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.

HD3SS460: HD3SS460 isolation questions

Part Number: HD3SS460

Dear all,

For HD3SS460, for the SSRX,SSTX pin, when the DC block cap should be used and what value should be used?

In datasheet, here is Figure 5/6 with 500nF cap and 100kohm resistor, what is the purpose of the 100kohm resistor? can the resistor no connected?

Thanks.

  • Hi John,
    DC block is used to isolate common mode voltages. As long as DC common mode voltage(VCM) is within HD3SS460 requirement, there is no reason to have AC coupling Caps. It is better to have AC coupling Caps to isolate common mode voltages from one another. Otherwise, if VCM don't match then this would affect return loss and thus signal integrity.

    Choice of AC coupling Cap values, depends on the run length or maximum number of zeros or ones within a data pattern. The higher DC or low frequency content - due to the long run length coding - the higher the Cap value. To calculate Cap value, you have to compute RC time constant and max run length of your data pattern. RC time constant should be higher than the max run length.

    100K-ohm Resistor is used to optimize return loss and thus reduce reflection and improve signal integrity.

    Regards,,,Nasser
  • Hi Nasser,

    Thanks for your reply.
    Do we have some test data for the insertion loss of HD3SS460 for the 5GB and 10GB usb3.1 signal?
    My customer use this mux in their system, but sometime usb3.1 can not work properly, could you please help give advice what should be checked?

    Thanks.
  • Hi John,
    Insertion loss at 5GB or about 2.5GHz nyquist frequency is within 2dB. However, I believe device has not been characterized for 10GB operation.
    1). So the first question is, your customer has problem at 5GB or 10GB?
    2). When you say it does not work properly what do you mean? Are you saying they are getting bit errors? If indeed we are getting bit errors, then we have to run different test or experiment to find the root cause of these bit errors:
    2a). Are we sure this is related to the HD3SS460 device? Is there a way to bypass the device to prove this is caused by HD3SS460 device?
    2b). Any potential cross coupling that could cause this issue? Have we closely reviewed pcb layout to rule this out?
    2c). Have we used lab power supply to make sure noise from the switching supply is not causing signal to noise degradation?
    2d). Could there be EMI causing this? Are we using strip line or micro strip traces?

    In summary, first we should try to find the root cause and then try to find a possible solution. Please let us know regarding items 1 & 2 so we can further understand the root cause of this problem.

    Regards,,nasser
  • Hi Nasser,

    The application is 5GB signal, and we don't have test equipment to test if it is bit errors or not, we just connect the USB3.0 U disk to the port and find sometimes it is not be recognized correctly.
    We have try to remove the mux and connect the connector to the platform directly, it can recognized the USB3.0 device always.
    So we are suspecting HD3SS460 may have some impact for the signal quality, but we don't know how to test the signal and to find the root cause.

    Thanks.
  • Hi John,
    Please note points 2b through 2d that i noted in my earlier email. Please run these experiment to see if you can find the root cause of the problem.

    Also, do you have scope to check output of the HD3SS460 to see if you notice any abnormality?

    Regards,,nasser
  • Hi Nasser,

    Thanks for you advice, I will check with them.
    Let us close this thread, anything need, I will start a new one.

    Thanks.