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

About this list Date view Thread view Subject view Author view

From: Larry Miller (ldmiller@rhapsodynetworks.com)
Date: Wed Apr 04 2001 - 14:02:02 PDT


Mitch,

I haven't had personal experience with broadside-coupled differential pairs.
I do know a couple of people who got burned with them, but not why they got
into problems.

The fab houses have tended to discourage broadside because of the number of
manufacturing variables they have to control. On the edge-coupled boards we
did the vendors were able to hold the impedance within 1 ohm over hundreds
of paths and a number of boards, which I thought was fabulous.

I think there was a DesignCon 2000 paper given on differential pairs that
concluded that if you matched the parameters correctly for each case there
was actually no measurable difference. One of the diff pair gurus on the
reflector will no doubt have pointers and/or opinions.

One thing that might be a factor: the common connectors are all edge-coupled
in design, so I would guess that theoretically in a broadside-coupled PCB
system you would have to change propagation modes going through the
connectors. Does this make a difference? I do not know. I seem to recollect
from 30 years ago that in rectangular wave guide E-plane bends were worse
than H-plane bends (or maybe vice-versa).

I used edge-coupled because AMP recommended it for their HS-3 connectors and
gave very detailed specifications for routing through pin fields, etc. This
made it easier to communicate with the PCB designer ("do it like the
picture").

The use of stripline was because of the number of channels I had to do. It
turned out that I could fully interconnect the diff pair part of the board
with 4 layers of stripline pairs and no vias. That would not have been
possible using only the surface layers. (I don't think I said microstrip
anywhere, but if I did I meant stripline. Very dyslexic morning.)

I was lucky in one respect: I did the back plane before the plug-in cards
were designed, so I had "dibs" on pinouts that made the back plane route
with no crosses. Actually, it turned out that this was good for the plug-in
part of the system as well. It certainly showed the advantage

Cheers,

Larry Miller

-----Original Message-----
From: cadpro2k@dacafe.com [mailto:cadpro2k@dacafe.com]
Sent: Wednesday, April 04, 2001 12:53 PM
To: Larry Miller; Larry Miller; 'si-list@silab.eng.sun.com'
Subject: RE: [SI-LIST] : RE: 2.5 GHz in FR4

---------Included Message----------
> From: "Larry Miller" <ldmiller@rhapsodynetworks.com>
> Subject: RE: [SI-LIST] : RE: 2.5 GHz in FR4
>
> Sorry, I meant EDGE-coupled.
>
> Larry Miller
>
> -----Original Message-----
> From: Larry Miller [mailto:ldmiller@rhapsodynetworks.com]
> Subject: [SI-LIST] : RE: 2.5 GHz in FR4
> Broadside coupled, keep goodly distance apart for crosstalk.

> That is about all that my mgmt is comfortable with....
>
Hi Larry,

Are you saying that anything around and above the 2.5Ghz range you do
edge-coupled? Do you also ONLY do it microstrip? What experience(s) have
you had with broad-sided stripline?

Appreciate your comments.

Mitch
_____________________________________________________________
Tired of limited space on Yahoo and Hotmail?
Free 100 Meg email account available at http://www.dacafe.com

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. 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 http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 21 2001 - 10:11:27 PDT