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.

LMX2572LP: Phase Shifting using TICSPRO

Part Number: LMX2572LP
Other Parts Discussed in Thread: LMX2594, LMX2572

Hello,

I'm trying to experiment with the phase shifting functions of the LMX2572LP using the dev board and TICSPRO software. I have it set up to output 400MHz from the standard 100MHz reference it comes with. There's a section on phase shifting in the instruction manual for the eval kit (section 3.2.1 of http://www.ti.com/lit/ug/snau217b/snau217b.pdf). As far as I can see, my settings are the same as for the example given there, but I cannot see any phase shift on the 400 MHz output when I write the MASH_SEED value. As I understand it, writing this register should give a phase change if everything is set up properly.

I am measuring phase by looking at it on a scope. The scope is triggered on the 100MHz reference and because the output is an integer multiple of the reference, I can see the two signals together. I had expected that the output would shift relative to the input when I wrote values to MASH_SEED, but this is not happening. I’ve attached a screenshot of the TICSPRO window to show my current settings. Is there something I’m missing here?

Thanks,

Gordon.

  

  • Hello Gordon,

    If you program the mash seed this will increment the value of the phase offset relative to the divider value N by increasing the propogogation delay.

    800 seems like a large value relatively.  Phase shift in degrees = 360 × ( MASH_SEED / PLL_DEN) × ( IncludedDivide / CHDIV )

    There is a good description of this in the LMX2594 datasheet that applies to the LMX2572 as well in sections 7.3.11 and 7.3.12.

    Regards,

  • Hello Liam,

    thanks for getting back to me. While I agree that 800 might seem a large value, it's exactly the value used in the example shown in the documentation for the dev kit, and is also the example used in the LMX2572LP datasheet. I've tried larger and smaller values and there was no difference with any of them.

    At this stage, I don't really care how much it shifts the phase, I just want to confirm that I can do some kind of phase shift - I'll tweak the numbers later.

    When I look at it on the scope, there is absolutely no change in phase between 100MHz reference and 400MHz output after writing MASH_SEED. The two possibilities therefore seem to be:

    * The intended phase shift hasn't worked

    or

    * Phase has been shifted then very quikly been shifted back to what it was

    I want to shift the phase and for that shift to remain for as long as necessary until I do another phase shift. At the moment, for whatever reason, that isn't happening. /If you have any suggestions as to why this might be, or if I might have missed some particular register write, please let me know.

    Thanks,

    Gordon.

  • Hello Gordon,

    No, it should increment the phase each time you write to the mash seed until you reset or synch.

    I am not sure what is going on here as this feature is well used.  Can you confirm the mash seed register reads back as expected?

    If you are not able to resolve this, please attach the TCS file then we may have to check next week in lab.

    Regards,

  • Hi Gordon,

    I just checked again with the following test procedure, please check if you can also make it.

    TICS Pro - Default configuration

    1. Vcc power up the board

    2. Provide a valid 100MHz reference clock

    3. TICS Pro - USB communications --> Write All Registers (you should get it locked at 1500MHz)

    4. Change R divider to 2 (fpd becomes 50MHz)

    5. Change VCO frequency to 6400MHz

    6. Change Channel divider to 16 (RFoutA becomes 400MHz)

    7. Click the CalibrateVCO button (The button will turn white after clicking it)

    8. Check the phases between the 100MHz reference clock (use it as the trigger channel in the scope) and the 400MHz output clock (they should be phase aligned with some deterministic delay)

    9. check the VCO_PHASE_SYNC_EN box to enable phase sync.

    10. The delay in step 8 above should be getting smaller. 

    11. check the MASH_SEED_EN box to enable phase adjustment

    12. change the MASH_SEED value to 800 and then press enter (you should see the phase of the 400MHz clock shifted to the left)

    13. If the cursor remain flashing in the MASH_SEED box, press enter again to shift the phase one more time

  • Hi Liam / Noel.

    When I follow the procedure above I do see the phase shifts I was looking for. I must have missed a critical step somewhere, but following the steps in order gives me what I was looking for, and I can change the value of MASH_SEED to give bigger or smaller phase shifts as necessary.

    Thank you both for your help with this - appreciated.

    Gordon.