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.

LMH6722: Layout suggestion for fanout input (opa V+)

Part Number: LMH6722
Other Parts Discussed in Thread: LMH6702, OPA3692

I adopt LMH6722*2 to implement analog fanout buffer.

Refer to here:

https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1111959/lmh6722-fanout-analog-signal

Now I encounter a problem: how to layout buffer input?

Scheme as here

V+ of OPA Input is my problem.

I know that <90 deg angle is not good layout rult.

So I adopt Y-Shape(~135deg) to layout V+ of OPA (green rectangular)

Is this layout good?

Or others layout suggestion for 1:4?

Actually, I've think that maybe I need more buffer to do fanout.

1->2->4->8  but 15 OPAs are necessary.

I wish I could implement 1:8 fanout with LMH6722*2 and LMH6702*1.

  • Hi Dewei,

    I would locate the 50R series termination resistors close to the outputs of OPAmps. The copper tracks between the 50R resistor and the connector should be made from 50R microstrip lines. And the connectors should be mounted closest to the outputs of OPAmps. Sometimes this can mean that you do not take all the four OPAmps from the quad chip but only those which allow shortest connections.

    And I would avoid making this "tree" in the green rectangle. Just make simple and short connections. Short connections with 90° angles can be better than long and snake like connections with 45° angles.

    Kai

  • Excellent feedback Kai.

  • Many thanks @kai klaas69 !!

    I follow your suggestion to fix shape:

    As that picture, connector(J7~J9) seems to close OPA, longest path < 20mm (through 50 ohm resistor, OPA-> R18-> R25-> connector)

    But about the "tree" you mentioned,  I'm not sure my new layout whether correspond your suggestion.

  • Excellent feedback Kai.

    Thank you, Sam Relaxed

    Hi Dewei,

    As that picture, connector(J7~J9) seems to close OPA, longest path < 20mm (through 50 ohm resistor, OPA-> R18-> R25-> connector)

    Unfortunately, in this frequency range 20mm is not "close".

    Kai

  • Thanks @Kai's reply.

    OK!

    Maybe fixed gain and integrated product is a good choice for short trace, such as [OPA3692].

  • Dear Dewei,

    this is something which cannot be totally decided only by theory. You will need to fabricate a printed circuit board and carry out measurements. Then you will see how the board is behaving and whether the layout is already acceptable or needs some modifications.

    A good idea is to make two identical boards and make them differ in only the property in question. Then you can very clearly decide what version is better. Yes, I know this is time consuming, but that's how it is when designing a complex HF-circuit.

    Divide and conquer Relaxed

    Kai