Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 19 subscribers
  • Views 654 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.

TXB0102: UART from 3.3v Raspberry Pi to ATMega2560

eljeffo
Intellectual 560 points
Part Number: TXB0102
Other Parts Discussed in Thread: TXS0102, SN74LVC1T45

Hello! 

I previously used a simple MOSFET N-Type Voltage Converter to link a Raspberry Pi 3.3V UART to an ATMega2560 5v UART.  It worked fine.

So I made a board using a TXB0102, enabled OE after boot, and tried to talk, but so far the devices don't talk to each other.

It seems both devices should be able to source enough current, but perhaps that's not the cause of my problem.

Do you think I need a TXS0102 instead?  It's not really like serial has any pullups though, so I'm not sure it's the right approach.

  • 0
    Intellectual 560 points

    As a note, voltage conversion was tested prior to install, and the TXB0102 is operational.

    Voltage Conversion from RPi Tx 3.3->5V is working, but when connected to the ATMega2560, the square wave of 5v turns into some strange 10v (0v to 10v) peak to peak sine wave.

    Tx from ATMega2560 to RPi Rx 3.3v appears as a 5v peak to peak sinewave on the 5v side.  

    Needless to say, these kinds of waveforms aren't going to be read as serial data.  Is it a pretty clear symptom of me picking the wrong chip?

  • 0
    Intellectual 560 points

    I tried some lighter/thinner/shorter wire from a CAT5 stranded cable just to reduce any possible capacitance.  Doesn't seem to help.  

  • 0 in reply to eljeffo
    Guru 278910 points

    What is the total capacitance of the cable and all connectors? The TXB does not work with more than 70 pF.

    The TXS0102 would probably work better (the idle state of UART signals is high, so the pullups make sense).

  • 0 in reply to Clemens Ladisch
    Intellectual 560 points

    I will stuck my meter on the wires with the circuit off and measure the capacitance and get back to you. 

    is there a better way to get a capacitance reading?

  • 0 in reply to eljeffo
    Guru 278910 points

    The actual waveform would probably be more helpful.

  • 0 in reply to Clemens Ladisch
    Intellectual 560 points

    Well... with just the wires it was 56pF, but with the ATMega attached, I got a whopping 150pF on the meter.

    The actual waveform is kinda... uh... I think useless.  Looks like a crazy out of control sinewave.  Nothing resembling a digital square wave (which I have confirmed is fine without the ATmega).

    I'll try to mess with the wires to reduce the capacitance and see what happens.

    So the TXS0102 doesn't have a capacitance limit?

    I see the TXS0102 VSSOP is completely sold out almost everywhere.  Since I'll probably be forced to change chips.

    Is there a favorite uni/bi-direction TXS kind of chip that you recommend?  I'm fairly open, but just gotta make sure in this crazy global shortage that I can get a few hundred chips and get my work done :)

  • +1 in reply to eljeffo
    TI__Mastermind 44655 points

    Hi,

    I typically recommend avoiding auto direction sensing devices for unidirectional signals due to them all requiring lower load capacitances. This may be a better accomplished with a device like the SN74LVC1T45 and then you would just use two (RX and TX signals). This device has strong buffered outputs making it more suitable for heavier capacitive loads. Looking at the TI store, there are quite a few SN74LVC1T45DCKT available. 

  • 0 in reply to Dylan Hubbard
    Intellectual 560 points

    Thank you for the guidance.  Indeed, magnet wire in a short length resulted in a functional circuit so as you said it is capacitance related.

    I'll swap over to the part you recommend.