• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 47 subscribers
  • Views 1212 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS54J60 2nd Nyquist

Justin Correll
Prodigy 60 points
Other Parts Discussed in Thread: ADS54J60, LMK04828

Hi Jim,

Another question about the ADS54J60...My signal of interest falls in the 2nd Nyquist and I have researched the setup in the datasheet  on pg. 43.  I have set 0x68004E to 0x80 and 0x6800042 to 0x01.  Further, following the board schematic and layout files, I have bypassed the ADT1-1WT and am using a more appropriate balun.  

This issue, is that I got great performance improvement using the 2nd Nyquist IL correction on two different occasions, but I am now not able to get the same result.  Even though the said registers read back the correct IL correction values mentioned above, the performance is not reflecting the same data taken with the correction was seemingly turned on.  

For example, at 750MHz the SNR (before vs. after) is 62.95dB and 31.57dB.  

Do you have any performance data available that were taken in the 2nd Nyquist?  

Also, is there a specific SPI write sequence that I should follow to implement the IL correction?

Lastly is the performance increase that I have seen possibly due to some other artifact of the HSDC Pro software?

Thanks,

Justin 

  • 0
    TI__Guru* 91947 points

    Justin,

    What is the sample rate you are using, 983.04? Let me know what Fs and IF you would like to see data from and I will see what we have. Can you send me the balun part number you are now using? If we have one, maybe we can duplicate your test. What mode are using? Also, if you can send the complete register configuration, we can look to see if everything is set properly. I do not believe your problem is HSDC Pro related or SPI write sequence. In most cases, it is usually an incorrect or missing register value. 

    Regards,

    Jim 

  • 0
    TI__Guru* 91947 points
    Justin,

    If you put the board back to its original configuration, do you get SNR of 62.95dB?

    Jim
  • 0 in reply to jim s
    TI__Guru* 91947 points

    Justin,

    After writing to the IL registers, you must do a digital core reset. Follow your commands with 680000 0x1, then 680000 0x0. See the attached config file for an example. I was able to get 64dB with a 750MHz tone sampling at 983MHz.

    Regards,

    Jim

    ADS54J60_20x_with_loop_2_nyquist.cfg

  • 0 in reply to jim s
    Prodigy 60 points

    Hi Jim,

    Thanks for your help, it is working now.  I must have been doing the 0x680042 and 0x68004E writes in the wrong sequence.  I was simply appending those writes to the provided ADS54J60_LMF_8224.cfg file, but resetting the digital core twice.  I got it to work once yesterday, but was not able to repeat.  The SNR at 750MHz in the original config was indeed 62.95dB, with the new balun I'm getting slightly better over 63dB.  

    Thanks,

    Justin

  • 0 in reply to Justin Correll
    Intellectual 375 points
    Sequence of setup writes. Below is from the LMF_8224.cgf file:

    Does the statement "0x690000 0x81 // set CTRL K " directly write to 0x690000 regardless of the setting of general register 4 (data sheet page 43)? If so, why do we need general register 4 ? Is "ADS54Jxx_DIGITAL" a command to the config downloader? If so, why don't we have "ADSjxx_JESD_Digital" for the 0x690000 writes?

    Thanks , John Reyland

    LMK04828
    0x10F 0x66 //Enable SYSREF to ADC
    ADS54Jxx_ANALOG
    0x0000 0x81 // LMF = 8224
    0x0011 0x80
    0x0059 0x20 // always write 1 to bit 5
    ADS54Jxx_DIGITAL
    0x6800f7 0x01 // digital reset
    0x680000 0x01 // reset digital
    0x680000 0x00 // clear reset this pulse must be the last write to main digital 0x680000
    0x690000 0x81 // set CTRL K
    0x690001 0x01 // JESD Mode (bits 2-0),20X mode, 4 lanes per ADC
    0x690006 0x13 // set K to 20
    LMK04828
    0x10F 0x06 //Disable SYSREF to ADC
  • 0 in reply to John Reyland62
    TI__Guru* 91947 points

    John,

    If the pointer was not set to this page, no. If the pointer was already pointing to this page, then yes. You must first set the page to write to. The command ADSjxx_JESD_Digital does the writes to address 4004 and 4003 in the background to setup for the page write. This was a unique command created for our GUI to make the configuration files easier to generate. As long as all addresses you are writing to are within this page, the GUI does not have to keep sending this command.  

    Regards,

    Jim