• State Answer Suggested
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 296 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.

TCA5405: Power-up Sequencing Requirement

Jejomar Ildefonso
Genius 9490 points
Part Number: TCA5405

Hi Team,

The TCA5405 datasheet doesn't seem to specify timing requirements for the pins in terms of power supply. Should VCC be provided before high level on DIN occurs? Our customer noticed that sometimes outputs are high for a short period of time during power on scenario of the final system with the master host that manages the IC.

They did performed a simple experiment:  They provided high level to DIN from external power supply 1.8V and triggered VCC (1.8V too) few times. In almost all of the time, inputs become high on power on. In most cases they remain high. It looks like there is a requirement not to have a high level on DIN before the IC is fully powered exist.

Unfortunately we can't locate a timing/sequencing requirement in the datasheet. Can you help us specify these requirements precisely to our customer? Thanks in advance!


Kind Regards,

Jejomar

  • 0
    Guru 316920 points

    The DIN input supports up to 4 V, even when powered off. It should never float, so if the controller is too slow to start up, add a pull-up resistor.

    As far as I can see, the datasheet does not specify the initial state of the output.

  • 0 in reply to Clemens Ladisch
    TI__Genius 9490 points

    Hi Clemens,

    Many thanks for your response. But we're not quite sure what exactly problem do you cover by saying "So if the controller is too slow to start up, try adding a pull-up resistor". Are you pertaining to pulling up DIN before VSS?

    The problem seems to be opposite. Our customer does not encounter the issue if DIN provided later than VSS to the TCA5405. The problem is when DIN gets high levels earlier than VSS.

    Hi E2E Team,

    Can you help us if there's any information about the output's initial state, or proper power-up sequencing to the device that was not mentioned in the datasheet? Thank you!


    Kind Regards,

    Jejomar

  • 0 in reply to Jejomar Ildefonso
    TI__Guru 73181 points

    Hi Jeojomar,

    I'll get back to you tomorrow. I believe I will need to locate the design file for the device to see if there is some kind of back biasing current from DIN into Vcc. 

    -Bobby

  • 0 in reply to Jejomar Ildefonso
    TI__Guru 73181 points

    Hi Jeomar,

    I took a look at some of the existing design files for the device. I was only able to find the CMOS inputs and a series ESD cell infront of the DIN pin. I didn't find anything that would suggest a leakage path from DIN. 

    The state diagram for the device also didn't show anything aside from the first falling edge on DIN being used to turn on the internal oscillator. 

    Judging from the customer's experiment there may be an issue with the device's PoR circuit when DIN has a bias voltage. Is it possible for the customer to have the DIN voltage low before powering up the device? 

    -Bobby

  • 0 in reply to BOBBY
    TI__Genius 9490 points

    Hello Bobby,

    Many thanks for your response. I've shared this accordingly with the customer and had their feedback as follows:

    "There is something about the falling edge that starts internal oscillator but any experiments of the evaluation board are difficult because DIN is too sensitive to any noise - basically it can switch ports when your hand is approaching the evaluation board. We may need to get back to this later after 3 months or so, when this will be integrated to the real design, which actually sounds not prospective from our first trial of this IC. It seems doesn't have any EMI protection and it doesn't have CRC. Correct?

    What do you mean by "when DIN has a bias voltage". Do you particularly say that the IC cannot be powered from 3.3V if DIN is 1.8v?

    Do you think it would be a good idea to pull up DIN with 10K from 3.3v (we have 5V tolerance pins on MCU, while MCU is powered from 1.8V)?"


    Kind Regards,

    Jejomar

  • 0 in reply to Jejomar Ildefonso
    TI__Guru 73181 points
    Jejomar Ildefonso said:
    It seems doesn't have any EMI protection and it doesn't have CRC. Correct?

    Correct. This device wasn't designed to have built in EMI protection and it's communication protocol is a custom one which doesn't include any CRC. 

    Jejomar Ildefonso said:
    What do you mean by "when DIN has a bias voltage". Do you particularly say that the IC cannot be powered from 3.3V if DIN is 1.8v?

    I mean, you shouldn't apply any voltage to DIN while the device is unpowered. Bias voltage here means any voltage that is not GND.

    Jejomar Ildefonso said:
    Do you think it would be a good idea to pull up DIN with 10K from 3.3v (we have 5V tolerance pins on MCU, while MCU is powered from 1.8V)?"

    You'll likely see more current draw with an external 10k. If the MCU is driving to 1.8V the VoH will likely slowly rise to about 3.3V but will depend on the switching frequency. It shouldn't damage the device but you may end up having some kind of leakage when you drive high with the MCU. If the output stage of the MCU doesn't have some kind of internal protection you may end up biasing the MCU to 3.3V minus Vth. 

    -Bobby