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.

LMX2492: PLLatinum Sim and VSS Lock Time Discrepancy

Part Number: LMX2492

Dear all,

I have been evaluating a few different simulation environments for PLL systems lately. As we are using a Texas Instruments PLL, the first choice has been PLLatinum SIM, as it is already loaded with the PLL model. I have set a sample simulation as shown below to evaluate the Lock Time of this specific set-up.

As you can see the transient response peaks at 2 usec and stabilizes to 1 KHz around 20 usec. Later, I tried to replicate this set-up in VSS and ADS software so we could be able to make some changes and implement the PLL in a more complex system. Simulations are as shown below and the results are identical in both Environments, but the response is different from that shown in PLLatinum SIM by an order of magnitude!

Simulation in VSS:

Simulation in ADS:

Results in ADS and VSS:

What are we doing wrong? Is there some effect that PLLatinum is taking in consideration that is not considered by VSS or ADS?

Kind Regards,

Alejandro.

  • Alejandro,

    The difference you are seeing is due to cycle slipping of the PLL, which your PLLatinum Sim is not doing. 

    For default startup, PLLatinum Sim uses an analog model because it is much faster to calculate.  Also, it typically is pretty good if for integrated VCO products that get close in frequency before analog lock.

    For the LMX2492, this has external VCO and has more cycle slipping, which is more worth considering.  PLLatinum Sim does model cycle slipping, but you have to enable it.  Go to the lock time tab and click on "Show Discrete Model".  Then you can click the button to calculate discrete lock time.  It is sort of hidden because it can take longer to calculate discrete lock time.  There is a box for "Maximum Calculation Time" that prevents it from taking too long.

    Regards,

    Dean