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.

CDCM7005 can't set to "div by 1" output

Other Parts Discussed in Thread: CDCM7005

Hello:  my customer has designed in the CDCM7005 into a system, and is having trouble seetting the output to /1.  EAU is 50/yr, they will ultimately use the space grade version.  regards, George Biner, Arrow FAE Los Angeles

CDCM7005  application: 

  • 50 MHz reference input
  • 1.6 GHz VCXO
  • P divider = 8
  • N divider = 4
  • All outputs configured for LVPECL
  • Yn_MUX configured for divide by 1 (1.6 GHz output)

We are aware that we are configuring the CDCM7005 #2 above its maximum specified 1500 MHz output.  However, the datasheet specifically states in footnote (7) on page 8 that the outputs can be configured above 1500 MHz with degraded performance.  Figure 5 depicts differential output voltage swing up to 1950 MHz output.

We are configuring the CDCM7005 device #2 (1600MHz) with the following configuration data:

WORD-0: 0x4000.3000

WORD-1: 0x0000.007D

WORD-2: 0xD000.14F2

WORD-3: 0x0000.0027

( WORD-x: 0xMSB…..LSB    For example, in Word-0, the 0x4 of the MSB refers to bits [28..31] with values [0010].  )

CDCM7005 configured as indicated above, presents a 800 MHz clock at its outputs (we’re expecting 1600 MHz).  If we re-configure the Yn_MUX to any other divider, we get the expected output frequency.  E.g., if we configure Yn_MUX to select the divide by 4 output, we get 400 MHz.  If divide by 8 is selected, we get 200 MHz.  If divide by 2 is selected, we get 800 MHz.  But if divide by 1 is selected, we still get 800 MHz.  Note that in all cases the CDCM7005 indicates it is locked, and the measured VCXO input to the device is a compliant 1600 MHz LVPECL clock.

With a 1600 MHz VCXO input, the CDCM7005 seems unwilling to be configured for the divide by 1 output.  We’re stumped.  Any ideas?

  •  Hello George,

    sorry for the late reply. I read through the mail but have not seen this issue before. Div by 1 should work well, even when output frequency is slightly higher than 1.5GHz. Your setting looks fine as well. We need to look into this again together with design. But in meanwhile, can you share more information about programming sequence ans are more units affected.

     

    Thanks and regards,

    Georg Becke <TI>