RE: [SI-LIST] : RE: 2.5 GHz in FR4

About this list Date view Thread view Subject view Author view

From: Zabinski, Patrick J. ([email protected])
Date: Tue Apr 03 2001 - 05:08:21 PDT


I would recommend 10 mil or wider traces for the 2.5 GHz signals.
To maintain roughly 50 ohm impedance, you will need to
thicken the upper/lower dielectric layers.

If the 622 MHz signals are LVDS or some other form of differential
protocol (LVDS seems to be common these days in SONET/ATM
components), then I suggest using stripline to get them
out of your way (my guess is that the outer microstrip layers
will be congested; so put what you can in stripline).

The 155 MHz stuff is low enough that good SI-practices should
allow you to put them in stripline regardless if they are
differential or single ended.

In terms of reference planes, having one set of signals reference
one plane and another set of signals reference another plane
should be okay, but I *strongly* recommend that:

        * the reference plane for a set of signals is the same
          supply used by the driving device's output!
        * in particular, make sure the supply is what the signals
          are being internally referenced to. For example, ECL
          signals should be referenced to VTT (-2.0V), *not*
          VEE (-5.2V). Similar, CML should be referenced to VCC
          (positive rail) not VEE (negative rail).
        * for signals that evenly reference two supplies (e.g., full
          swing CMOS), reference one plane and add plenty of
          decoupling at the driver.


> >* Use wide traces. FR4 is lossy at 2.5 GHz, so try to compensate
> >for the high dielectric loss by decreasing the metal loss. Use
> >wider traces which will reduce the resistive losses in the metal.
> the general rule i am planning to give is 5 mils track with 5
> mil spacing.
> i have few 155mhz / 622mhz signals also and due to constaints of
> BGA and microstrip, i was planning to put them also in microstrip with
> 5 mil track and 5 mil spacing.
> how much more do i need to increase? is 8 - 10 mils ok?
> as with different trace width the impedance varies, in order to get a
> uniform
> impedance across the board, i may need to make two power
> planes, one which
> references the 155 mhz signals (5 mils) and one which
> references the 2.5 ghz
> signals. will this be ok? or is it better to put the 155 mhz
> signals onto
> strip lines and use 2.5 ghz as microstrip and manipulate the planes?

**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected] In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at

About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 21 2001 - 10:11:24 PDT