Due to the U.S. Thanksgiving holiday, please expect delayed responses during the week of 11/22.

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.

TIDA-01021: Available For Purchase?

Part Number: TIDA-01021
Other Parts Discussed in Thread: ADC12DJ3200EVM, TIDA-01022, ADC12DJ3200, TIDA-01023, TIDA-01024, ADC12J4000, ADC12J4000EVM

Hi,

My group at Lawrence Livermore National Laboratory, a research laboratory and not a company, is trying to synchronize two ADC12DJ3200EVM with less than 20ps board to board clock skew to act as a high performance oscilloscope. I've noticed the TIDA-01021 Design and it seems to have all the appropriate specs for our experiment. I notice when I looked up the TIDA-01021 on your website, it says this: Fully assembled board (shown above) developed for testing and performance validation only, not available for sale.

I was wondering, since my group is not part of a company, that we could purchase or lease a TIDA-01021 board for our experiment? If not, do you know of any other company that might sell a similar style clocking board that could synchronize two ADC12DJ3200EVM with less than 20ps board to board clock skew?

Thank you for your help.

Best,

Kevin

tidud80a.pdf

  • Hello Kevin,

    Welcome to the E2E Support Forums, and thank you for considering TI Products and TI Reference Designs.

    Kevin Chaves said:

    I was wondering, since my group is not part of a company, that we could purchase or lease a TIDA-01021 board for our experiment? If not, do you know of any other company that might sell a similar style clocking board that could synchronize two ADC12DJ3200EVM with less than 20ps board to board clock skew?

    A: Due to regulations, and as you point out reading the notice displayed in the TIDA-01021 Folder, the TIDA-01021 Reference Design is not for sale.

    However, note that we do provide all of the files (Gerber, CAD, Schematics, BOM, etc) that you'd need if you'd like to get one built yourself, to find these files go to the 'Technical Documents' Section of the TIDA-01021 Folder

    If you do decide to build your own, we do have a couple of videos talking about how to set up the Hardware and Software: 


    I wanted to also point out that given your requirements, we also have a few other TI Reference Designs that might we of interest to you, and that also include the Design Files.

    TIDA-01022, is similar to TIDA-01021 synchronizes two ADC12DJ3200 but in this case the ADCs are on the same board, and the design achieves a skew of < 5 psec (Test Results are available in Section 7.3 of the TIDA-01021 Design Guide).

    We also have two other TI Reference Designs (TIDA-01023 and TIDA-01024) that show how you can scale up multi-channel clocks using Tree and Daisy Chain configurations:

    TIDA-01023 features the 'Tree Configuration', and it achieves a channel-to-channel skew of < 11 psec (Test Results are available in Section 3.2.2 of the Design Guide).

    TIDA-01024 features the 'Daisy Chain Configuration', and it achieves a channel-to-channel skew of < 6 psec (Test Results are available in Section 3.2.2 of the Design Guide).


    Let us know if you have any follow-up comments or questions.

    Note: that TIDA-01021, TIDA-01023, and TIDA-01024 share the same hardware.

  • Thanks Victor, that was really helpful!

    My group has two ADC12DJ3200EVM and then we have six other ADC12J4000EVM. Would the TIDA-01021,TIDA-01023,TIDA-01024 be able to synchronize the ADC12J4000 as well or does the hardware not match up?

    Also how does the clock skew change as you try to synchronize more than just one board for the TIDA-01023 and the TIDA-01024? Does the clock skew standard deviation get higher and by how much?

  • Hello Kevin,

    Kevin Chaves said:
    My group has two ADC12DJ3200EVM and then we have six other ADC12J4000EVM. Would the TIDA-01021,TIDA-01023,TIDA-01024 be able to synchronize the ADC12J4000 as well or does the hardware not match up?

    There is a variation on the clock input connections between ADC12DJ3200EVM and ADC12J4000EVM. So it may require a different architecture/hardware change for synchronization. We have not tested with TIDA-01021 yet.

    Please take a look at TIDA-01017 & TIDA-00432 reference designs, which uses 2 numbers of ADC12J4000EVMs and LMK04828EVM to achieve synchronization.

    Kevin Chaves said:
    Also how does the clock skew change as you try to synchronize more than just one board for the TIDA-01023 and the TIDA-01024? Does the clock skew standard deviation get higher and by how much?

    TIDA-01023 and TIDA-01024 designs were released with external high frequency inputs reference, where channel to channel skew validation done up to 2.4GHz input signal with less than 10ps. Hence, it doesn’t affect much clock skew for 2 boards.

    For more than 2 clocking boards synchronization, same should be possible but calibration should be done iteratively.

    Please let us know if you have any further questions on the same subject.

    Best Regards,
    Ajeet Pal

  • Kevin,

    Please see the response from our High Speed Data Converters team on your other E2E thread:

  • Hi Ajeet,

    Maybe I phrased my question poorly. Specifically for the daisy chain architecture (TIDA-01024), you are synchronizing the clocks of a single ADC with the preceding one and there is a channel to channel clock skew of 10 ps. Does that mean for a system of 4 ADCs, there can be up to a 30 ps channel to channel clock skew between the first ADC and the fourth ADC? Or with this system, you can ensure just a 10 ps channel to channel clock skew among all the channels?

    Thank you for your help.

    Best,

    Kevin
  • Hello Kevin,

    In Daisy chain architecture (TIDA-01024) and Tree architecture (TIDA-01023), both clocking boards (called, TIDA-01021 clocking boards) are synchronized to each other, where clock to clock skew is better than 10ps to all 4 clock outputs.

    With these synchronized clocks, all 4 ADCs (to each other) channel to channel skew is less than 10ps in a system.

    Hope it clarifies your queries. Let me know, if you have any further questions.

    Best Regards,
    Ajeet Pal