• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 33 subscribers
  • Views 271 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.

LMH0046 not lowering the jitter enough

Wm Vancura
Prodigy 40 points
Other Parts Discussed in Thread: LMH0030, LMH0046, LMH0002

I have an LMH0046 being driven by a LMH0030 direct coupled with two 75 ohm pullup resistor on the inputs. The output of the LMH0046 is direct coupled to an LMH0002 with a 100 ohm resistor between the inputs. The 27Mhz is supplied by a 20ppm oscillator. The SDHD output is connected to the LMH002. LF1& LF2 have 56nf cap. lock tied to Mute. Other than power and ground, all other connections are open. The measured input (1080i) to the LMH0046 UI is 0.45, and the output is 0.345UI. This is much too high. What is being done wrong?

Bill

  • 0
    TI__Expert 4400 points

    I apologize for the delayed response.  This post would be better seen in the High Speed Interface Forum so I have moved it there.

    The LMH0046 reclocker is essentially a low-pass filter system that passes jitter at frequencies below the loop bandwidth and attenuates jitter for frequencies above the loop bandwidth.  The LMH0046 loop bandwidth is 2 MHz for 1.485 Gbps operation.  I would suspect that significant components of this jitter you are attempting to remove are at frequencies < 2 MHz, which would explain why the reclocker is not filtering as much of this jitter as you are expecting.

    I think the real problem stems from the high LMH0030 output jitter you are observing.  The LMH0030 output jitter should be < 0.2 UI as long as its input VCLK jitter is ~30 psP-P as recommended in the datasheet.  If the LMH0030 input VCLK jitter is too high, this jitter will be multiplied and passed on by the serializer.  The LMH0030 output jitter may contain low frequency components that are below the reclocker's loop bandwidth and are then passed on by the reclocker.

    Please check the jitter on the input VCLK to the LMH0030 serializer, and if it is much greater than 30 psP-P, then the best solution for this application would be to either use a lower-jitter clock source for the serializer or to clean up this VCLK jitter by using an external jitter cleaner.  Then the reclocker would likely not be needed following the serializer in this application.

    Gary Melchior