TECHNET Archives

March 1998

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text 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:
John Waite <[log in to unmask]>
Reply To:
John Waite <[log in to unmask]>
Date:
Mon, 30 Mar 1998 15:01:00 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (101 lines)
Hi Joe,
    Life would be nice if we could use a constant of 4.4 or 4.5, but
unfortunately, many factors effect the impedance of a circuit line,
including variation in the Er.  Depending upon the material type, (FR4,
"Gtek", Teflon, etc) a general Er can be derived from the resin/glass ratio.
The following numbers will be "generic" in sense and pertain to FR4
laminates.   4.4 is representative of a 2116 style prepreg material at
approx.. 54% resin, 4.2 is representative of a 1080 style material at
approx.. 66% resin, 4.6 is representative of 7627 material at approx.. 42%
resin content.  Other variables in the prepreg material can include,
moisture level, design of circuit traces and copper weights, lamination
techniques, etc.
    It almost sounds like you have been given variation numbers for the
innerlayer and outerlayer to "fudge" certain template designs and have
worked? for you.  As your product performance increases, all the "little"
things will create a major impact.
    Don't let these variations scare you too much.  If you work with your
board vendor, They should be able to guide you through the "mechanics" of
the lay-up and give you the necessary numbers.  Insure that you use the
proper configuration for microstrip, embedded microstrip, stripline, and
dual stripline.  If the board has a differential impedance, there will be a
different set of "magical numbers" necessary.
   If you're dealing with high speed signals, your Operating level of the
material will change in regards to the Er.  If this is the case, evaluate
some high performance materials.
    Your choice of material becomes more critical as you deal with lower
dielectric/higher layer count product  (not as much dielectric to act as a
"buffer" for material variation.
    Line widths and copper thickness will play a major part, insure that
these are properly defined. The type of soldermask and thickness will have
some impact.  (typically, not MAJOR).
    I apologize if I'm starting to write a book, but impedance has been
viewed by some as a "black magic", by some as "rule of thumb", and by others
as overly difficult.  If you work with your vendor and outline some ground
rules, you should be ok.  Insure that your vendor has reliable software,
test equipment (calibrated!), and someone that has a decent understanding of
Impedance.    IPC has a tremendous amount of information on Impedance and
some general guidelines (IPC std 275?).  If You Email me with your Tel
number, I can discuss further and hopefully help clear up some questions
(that's if I haven't confused the ---- out of ya already.  Good luck JOHN
WAITE
    It
-----Original Message-----
From: Hurst, Joe <[log in to unmask]>
To: [log in to unmask] <[log in to unmask]>
Date: Monday, March 30, 1998 2:21 PM
Subject: [TN] FR4 Relative Permittivity


>Hello Folks:
>
>
>Since I often design PWBs having controlled impedances, the question
>often arises as to what value to use for the relative permittivity (Er)
>of FR4 material.
>
>1. Some consultants have said to use 4.8 for outside layers and 3.5 for
>inside layers.
>Impedance calculations based on these numbers give results closely
>matching actual measurements.
>
>2. Another source has said that due to the pressing process the Er
>around the traces on inside layers actually is 0.5 less than the Er of
>the FR4 material being used.
>
>3. Others insist that an Er of 4.5 be used for all calculations.
>
>Considering that these materials and processes have been around so long
>the entire issue must have been settled by someone.    What are other
>people using in these calculations?
>
> Thank You
>Joe Hurst
>
>################################################################
>TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
>################################################################
>To subscribe/unsubscribe, send a message to [log in to unmask] with following
text in the body:
>To subscribe:   SUBSCRIBE TechNet <your full name>
>To unsubscribe:   SIGNOFF TechNet
>################################################################
>Please visit IPC web site (http://jefry.ipc.org/forum.htm) for additional
information.
>For the technical support contact Dmitriy Sklyar at [log in to unmask] or
847-509-9700 ext.311
>################################################################
>

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC web site (http://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################


ATOM RSS1 RSS2