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.

DS100BR111: Compatibility with 10GBASE-R protocol?

Part Number: DS100BR111
Other Parts Discussed in Thread: DS110DF111

Hi, I would like to ask if the DS100BR111 redriver would be appropriate for supporting 10GBASE-R interfaces at an SFP+ port?

The 10 GBe signals are 10GBASE-KR from a 3U VPX general purpose processor (GPP), interfaces a backplane, traverses through a rear transition module (RTM) and terminates at an SFP+ port. A rough block diagram is shown below. I certainly have signal integrity concerns with insertion loss due to (4) 3U VPX connectors and trace lengths, but more importantly is that the link partner on the end of the fiber is expecting 10GBASE-R (no medium specified) with no FEC or auto-negotiation supported. I think a redriver solution may be suitable due to its protocol agnostic nature? If not, the other options I can think of are to use a separate FPGA or a -KR to SFI PHY chip to handle protocol translations.

I really appreicate any input on this.

  • Hi Jack,

    The DS100BR111 would be an appropriate selection for this situation.  It would help compensate for insertion loss due to your connectors and channel length.  However, if you believe additional signal conditioning might be required, you might also consider DS110DF111.  Since this is a retimer, it has more advanced signal conditioning capabilities.  However, since the retimer has a non-linear signal chain, you would have to disable link training.

    You might consider running IBIS-AMI simulations of the devices with your channel and connector models in order to add confidence to your design.  Also, if DS110DF111 is of interest, it is possible to design a socket supporting both DS100BR111 and DS110DF111 through the use of some zero ohm resistors and passives specific to each device.

    Thanks,
    Drew