• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 1488 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.

TXB0102: only translating in one direction

Justin James
Prodigy 60 points
Part Number: TXB0102
Other Parts Discussed in Thread: SN74LVC1G125, SN74AHC1G125, SN74LVC1T45, SN74AXC1T45

First time using the TXB0102 and I can only get it translating in one direction.  I'm using it to interface a 5V ATMEGA2560 hardware UART port to a 3.3V device with VOH 2.4-3.1V (TXB0102 table says VCCI x 0.65 so is well within spec) and VIH 2.0-3.6V.
I have VCCA @ 3.3V and VCCB @ 5V with OE controlled by an IO on VCCB and can get data from 5V>3V but not the other direction.  Any ideas?

  • 0
    Prodigy 60 points
    I should add that the screenshots are taken on the matching A/B pins of the TXB
  • 0
    Guru 316780 points
    Do you have any pull-up resistors? Please show the schematic of your circuit.

    How exactly did you configure the ATmega's pins?
  • 0 in reply to Clemens Ladisch
    Prodigy 60 points

    No other components are on the labelled nets, they go direct to the ATMEGA.  RX/TX is configured as hardware UART and GPEN/GPWR are configured as outputs pulled high or low.

  • 0 in reply to Justin James
    TI__Guru 62080 points
    Thanks for the schematic Justin.
    Dont see any pullup or pulldown resistors on the device. So most likely it could point to the fact that the GPS module may not have enough drive strength to overdrive the line to switch from low to high.
    As pointed in the datasheet, the driving device requires at least 2mA (worst case) drive strength in order to switch.
  • 0 in reply to Justin James
    Guru 316780 points
    The L80 "Hardware Design" document does not specify the drive strength of the TXD pin. Try asking their support, but it appears to be too low.

    Anyway, you do not actually need the automatic direction sensing. Just use two unidirectional translators, e.g., SN74LVC1T45 or SN74AXC1T45; for downtranslating, you could also use a buffer with overvoltage-tolerant input, e.g., SN74AHC1G125 or SN74LVC1G125.
  • 0 in reply to Clemens Ladisch
    TI__Guru 62080 points
    Clemens,

    As always, thank you for your valuable inputs.

    Justin,

    One thing to note here is that the AXC1T45 device can work for less than 3.6V and cannot support up to 5V.
    The SN74LVC1T45 will do 3.3<->5V without issues.
  • 0 in reply to ShreyasRao
    Prodigy 60 points
    Thanks for the responses.
    I'm having the exact same problem with an Iridium9603 modem too (Working from 5V -> 3V but not 3V ->5V) and that has a drive strength of 4.4mA/5.5mA for IOL/IOH. I'm wondering if it has anything to do with the voltage I'm pulling OE to and it being VccB's voltage and not VccA's? The datasheet says it's referenced to VccA but it can take 5.5V on OE which I assumed to mean it just needs to be pulled above VccA. Do you think it could be that it being pulled to VccB is causing this issue as the layout with the Iridium is the same?
    I know I didn't need the direction sensing and bidirectional translation but I was trying to simplify circuit design and size by having everything on a single IC and on paper the TXB0102 should have worked (and did work about 5% of the time).
    SN74LVC1T45 x 2 was going to be my next step if I can't get this working but that means binning these PCBs and another 2 week wait which I was hoping to avoid if there were a workaround (I'm thinking of tying OE to VccA and seeing if that makes any difference)
  • 0 in reply to Justin James
    TI__Guru 62080 points
    Justin,

    since the OE pin is overvoltage tolerant, it can have a voltage above Vcca without issues.
    You can definitely try having the OE pin to Vcca but I would doubt it will change anything.