• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 21 subscribers
  • Views 1542 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.

LMK03328: 1.8V LVCMOS output interface - Load drive capability and termination

Kawai
Guru 19785 points
Part Number: LMK03328
Other Parts Discussed in Thread: LMK00105

Hi Team,

This is my question divided from former post.
1.8V LVCMOS connection between LMK03328 output and LMK00105 input.

https://e2e.ti.com/support/clocks/f/48/t/664966

Below is the characteristics for 1.8V LVCMOS output for LMK03328.

Q1). Do you only guarantee the device operation only with source 50 ohms termination ?

Electrical characteristic specifies VOH(MIN)=1.35V with the condition of VIH=-1mA.
I believe that when there is load over 1mA, VOH (MIN) would decrease.

Q2). In the above former post, I had the answer that 50 ohms termination could be applied if there is DC blocking capacitor. Is this true ?

Even if there is a DC blocking capacitor, there should be AC current flowing from the LMK03328 output to 50 ohms termination.

Q3). In the application where they have long trace, for example, connection between LMK03328 output and LMK00105 input is 0.5m.
How should the user use the device ? What would be the recommended schematic ?

- Since the tr/tf is so fast for LVCMOS interface, end termination may be required when there is a long trace more than 0.075m.
- There is a parasitic capacitance in the LVCMOS input circuit and this capacitance could introduce dip when the voltage is rising.

Best Regards,

Kawai

  • 0
    TI__Mastermind 32575 points

    A1) The output voltage spec is given for the test conditions (Iout=1mA) shown in the table, which is NOT for 50 ohm load to GND.

    A2) Even with DC blocking cap, there could be output swing attenuation due to the voltage divider formed by the driver's internal 50 ohm output resistance and the 50 ohm load termination.

    A3) Per the previous post,  I recommend Figure 7 with the following values:

    * Rs = 0 ohms

    * Do NOT install 50 ohms load termination (voltage divider) since the 1.8 V output swing is compatible with the maximum single-ended input swing of 2 V.

    * Rb1 = 10 k (pull-up to 3.3 V) and Rb2 = 3.9 k to set CLKIN- to ~0.9 V.

    The LMK03328 has 50 ohms internal output resistance (source termination) to provide the required impedance matching for a 50 ohm trace.  No need to add an extra 50 ohm load to ground (avoid double termination).

    Alan

  • 0 in reply to Alan O
    Guru 19785 points
    Hi Alan-san,

    Thanks for continuous support.

    For A2), my thought is that the device cannot drive 50 ohms termination since IOH is 21mA.
    1.8V / 0.021A = 85.7 ohms
    The load must be higher than this resistance.

    I understand that input voltage at LMK00105 side would be half due to source and end termination.


    For A3), I will recommend customer to use the way described above.
    Since, they are observing reflection in their system, first I will ask them to try similar set up.

    Best Regards,
    Kawai