• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 568 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LMX2594EVM: LMX2594EVM output spectrum spur at 6.7KHz

Jijun Yao
Prodigy 10 points
Part Number: LMX2594EVM
Other Parts Discussed in Thread: LMX2594,

Dear Sir/Madam:

we are currently evaluation the LMX2594 using the TI evaluation board LMX2594EVM with the TI's reference Pro board; we connected the boards as shown in the instruction as figure 2 of the LMX2594EVM document, powered on and loaded the default setting; it locked and output at default 7GHz (divider output, the ocs output is 14GHz);

the phase noise measured is in according with the data provided; but when zoom in to narrow spectrum, we noticed the output signal contains spurious in multiple of about 6.7KHz spacing, the level is about -30dBc to -50dBc; after comparing with the equipment signal source output spectrum, it is confirmed the spurious of the LMX2594;

changing settings to the chip (eg. fpd change from 200MHz to 150MHz or other values, change output frequencies, change calibration  and even into detailed register settings), locking is no problem, but the close band spurious remains.

Please let us know, do you have a solution for 6.7KHz spurious; or is possible, could you provide a full register value file that we can load and test if it is our setting problem?

BTW, we tried external 100MHz reference provided by signal source, only the spurious spacing changed to very close 160Hz, but it would not solve the inband spurious;

thanks

  • 0
    TI__Mastermind 28021 points

    Jinjun,

    It would be good to have more solid confirmation that this spurious is indeed generated from the LMX2594, not the setup, power supply, or input reference.  We have not seen this in our setup, and spurs in the -30 to -50 range are hard to miss.  For the LMX2594, it would be good to eliminate the possibility that this is a fractional spur.  To do this, set the part to an integer channel and set MASH_ORDER to integer mode.  If the spurs go away, then this suggest that indeed these spurs could be inherently generated by the LMX2594 and something can be done with the fractional settings in the device.  But if the spurs stay the same, it suggests something about the setup and the solution will be changing something external to the device.

    Here are some external setup factors that could influence the spurs.

    1.Input reference:

    You mention that if you change the input reference, you see 120 Hz spurs appear, but it was unclear if the 6.7 kHz spurs went away.  If so, then this suggests that the input reference has 6.7 kHz spurs.  Any spurs on the input reference are multiplied up. For instance, if the input is 100 Mhz and the output is 10 GHz, then the spurs are magnified at the output by 20*log(10GHz/100MHz) = 40 dB.

    2.Power Supply

    Any spurious on the power supply can go to the output.  The LMX2594 has a pull-up component directly to Vcc.  So if the supply for Vcc you give has spurious content (say at 6.7 kHz), then this will modulate teh carrier and this is happening completely external to the LMX2594.  The EVM is sensitive to power supply noise.

    3. Inherent to measurement equipment

    Some equipment shows spurs that are not real.  If nothing seems to make the spur go away, then consider putting a signal generator into the spectrum analyzer at the same frequency as the LMX2594.  In other words, completely take the LMX2594 out of the setup condition.  If the spurs are still there, then I would question your spectrum analyzer.  That being said, with spurs of -30 to -50 dBc, these are pretty high, so I don't suspect that this is the cause of your spurs.

    Regards,
    Dean