• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 729 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LMX2581E: Time Taken to change frequency

H Northey
Prodigy 190 points
Part Number: LMX2581E
Other Parts Discussed in Thread: LMX2581, , CODELOADER, USB2ANY, LMX2592, LMX2594, LMX2572, LMX2582

Hello,
I have a question about the time it takes to change frequency on the LMX2581E.   This is a similar concept to the lock time but slightly different as I will explain below.
I have been measuring this by triggering off the Latch Enable signal and measuring the time for Lock Detect to go high.  As I measure this I view the clock, Latch Enable, and Lock Detect on the oscilloscope. This works well for cases where only one register needs to be changed but not for when two or more registers need changed.

Device;LMX2581
Evaluation board: LMX2581EVM in original configuration with internal VCO. 
Software; Codeloader program version 4.20.02, Part file version 2014-11-09
Reference; 47.5MHz
Windows 7 SP1 64 bit
Connected with USB2ANY

I notice for some frequency changes that two registers need to be programmed.  For example if I change PLL_NUM even just by one number, this changes registers 0 and 1 as there are bits of PLL_NUM in both registers.  When viewing the CLK signal for these changes, I notice the delay between the programming of the two registers to be about 20ms.  This blows out the time taken for the frequency to change to above 20ms.   

In cases where only one register needs to be programmed, the time between programming and actual frequency shift is closer to 150uS as mentioned in the datasheet p7 under VCO Calibration Time. 

I set up a sequence in TICs Pro burst mode to program the registers 0 and 1 straight after one another with no delay in between and I still got the same delay of around 20ms when I looked at the clock signal on the oscilloscope. 

Is there any way to reduce this time between programming registers?

Many thanks
Helen

  • 0
    TI__Expert 5370 points

    Dear Helen,

    thank you for your ineterst with LMX2581E. I would like to bring your attention to other great devices in our portfolio. LMX2582, LMX2592 and LMX2594. FInally, a device that will be released this month (and already visible on TI.com) is LMX2572. All of them are excellent synthesizer while LMX2572 will have excellent phase noise with very impressive small power comsumption. Max frequency will be 6.4 GHz. The calibration time will also be quite fast.

    The problem you describe with the register change seem to be more with the USB2ANY (gray box) that comes with the evaluation kit. The speed of this SPI interface is not very fast and offered for evaluation purposes only. I suspect it could be improve as the base for this interface is a MSP430. We expect the final implementation will have the SPI interface with the correct speed. For high speed test we use a data generator so we can control the speed very well and even add trigger signal at will for verification

    Hope this helps,

    regards, Simon.

  • 0 in reply to Simon Damphousse
    Prodigy 190 points

    Thanks Simon, that helps,

    Best Regards,

    Helen