• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 21 subscribers
  • Views 533 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.

TXS0104E: output amplitude dropped below specific output bias voltage

lam huynh
Prodigy 190 points
Part Number: TXS0104E
Other Parts Discussed in Thread: SN74CBTLV3257, TS3A5018, TMUX1574, SN3257-Q1, SN74AXC4T774

Hello

I use the TXS0104E to translate voltage for a Quad-SPI bus from MCU STM32F777 run at 3.0V (VCCB)  to a 2:1 MUX P/N SN74CBTLV3257  run at 1.8V (VCCA) (*). The MUX output is connected to a 1.8V Micron flash.  The voltage translator is used as is without external pull resistor.

We run on a clock ~ 16.7 Mhz.  I notice that the output signal amplitude of the TXS0104E reaches 1.8V at the beginning of the signal for a very short time and then drops quickly to stabilize around 1.5V.  That observation is for SPI bus data outputs.  For clock signal, its amplitude is constant around 1.6V.   Input signals are clean transitions between 0V and 3V.

Question: why can the amplitude not reaching and stabilizing at 1.8V?

(*)  the MUX is specified for 2.3V to 3.6V operation, I assume running it at 1.8V would incur higher Ron and slower switching time but since we do not MUX the Quad SPI bus in real time but only occasionally between MCU and FPGA so it should not matter much

  • 0
    Guru 317480 points

    Can you show oscilloscope traces showing input and output?

    16 MHz is 32 Mbps, which is more than the TXS can handle at 1.8 V. Are you using 16 MHz or 16 Mbps?

    When you run a device outside its recommended operating conditions, there are no guarantees at all; it's possible that it will not work at all. Other muxes like the SN3257-Q1 (not yet available), TS3A5018, or TMUX1574 are specified for 1.8 V.

  • 0 in reply to Clemens Ladisch
    TI__Mastermind 46195 points

    Hi lam,

    As Clemens has said, that data rate is exceeding the TXS0104E capabilities. I would also suggest taking a look at the SN74AXC4T774 as it operates at those voltage ranges and can handle that data rate.

  • 0 in reply to Clemens Ladisch
    Prodigy 190 points

    Hello Clemens,

    I would need some clarification:

    The specs of TXS0104E says max 24 Mbps.  I thought in Single data rate mode I would need 24 MHz clock to achieve a transfer of 24 Mbps.  Unless that 24 Mbps max specs indirectly mentions that the chip cannot withstand any edge transition faster than 41.7uS, which implies the max signal frequency the chip can handle is 12 MHz.

    Thanks for the tip on TMUX1574, it would work for me

    Thanks

    Lam

  • 0 in reply to lam huynh
    Guru 317480 points

    Section 6.7 of the datasheet shows that the minimal pulse duration is 41 ns. This corresponds to a square wave with a frequency of 12.2 MHz. Transmitting 24 Mbps is not possible with SPI, but only with some other protocol that does not need two signal edges per data data bit (e.g., UART, or some self-clocking signal).