• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 21 subscribers
  • Views 999 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.

LMK04821 - Are DC-coupled LVDS outputs allowed?

Daniel Bjorklund
Prodigy 30 points
Other Parts Discussed in Thread: LMK04821

The electrical characteristics of the LVDS outputs in the LMK04821 datasheet note "AC-coupled to receiver input" as a test condition. Additionally it mentions in a small footnote that LVDS needs a "560 ohm shunt". The EVM seems to use a 560 ohm resistor across the differential pair along with AC-coupling capacitors for all LVDS outputs.

Question 1: Is DC-coupled LVDS allowed? We would like to skip the AC-coupling capacitors (to lower the component count/space required) and directly connect the LVDS pairs to 100 ohm terminations in the receiver.

Question 2: If DC-coupling is allowed, will the differential swing and common-mode voltage match the numbers in the datasheet which are only given for AC-coupled output?

Question 3: If DC-coupling is allowed, can we then skip the 560 ohm resistor? Why is the 560 ohm shunt required? 

Thanks and best regards,

Daniel

  • 0
    TI__Mastermind 28560 points

    Daniel Bjorklund said:
    The EVM seems to use a 560 ohm resistor across the differential pair along with AC-coupling capacitors for all LVDS outputs.

    For the record, the 560 ohm shunt resistor for LVDS and HSDS outputs ideally should be placed as close as possible to the output of device, or right next to the load.  On the Rev C eval board, they are not in an optimum location and could be moved to where the emitter resistor location is.

    Daniel Bjorklund said:
    Question 1: Is DC-coupled LVDS allowed? We would like to skip the AC-coupling capacitors (to lower the component count/space required) and directly connect the LVDS pairs to 100 ohm terminations in the receiver.

    Yes.  This is allowed for LVDS.  For the record, HSDS type outputs have common mode voltages which may not be suitable for DC coupling but it is technically allowed to DC couple them too.

    Daniel Bjorklund said:
    Question 2: If DC-coupling is allowed, will the differential swing and common-mode voltage match the numbers in the datasheet which are only given for AC-coupled output?

    Yes.  Note these are made using DC measurement.

    Daniel Bjorklund said:
    Question 3: If DC-coupling is allowed, can we then skip the 560 ohm resistor? Why is the 560 ohm shunt required? 

    Without the shunt 560 ohm resistor it is possible in some AC coupled load termination configurations that the driver may not start operating or have a delay before normalized operation is achieved.  The DC path prevents this.  Of course if you DC couple your LVDS, there is no issue due to the 100 ohm termination resistor.

    73,

    Timothy