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.

LMX2531: latency between reset and lock to previously saved frequency setting

Part Number: LMX2531

Hi guys, 

You already helped me out on this thread, I have a follow up question.

The LMX2531 saves its previous settings through a hard reset of the CE pin, can you guys give any information about the latency between the moment CE is de-asserted, and the PLL locks to it's previous frequency setting?

Thanks,

Brian 

  • Dear Brian, I will have someone get back to you asap. Regards, Simon.
  • Brian,

    For the LMX2531, when you bring the CE pin low, all the programmable register information is retained. However, there is information in the state machine for the VCO capcode and amplitude settings that are lost. In other words, when you bring the CE from low to high, you need to program the R0 register again to re-run the calibration, which takes on the order of 500 us, depending on the input frequency and some other factors.

    For our newer devices after the LMX2531, they do retain the state macine information for VCO calibration, but not the LMX2531.
    If you look under the description under the CE pin in the datasheet, it says this:

    "
    Chip Enable Input. High impedance CMOS input. This pin must not exceed 2.75 V. When
    CE 11 Input CE is brought high the LMX2531 is powered up corresponding to the internal power control bits. Although the part can be programmed when powered down, it is still necessary to
    reprogram the R0 register to get the part to re-lock."

    Regards,
    Dean
  • Ok, so the R0 register must be reprogrammed *after* the CE pin is de-asserted, and then from the time R0 is reprogrammed to the time the PLL locks, it is about ~500us.

    Final question, I received guidance on another post that we should wait about 100us after de-asserting the CE pin before programming registers. Does that sound right?

    Thanks,

    Brian

  • Brian,

    After you pull CE low, I don't see why you would need to wait to program the registers, but it woudn't hurt to wait the 100 us.

    However, after you pull CE high, it does make sense to me to wait before programming the R0 register to ensure all the capacitors get to their proper bias levels.

    Regards,
    Dean
  • Dean,

    Your comments have confused me a bit.

    On the LMX2531, once you pull CE pin low, R0 loses its settings and must be re-programmed, but you can actually do the reprogramming while CE is still low?

  • Brian,

    Sort of. You can program the device while it is powered down and all the programmable register information is retained. However, this is not all the information needed for the VCO. The VCO also has internal state settings (not disclosed in the datasheet) like the frequency band and amplitude setting. When you power the device down, these internal state settings are lost.

    The only way to get the internal states settings for the VCO back is to run the VCO calibration, which is done by programming the regsiter R0, and the VCO calibration does not run when the device is powered down. Therefore, although you can change all the registers (to potentially a different frequency setting) you must program the register R0 after the chip is powered up so theVCO calibration can be properly run.

    From an applications standpoint, if I am locked to a frequency and pull CE low, then I would need to pull CE high and after that program the regsiter R0.

    What you could also do is power down the device with CE, completely change the settings and frequency, pull CE high, and then run R0 and then you could go to a new frequency.

    Regards,
    Dean