• State TI Thinks Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 6 answers
  • Subscribers 32 subscribers
  • Views 237 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DS110DF111: Would it be a suitable part to connect 2 SFP+ ports TX lane (or RX line) alone

Felix Arul Rajesh Francis
Intellectual 470 points
Part Number: DS110DF111
Other Parts Discussed in Thread: DS100DF410, DS110DF410, DS125DF1610, DS250DF230

Hi

Our requirement is to use one DS110DF111 part and connects only the TX ports (or RX line) of 2 SFP+.

So the topology(with SFP TX) would be as below,

ASIC Channel 1 RX <== Retimer Channel 1 OUT <==Retimer Channel 1 IN <== SFP+1 TX

ASIC Channel 2 RX <== Retimer Channel 2 OUT <==Retimer Channel 2 IN <== SFP+2 TX

Would this part, DS110DF111, be a suitable one for this requirement? Kindly confirm

Regards,

Felix.

  • 0
    TI__Mastermind 19135 points

    Hi,

    The DS110DF111 is a one-lane bidirectional retimer. Thus you would need one DS110DF111 IC chip per SFP+ port if you want retiming function to be implemented both in the host egress and ingress directions.

    Regards,

    Rodrigo

  • 0 in reply to Rodrigo Natal
    Intellectual 470 points

    Hi Rodrigo,

    Here our intention is to directly drive the SFPs from the ASIC without any PHY or EDC device...

    In certain cases of power-limited tests with positive dispersion of 1600ps/nm(which is equivalent to 80km of fiber) the serdes cant work.

    Basically, we can't get zero BER with the limited power of -23dBm.

    The issue has nothing to do with ASIC Tx and is only for the ASIC Rx. Hence we prefer to use the retimer device only on the ASIC Rx path.

    Let me know whether in this scenario the above device and the configuration that we asked for would be fine to have.

    Also, we were exploring DS110DF410 or DS100DF410 parts also....they are 4 TX & RX channels and we need to connect only 2 RX channels...we  felt its an overkill in terms of size, cost and application....that's why we prefer to have  DS110DF111.

    Kindly confirm the suitability...

    Regards,

    Felix.

  • 0 in reply to Felix Arul Rajesh Francis
    TI__Mastermind 19135 points

    Hi Felix,

    The DS110DF111 may be used for 10G SFP+ ZR applications. We have OEM customers already using this retimer part numbers for that application case. If you are only using retimer on ingress path then a single DS110DF111 would technically be sufficient for two SFP+ ports. However, one potential issue would be that your board routing might a bit unideal as the retimer channels would be oriented differently. As seen below channels A and B have their respective Rx and Tx pins on opposite sides of the chip. if you can live with this issue then DS110DF111 could be option for you.

    The DS110DF410 on the other hand is a unidirectional device.

    Regards,

    Rodrigo Natal

  • 0 in reply to Felix Arul Rajesh Francis
    TI__Mastermind 19135 points

    Please share your company name and work email and I can share some representative DS110DF111 BER test results for 80km link case.

    Thanks,

    Rodrigo Natal

  • 0 in reply to Rodrigo Natal
    Intellectual 470 points

    Hi Rodrigo

    Thanks for the input. It helps...

    Meanwhile, bringing up another clarification on this part regarding the thermal...

    We may end up using this part with an internal unit ambient of around 103degC...

    Considering the power dissipation of around 700mw, the simulation gives the junction temperature as 108degC(not sure its correct)...

    The datasheet doesn't have the recommended junction temperature or max junction temperature....

    Could you please share the Recommended Tj and Max Tj for this part?

    Also, let us know if we have any Retimer parts that can withstand the higher ambient temp >100 

    Regards,

    Felix.

  • 0 in reply to Felix Arul Rajesh Francis
    TI__Mastermind 19135 points

    Hi Felix,

    The maximum operating junction temperature for this part number is 105degC. Please refer to the TI product page for quality and reliability test result details (see link below)

    https://www.ti.com/quality-reliability-packaging-download/report?opn=DS125DF111SQ

    From the TI 10G portfolio the DS125DF1610 supports max junction temp of 115C. It is a 16-channel device though, much larger than what you are targeting.

    https://www.ti.com/product/DS125DF1610

    From the TI 25G portfolio the DS250DF230 is a 2-channel device with max junction temp of 110C. This retimer can operate at 10Gbps in addition to 25Gbps. Form factor wise the DS250DF230 would be a good fit for your application as it is 2-channel unidirectional. 

    https://www.ti.com/product/DS250DF230

    Regards,

    Rodrigo Natal

  • 0 in reply to Rodrigo Natal
    Intellectual 470 points

    Hi Rodrigo,

    Thanks for the Tj Max data for these parts....we are much interested to know any part that has Max ambient temp greater than or equal to 100

    We are not seeing any one from TI website. Kindly share if there are any additional details.

    Regards,

    Felix.

  • 0 in reply to Felix Arul Rajesh Francis
    TI__Mastermind 19135 points

    Hi,

    All of TIs retimers are specified to support max ambient operating temperature of 85C and a max junction temp in the range of 105C to 115C. This performance meets the requirements of both commercial and industrial applications. 

    Thanks,

    Rodrigo Natal

  • 0 in reply to Rodrigo Natal
    Intellectual 470 points

    Hi Rodrigo,

    Thanks for this update...one quick check on the reliability data

    Here the usage temp is 55degC and Test temp is 125degC...how does both differ? Is the part exercised with 125degC.

    When we exercise higher ambient (100degC) apart from reliability (life time)  degradation, would that be any functional impact?

    Regards,

    Felix.

  • 0 in reply to Felix Arul Rajesh Francis
    TI__Mastermind 19135 points

    Hi Felix,

    Here the usage temp is 55degC and Test temp is 125degC...how does both differ? Is the part exercised with 125degC.

    This is an accelerated stress test. You are actually testing the device at a temp of 125C for 1000hrs to simulate operation at 55C for a long period of time (years.)

    When we exercise higher ambient (100degC) apart from reliability (life time)  degradation, would that be any functional impact?

    We do not guarantee compliance to all datasheet parameters for device ambient operating temp higher than 85C.

    Thanks,

    Rodrigo Natal

  • 0 in reply to Rodrigo Natal
    Intellectual 470 points

    Hi Rodrigo,

    Thanks for the input. 

    Meanwhile, we explored DS100DF410 and it has max Tj as 150degC and believes this should be a good choice now...if we are not considering the cost.

    Any comment on this...

    Regards,

    Felix.

  • 0 in reply to Felix Arul Rajesh Francis
    TI__Mastermind 19135 points

    Hi Felix,

    The Tj = 150C on the DS100DF410 is an absolute maximum figure i.e. beyond this temp you risk damaging the device. The max operating junction temperature for which DS100DF410 datasheet specs is guaranteed is 105C.

    Thanks,

    Rodrigo Natal