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.

HD3SS3412: Crosstalk Data

Part Number: HD3SS3412

Hello,

Do we have any data relating to crosstalk performance of the HD3SS3412?

Don

  • some data on page 10 of datasheet, is this the data you are looking for?
  • Thank you.   

    Figure's 5 and 6 detail crosstalk and off-isolation:

    Figure 7 details the cross-talk setup:

    Figure 8 details the off-isolation setup:

    Some questions:

    1. Are these measurements taken at room-temperature?   If so, does crosstalk and off-isolation vary over temperature?   By how much?

    2. The figures show adjacent channels, thus it should represent a worst case.    Do we have the same data from channels further away (i.e. A0 and A3)?

    3. What is the magnitude of the signal injected into the device?   I understand that the measurements are ratiometric, but would the ratio change for large/small signal swings?

    4. Do we have any additional characterization data available that's not included in the datasheet?

    Don

  • sorry for late response, the data on datasheet is typical data took at root temperature. We can't find data over temp and voltage any more.
  • Hi Brian,

    I appreciate that you eventually answered a portion of my question.   There are a few additional facets to the previously posted question, many of which are not addressed with your short response.   Would you please re-read and provide a comprehensive answer?

    Many thanks in advance.

    Don

  • Hi Don,

    1. I would not expect the crosstalk and isolation to vary over temperature.  The package and device structures are stable over process and temperature so I would expect very little change in measurement. Likely less then 1 dB, but it may be a little higher depending on setup repeatability.

    2. I have not taken data on this device, but other devices in QFN packages show isolation improving by 10-15dB for each extra pin of distance between the victum and aggressor.

    3. Typically signals injected are fairly small - 50 to 500mV - just depends on the default.  For the passive switches this will not change the results unless the signal was too very large and started to change the electrical performance of the switch.  This would likely mean lower bandwidth seen in the measurement.

    Regards,

    Lee

  • Thanks Lee,

    Some additional questions:

    1. In the case of the data presented in figures 5 and 6, do we know what the test conditions are?

    1a. Can I expect this kind of performance so long as I keep the single-ended voltage of each line within the VDD supply range?

    Thanks again.

    Don
  • Hi Don,

    I do not have the specific setup information for Fig5 and Fig6, but data was collected for all possible differential paths (A0 - B0, A0-C0, A0-A1 etc.) across VDD and temperature ranges.

    For single ended operation, you can expect similar performance if your single ended channels are in separate pairs (like A0+ to A1+).  Within a single pair there will be more coupling due to package bondwires and on-device coupling capacitance being in close proximity.

    The recommended voltage range extends from -0.9V to 2.9V, above 2.9V you will likely experience a small reduction in bandwidth.  This should not present any issue with LVCMOS single-ended signals since the typical switching threshold is usually set at 1.5V or ~ VDD/2

    Regards,

    Lee