TECHNET Archives

July 2003

TechNet@IPC.ORG

Options: Use Monospaced Font
Show HTML Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Happy Holden <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Wed, 9 Jul 2003 10:10:26 -0400
Content-Type:
multipart/mixed
Parts/Attachments:
text/plain (2100 bytes) , text/html (2580 bytes) , 10 Layer Stackup.xls (24 kB)
John,
There is a tutorial on Impedance Analysis on the Web at:
www.westwoodpcb.com   It includes Impedance calculators on the Web as well
as Field Solvers you can buy.  A comparison is made between the
calculators (that use the IPC equations), the Field Solvers, and the 'AS
BUILT' board.  The equations can be over 10% in error compared to the
Filed Solvers and 'AS BUILT'.  This is because of solder mask and
different Dk for cores and prepregs.
The stackup and line widths you talk about below are very common.  BUT,
what is the overall thickness of the board.  The other question is, do you
mean 100 ohm edge-coupled differential pairs (odd mode) and 50 ohm
single-ended traces or 100 ohm single ended traces.  Are these both
microstrip and stripline structures?  Solder Mask Coated edge-coupled
microstrips are not covered by equations, only the Field Solvers can do
these.  Included is a typical 10 layer 0.062" board that matches your
stackup and line widths.  Layer 10 has the upper range of copper plating
and solder mask thickness.  This lower the impedance quite a bit which is
why differential microstrips are not a good idea.

Many of the field solvers have a 10 day trial use period when you download
them.


Happy Holden
Westwood Associates


-----Original Message-----
From: John Parsons [mailto:[log in to unmask]]
Sent: Tuesday, July 08, 2003 10:37 PM
Subject: FAB: Impedance Calculations


Greetings all,

I could use a little help here.  I don't have an impedance solver at hand
and I have a customer looking for a suggested build for a 10 and 12 layer
board with signal traces running at 100ohm (single ended and
differential).
They will have signal traces at 4-6mil's (I am using 5mil for my rough
calculations) but I can't see how to do this with regular FR-4 laminate
(Er
4.2).  Has anyone build/designed boards with this requirement?  Can you
give
me some clues as to lay-up or laminate spec's that could save me some leg
work?

Stack is as follows; 1/2oz base copper all layers
10 layer, Sig,Plane,Sig,Plane,Sig,Sig,Plane,Sig,Plane,Sig
12 layer, Sig,Plane,Sig,Sig,Plane,Sig,Sig,Plane,Sig,Sig,Plane,Sig

Regards
John Parsons



ATOM RSS1 RSS2