• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 19 subscribers
  • Views 1115 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.

Original question:

TXB0104: Unable to drive outputs to zero volts

TXB0104: B3 channel output could not go down to 0V

Rocson He
Intellectual 1830 points
Part Number: TXB0104

Hi Team,

I have customer using TXB0104-Q1, and now meet issue of VOL=0.4V issue, and also overshoot/undershoot.

While, replacing with P2P TXS0104-Q1 could help resolve the VOL=0.4V issue, but still have overshoot/undershoot.

See summary and waveform details as below.

It’s for UART signal level shifting, and the B3 channel is found with VOL of 0.4V rather than GND/0V, while the A3 is OK. And both RX/TX of B side, that is B3, B4, have overshoot and undershoot, while the 1.8V A side is good. The PCB trace length is around 18mm from B side to SOC/IO.

For the overshoot and undershoot, I think it should be relative to parasitic trance inductance and capacitance, but the trace length seems is not that long.

Could you help check if any suggestions to improve for this case, of over/undershoot and 0.4V/VOL step voltage issue ?

Thanks.




  • 0
    TI__Guru* 97046 points

    Hi Rocson,

    I would recommend in the future labeling your scope shots -- I don't know what signals are what on here.

    Also, giving a single signal for a device is typically not very helpful.  Try to always include an input and an output signal in the same scope shot -- otherwise it is impossible to determine timing and other I/O characteristics.

    *****

    Overshoot and undershoot are _extremely_ common in logic circuits.  This is typically caused by having a strong driver connected to a light load, and is explained by fairly complex control theory. The simple answer is, either increase the series line impedance (add a series resistor) or increase the line capacitance (add a capacitor to ground) to reduce ringing. Try not to exceed 70pF total on the line, or you could run into other problems.

    The V_OL issue is related to how the TXB device works.  The output is forced to 0V by a very strong output driver for a very short time period (~30ns), but it is then released and driven by the reduced strength output driver, which is shown here:

    With a 4kohm output driver, it doesn't take much current to produce a 0.4V drop:  I = 0.4/4000 = 100uA.

    It's not uncommon to see leakage currents in the 100uA range, so I would recommend looking at the devices connected to the TXB to determine where the current is coming from.