SUBJECT too long. Original SUBJECT is
[SI-LIST] : Re: Embedded microstrip calculations, Ultracad Calculator
---------------------- Original Message Follows ----------------------
Text item:
Doug,
First, I want to apologize for the "public criticism". For one, I didn't
mean
to criticise. My intention was to alert the innocent user, because I am
still
not convinced that the results of your impedance calculator is correct.
I am not a numerics guy, so I cannot comment on the equation itself. Your
explanation seems to be right, but something seems to be wrong with the
equation. If you check your results, you will understand what I mean.
microstrip in air embedded microstrip
H = 5.0 mil H = 5.0 mil
T = 1.4 mil T = 1.4 mil
W = 10.0 mil W = 10.0 mil
Er = 4.7 Er = 4.7
H1 = 11.4 mil
Z = 40.7275095 Ohms < Z = 41.1925677 Ohms
If I change H1 to a large number, such as 10,000,000.00 the impedance
results
will both be 40.7275095 Ohms.
This implies that a microstrip with air cover layer has the same impedance
as
the same microstrip conductor with an infinitely high dielectric cover,
which is
simply NOT true. Something is clearly wrong in the calculator and this is
all I
wanted to report in my first EMAIL.
Again, I want to apologize for the public response, but I feel that the
users of
this program need to know about this. Please do not take it personally.
Sincerely,
Arpad Muranyi
Intel Corporation
============================================================================
====
I apologize, but I should have carried this argument one step
further .........................
To repeat:
I quote from IPC-D-317, Design Guidelines for Electronic Packaging
Utilizing High-Speed Techniques, p 22
5.5.2 Embedded Microstrip Line .... The equations for embedded
microstriplines are the same as in the section on (uncoated) microstrip,
with a modified effective permittivity..... the effective permittivity
can be determined as in sction 5.2
Section 5.2 (equation 5.17 on p 17) gives this relationship as
E'r = Er[1 - exp(-1.55H1/H) ]
if H1 becomes infinite, the exp term goes to zero and E'r becomes Er
Therefore, according to this reference, which I relied on for the
calculator, the results ARE THE SAME for microstrip and embedded microstrip
if the thickness of the coating is very thick.
THIS REFERENCE AND FORMULA ARE FULLY DISCLOSED IN THE HELP FILE.
To Continue -------------------------
Now, if
E'r = Er[1 - exp(-1.55H1/H)]
if H1 > H then the exp term is <1
Therefore E'r < Er (which makes sense, because it will be between
the Er of the material below and Er of air, which is 1. So, it will
be between 1 < E'r < Er
Now Zo is an inverse function of the square root of Er
So if E'r goes down, Zo will go up (not down as Arpad alleges)
I rely, for my source, on the referenced IPC manual
I believe you all will find our little calclator (AND its Help
file where all this is disclosed and referenced) a useful
addition to your tool set.
Doug Brooks
President
UltraCAD Design, Inc.
www.ultracad.com
Text item: External Message Header
The following mail header is for administrative use
and may be ignored unless there are problems.
***IF THERE ARE PROBLEMS SAVE THESE HEADERS***.
Cc: Arpad Muranyi <[email protected]>
Subject: Embedded microstrip calculations, Ultracad Calculator
From: Doug Brooks <[email protected]>
To: [email protected]
Date: Thu, 11 Dec 1997 09:27:52 -0500
Content-Type: text/plain; charset="us-ascii"
Mime-Version: 1.0
X-Mailer: Windows Eudora Light Version 1.5.2
X-Sender: [email protected] (Unverified)
Message-Id: <[email protected]>
Received: from doug.ndip.eskimo.net (doug.ndip.eskimo.net [204.122.25.105])
by mail.eskimo.com (8.8.8/8.8.8) with SMTP id JAA24995;
Wed, 10 Dec 1997 09:25:32 -0800 (PST)
Received: from mail.eskimo.com (mail.eskimo.com [204.122.16.4])
by thalia.fm.intel.com (8.8.6/8.8.5) with ESMTP id JAA16762
for <[email protected]>; Wed, 10 Dec 1997 09:26:29 -0800
(PST)
Received: from thalia.fm.intel.com (thalia.fm.intel.com [132.233.247.11])
by fmm
ail.fm.intel.com (8.8.5/8.7.3) with ESMTP id JAA09918 for
<[email protected].
intel.com>; Wed, 10 Dec 1997 09:25:32 -0800 (PST)
Return-Path: [email protected]