• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 21 subscribers
  • Views 786 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.

LMX2594: Spurs technical quesiton

YANG LI
Prodigy 10 points

Part Number: LMX2594
Other Parts Discussed in Thread: TPS5430

Hi, Dear TI Sir, 
I need your help about the design of LMX2594.There is a technical question  about  its design.

In our design, there are dual ~15KHz offset spurs besides of Main output peak. Such as following.  Eg. 5GHz peak.

Which factors and reasons are contribute to this?  Very Look forward your arrange and technical reply. Thanks a lot.
Attached files is 5G-15kHzspur and 8G-12kHzspur.

  • 0
    TI__Guru** 122550 points

    Hi Yang,

    Could you provide your synthesizer configuration, preferably in TICS Pro format like below?

    TICS Pro is available to download in ti.com.

    https://www.ti.com/tool/TICSPRO-SW

  • 0 in reply to Noel Fung
    Prodigy 10 points

    Hi，Dear Noel,

    Our config is attached. Oscillation source is 10MHz input. The output default is 1GHz;  When output is 5GHz; the configura as following. As same as the design of a demo brd that we purchased as a reference. The demo has not this spure. But our design have this spure. Our spure gueesed by hardware, but we have not dector which factors  contribute this. Hope you help.

    5GHz-configure:

    1GHz-Default:

  • 0 in reply to YANG LI
    TI__Guru** 122550 points

    Hi Yang,

    You have set the multiplier to x5, this is not a valid configuration. You should set the Doubler to x2 so that your highest possible fpd is 20MHz. You should also set the FCAL_HPFD_ADJ according to fpd, otherwise you might have VCO calibration issue.

    The configuration is an integer channel, so there should be no spurs at 15kHz offset.

    This 15kHz spurs could be coming from your system, for example, are you using DC/DC converter with switching noise at 15kHz?

    What is the format of the 10MHz input clock? Since the frequency is very low, suggest use square wave signal to get a better PLL noise performance.

  • 0 in reply to Noel Fung
    Prodigy 10 points

    Hi, Noel, 

    Thanks for your point out the spot. Although The demo brd and our design brd is work. It should not work at optimal condition as x5 configu. Is this relate with spure issue?

    I checked tha DC/DC(TPS5430) freq on brd is 500KHz. But the number is not correlation. So, it may not directly relationship between this spure to DC/DC on brd. But maybe environment noise or CE emmision into our brd.  The 10MHz Clock input is LV-CMOS that supply is 3.3V.

    If it is caused environment noise, it is hard to elimiate the noise; As U know, there are noise source, crosstalk paths, victim. So, which parts of LMK2594 device are sensitive, and may cause this issue?  And how  to protect this sensitive parts?

    Hope your valable suggestions and guidance~ Thanks a lot
     

  • 0 in reply to YANG LI
    TI__Guru** 122550 points

    Hi Yang,

    Wrong Multiplier setting would not cause this issue, I think you have verified this in the EVM (demo board). 

    Your spurs signal is very strong, it should be coming somewhere in your board. To debug, you could try to use external power supply, use external reference clock. It could also be a coupling issue from pcb layout. 

    Since your spurs frequency follows VCO frequency, I suspect the spurs is coming from the reference clock. Try replace the reference clock first.