• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 34 subscribers
  • Views 426 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TCA9406: TCA9406

Rok Pirnat
Prodigy 10 points
Part Number: TCA9406
Other Parts Discussed in Thread: TXS0102

Dear Sir or Madam,

TCA9406 datasheet states VCCA must be less than or equal to VCCB. What are the tolerances when VCCA equals VCCB? Especially when both are powered from different 3.3V (nominal) power supplies.

Thank you for you reply in advance

Regards,

  • 0
    TI__Mastermind 49370 points

    Hey Rok,

    I can run some test and see if the performance of the device is affected when VccA is larger than VccB. Typically though, when I see that there is a requirement like this, it involves an internal diode or ESD cell which can be turned on when VccA larger than VccB by Vth (typically 0.7V). I can try looking for diode drops between VccA and VccB to confirm if this is possible though the datasheet shows that VccA can be larger than VccB during turn on so I suspect this isn't why the device is spec'd this way.

    I will look for an EVM to see if I can perform any testing. I will let you know if I am able to locate one tomorrow.

    Thanks,

    -Bobby

  • 0 in reply to BOBBY
    TI__Mastermind 49370 points

    Hey Rok,

    I was able to find an EVM, I did some initial tests of probing for internal diodes and found SDA/SCL and the Vcc pins have diodes coming from GND into their respective pins. No other diodes between pins were detected. This means the theory of a diode between VccA and VccB is ruled out.

    Also found that at 3.3V on VccA and about 2.6V on B side, supply current from Vcca went from uAs into about 1mA and after going higher or lower on VccB the supply current returned back to uAs. This is probably something to do with some partial conduction (turn on) of some CMOS inputs (gates).

    I did some thinking on why VccA should be lower than VccB.... I believe this device uses A side as the reference voltage for ViL and the rise time accelerator voltage triggered threshold. I will do some testing with this tomorrow and get back to you.

    Thanks,

    -Bobby

  • 0 in reply to BOBBY
    TI__Mastermind 49370 points
    Hey Rok,

    It looks like if VccB goes below ~2.6V while VccA is 3.3V the device will stop working (to me it seemed like OE was disabled). I also did a test where VccA was 3V and found if VccB went below 2.3V the device stopped working. To me this means VccB cannot fall below VccA-0.7V or else the device goes into reset.

    Also, I found that the device does reference the rise time accelerator trigger point at 30% of VccA (even if VccB is lower).

    Now to answer your original question, to me it 20% tolerance (At VccA=VccB=3.3V) is allowed but I would suggest keeping it to 10% or lower.

    Thanks,
    -Bobby
  • 0 in reply to BOBBY
    Guru 244220 points

    The TCA9406 is identical with the TXS0102. VCCA and VCCB Bias Requirements for Bidirectional Translators (SCEA060) says:

    TXS and TXB-type translators require that VCCA is less than or equal to VCCB. This is due to an internal protection diode that can become forward biased when the voltage on VCCA exceeds the voltage on VCCB. […] VCCB fixed at 2.5 V and VCCA swept from 2.5 V to 3.5 V:

    As shown in Figure 4, … when the voltage on VCCA exceeds the voltage on VCCB by roughly 0.6 V, the diode begins to conduct and the ICCA current begins to drastically increase as VCCA increases.

    So a difference of less than 0.5 V is safe.