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.

LMX2592: calculation parameter for RFoutA with respect to TicsPro GUI

Part Number: LMX2592
Other Parts Discussed in Thread: TIDA-00626, PLLATINUMSIM-SW

respected,

i have been working on the reference design of TI i.e. TIDA-00626 which incorporates LMX2592 IC. i want a proper mathematical calculation that how do we achieve RFoutA in TICSPRO GUI and what is the equation to manually calculate the rfoutA by changing other parameter. the reference is 100 MHZ and suppose RFoutA is 2.5 GHz what is the internal calculation that makes us achieve 2.5 GHz please help me with mathematics of this as i am highly interested in the same. Please provide proper reference or calculation regarding this

  • Pratik,

    It would go like this:

    RFoutA = 2500 MHz
    Fvco = 2x2500 MHz = 5 GHz
    (you need to choose the output divider so that the VCO is a multiple of 2500 MHz and in the VCO range).

    The phase detetor choice has many options, it has to be either 2X the OSCin frequency, or divide the OSCin frequency. 2X OSCin frequecy is 200 MHz, so I would choose this as higher phase detector frequency is for better phase noise.

    Regards,
    Dean
  • respected,

     ia appreciate the answer provided by you kindly clear some of my concerns.

    1. in TICS PRO GUI what does charge pump gain does what is the affect on output while varying the charge pump gain. to test the output i only have spectrum analyzer and dont have phase nosie analyzer.
    2. what do channel divide MUX does when we select seg 1 and 2. what is the significance of each segment.
    3. what is the effect on changing output MUX to VCO rather than CHDIV.
    4. what we can achieve with VCO calibration block in TICS PRo gui
    5. what is the importance of MASH and MASH_dither

    kindly clear my doubs as soon as possible as i need to prepare everything about the working of the PLL. sorry for the inconvenience caused

  • Pratik,
    1. Charge pump gain is a critical parameter to the loop filter bandwidth. The bandwidth impacts the spurs and phase noise. If you choose a narrow span, you might be able to see the loop bandwidth. If you change the charge pump gain, it impacts this bandwidth.

    2. The channel divider is broken into 3 segments for optimal performance and to minimize current consumption. For smaller divide values, the channel divider may not need all 3 segments, so it powers the unused ones down. So when a segment is powered down, you do not want to select this. The TICSPro gui shows how this works; if you enter in a frequency in the output, it configures the channel dividers and the channel divider mux.

    3. If you choose the output divider to VCO, then the VCO is routed to the output, regardless of the state of the CHDIV mux or the states of SEG1, SEG2, and SEG3. When you want a frequency of 3350 MHz or above, you don't want to use the channel divider, so you set this mux to VCO.

    4. VCO calibration is done whenever you program register R0 with FCAL_EN=1 (or you can just load all registers). There are some settings on the User Controls page that allow one to improve the VCO calibration time, but this requires some experimentation and can degrade phase noise.

    5. MASH and MASH_DITHER impact fractional spurs and phase noise. These are modeled in the PLLatinum Sim software at: ti.com/tool/PLLATINUMSIM-SW and also discussed in great depth in our online PLL reference as ti.com/tool/PLL_BOOK.

    Regards,
    Dean