• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 18 subscribers
  • Views 470 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.

LSF0204: LSF0204: Does LSF0204RUTR need a large current to sink?

CS SU
Prodigy 20 points
Part Number: LSF0204
Other Parts Discussed in Thread: TCA9800, TXU0304, TCA9509

Hi Team,

In EVT board we used TXS0104EPWR for i2c bus, A side pull up resistors with 2.4K, B side pull up resistors with 4.7K, all i2c devices test passed.

In DVT board we changed P/N to LSF0204, A/B side pull up resistors are the same as EVT board, but we saw i2c test random fail, we tried to adjust pull up resistors to lower, A side pull up resistors with 1K, B side pull up resistors with 2.5K, all i2c devices test passed.

XS0104EPWR and LSF0204 are level translators and can be used for I2C bus, I am curious why LSF0204 used higher value pull up resistors, i2c bus is not working, my questions are:

1. what the differene for XS0104EPWR and LSF0204?

2. Does LSF0204RUTR need a large current to sink?

 

Thanks,

CS

  • 0
    Guru 295240 points

    The TXS has integrated pull-up resistors and edge accelerators. So the TXS and LSF will have different electrical characteristics.

    What pull-up resistors can be used depends on the total bus capacitance, and is limited by the signal frequencies you want to use and by the drive strength of all I²C devices. See the I2C Bus Pullup Resistor Calculation application report.

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    Hi,

    We already adjusted pull up resistors with lower value to enhance I2C driving current, but i2c bus was still unstable at 65C, it looks like LSF0204RUTR is very sensitive, now we have a chance to respin schematic, can you suggest the suitable part number for i2c bus for level translator (1.8V to 3.3V, industrial part).


    Also, we also used LSF0204RUTR for SPI interface in my design, we also have SPI unstable issue, please kindly suggest the suitable part number for SPI interface for level translator (1.8V to 3.3V,industrial part).

    Thanks,

    CS

  • 0 in reply to CS SU
    Prodigy 20 points

    Hi,

    One more question, please provide LSF0204RUTR VIL/VIH/VOH/VOL spec below.

    A Side

    LSF0204RUTR

    B Side

    　

    Min

    Max

    　

    Min

    Max

    VIL

    ?　

    ?　　

    VOL

    ?　

    ?　　

    VIH

    ?　

    ?　　

    VOH

    ?　

    VOL

    ?　

    ?　　

    VIL

    ?　

    ?　　

    VOH

    　?

    VIH

    ?　

    ?　　

    Thanks,

    CS

  • 0 in reply to CS SU
    Guru 295240 points

    The LSF is a passive analog switch; it does not have digital inputs or outputs and therefore does not have these specifications.

    I do not know why your bus is unstable. The pull-up resistor values are either too low or too high; check the waveforms with an oscilloscope.

    I suspect that not all your devices have enough drive strength for such strong pull-ups. Use a buffered I²C level shifter like, for example, the TCA9800/1/2/3 (it must not have pull-ups on the B side; place the B side towards the master because the current source it is not compatible with an I²C switch).

    SPI uses unidirectional signals; use a unidirectional level shifter like the TXU0304.

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    Hi,

    Here is i2c topology below, CPU is master with 1.8V, RTC is slave with 3.3V, we used a LSF0204RUTR as level shifer.

    If LSF0204RUTR have no digital inputs or outputs, LSF0204RUTR how to identify signal Hi or Lo from master/slave I2C device.

    I made a table for static characteristics, is it correct? please review and advise.  

    Why I used TXS0104EPWR in EVT board, we don't see any I2C unstable problem?

    Why I used LSF0204RUTR in DVT board, we see I2C unstable problem?

    Schematic/layout design is the same for EDVT, only change level shifer, why is there such a big difference?

    Thanks,

    CS

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    Hi,

    TCA9800/1/2/3 can not have pull up on B side, it is not flexible for debugging, that means designer can not adjust i2c drivie current by pull up.

    can you please suggest another part number for i2c level translator (1.8V to 3.3V, industrial part).

    Thanks,

    CS

  • 0 in reply to CS SU
    Guru 295240 points

    For voltages below Vref_A, the LSF works as an analog switch; the switch itself has an on-resistance of about 5 Ω. (For high voltages, the switch is off, and the output voltage is determined only by the pull-up resistor.)

    I do not know why your I²C is unstable; it is necessary to look at the waveforms with an oscilloscope.

    To change the TCA980x pull-up current, just use a different 0/1/2/3 device.
    It might also be possible to use the TCA9509, which has the voltage offset restrictions on the A side.

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    Thanks Clement.

    There is a glitch at falling edge on scl/sda signals when read/write i2c, do you have idea how to eliminate the glitch?

    Thanks,

    CS

  • 0 in reply to CS SU
    Guru 295240 points

    Please show the signals at both sides of the translator.