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.

LMX2486: LMX2486 PLL RF_FN Question

Part Number: LMX2486

Hello,

My customer has a question regarding registers functionality in the LMX2486 PLL. According to the data sheet the R0 register has 2 fields: RF_N and RF_FN. If I decided to use RF_FN in 22-bit mode what would happened with RF_N field, would it be ignored?

Thanks,

Chuchen

  • Chucken,

    No, the RF_N is always considered.

    In 12 bit mode, the MSB bits of RF_FN  (RF_FN[21:12]) are ignored.  in 22 bit mode, they are considered.

    The complete N value is:

    N divider = RF_N   +   RF_FN/RF_FD

    Regards,

    Dean

  • Hello Dean,

    We have only one 24-bit R0 register, which in 12-bit mode is divided as shown in Registers Map, Table 14: RF_N between bits 13 and 23 (11 bits) and RF_FN between bits 1 and 12 (12 bits).

    In the 22-bit mode RF_FN will cover space from bit 1 to bit 22 which will overlap space for RF_N Counter?

    Thanks, Steven.   

  • Hello Dean,

    I still didn't get your answer to my question. Will repeat again:

    We have only one 24-bit R0 register, which in 12-bit mode is divided as shown in Registers Map, Table 14 where: RF_N between bits 13 and 23 (11 bits) and RF_FN between bits 1 and 12 (12 bits).

    In the 22-bit mode RF_FN will cover space from bit 1 to bit 22 which will overlap space for RF_N Counter.

    What is wrong with my interpretation of LMX2486 data sheet?

    Thanks,

    Steven.

  • Dear Steven,

    excellent question.

    the numerator extended fraction portion does not extend into R0 but is located in R5. Check table 13 on page 26.

    Regards, Simon.

  • Thank you Simon,

    To make sure I correctly understood:

    1. RO does not extend RF_FN to 22 bits (data sheet error).
    2. If we want to have extended, 22-bits RF_FN we need to use R5.

    What if we need to use extended both RF_FN and RF_FD, R5 can have only one of them to be extended?
    And how we differentiate which of RF_FN and RF_FD was extended?

    Regards,
    Steven.
  • Dear Steven,

    no problem, I see how it can be misleading in the documentation.

    In register R3 position [23:20] there is a field called access[3:0]. depending on access[1] (R3[21] and the 2nd bit of access from the LSB), the fraction will be a 12 or 22 bit quantity:

    if access[1]=0,

    fraction will be 12 bit
    The numerator will be the field name RF_FN[11:0] in the register R0[12:1].
    The denominator will be the field name RF_FD[11:0] in the register R1[15:4]

    if access[1]=1,

    The numerator will require the concatenation of two fields in two different registers. I use the {} to indicate concatenation.

    {RF_FN[21:12] , RF_FN[11:0]} (22 bit long)

    now expressed as register name and position it is:

    {R5[13:4] , R0[12:1]}

    The denominator will require the concatenation of two fields in two different registers.

    {RF_FD[21:12] , RF_FD[11:0]} (22 bit long)

    now expressed as register name and position it is:

    {R5[23:14] , R1[15:4]}

    Regards, Simon.
  • Thank you very much Simon for so nicely, detailed explanation.
    Now I don't have any questions.
    If data sheet at least added one more sentence, that extension of register is result of concatenation with the fields of other registers,
    it would save me much time. But the best way would be to give example as you did.

    Thank you so much,
    Steven.