of external pullup resistors. 

    The summary is that an external pullup while driving low is in parallel with the internal 40kOhm resistor resulting in a Req = (40*2.2)/(2.2+40) = 2.08kOhm resistance. Not only this but you also have a pullup on the output which will cause the same problem.

    This is the impact on VOL from an external pullup resistor taken from the appnote I attached. Extrapolating this data, a 2.2k or 3.3k pullup resistor that exists on this system will cause a .4V VOL just as you are seeing. Solution is to remove these Rpu.

    I have deleted the copy of your schematic as I have helped your issue, if you any other questions related to this please only share the screen captures of the parts I need to see and not the whole schematic if it is confidential.

    Thanks!

    - Karan

  • Hi Karen,
    As you can see in the schematics, the value of R with "X" in front of it, it means no use on system.
    Therefore, the R101,R102, R125, and R126 is not use.
    Could you have any idea about the reason of issue?
    Thanks.
  • Hello Peter,

    It is most likely that your driver is not being able to sink the current causing it not be able to force a 0V on the line.
    Is this issue a problem for you? The logic level low threshold is much higher than .4V so any devices on the SMB should go low at .4V.

    Secondly, if this is an issue, you can try using the LSF0108 with 10K pullup resistors. This would be double the pullup resistance to the TXS0108 meaning you should get a low voltage of .2V instead of the .4V you see.

    Thanks!
    -Karan