Received: |
by ipchq.com (Smail3.1.28.1 #2)
id m0s9ZFi-0000GeC; Thu, 11 May 95 09:30 CDT |
Old-Return-Path: |
<miso!rhino.qpsx.oz.au!LOTHAR> |
Date: |
Thu, 11 May 1995 10:16 WAT |
Precedence: |
list |
X-Loop: |
|
From [log in to unmask] Thu Apr 25 15: |
32:12 1996 |
Resent-Sender: |
|
X-Status: |
|
Status: |
O |
X-Mailing-List: |
|
TO: |
|
Return-Path: |
|
Resent-Message-ID: |
<"kmvxy3.0.SD7.hzXil"@ipc> |
Subject: |
|
From: |
|
Resent-From: |
|
Message-Id: |
|
Parts/Attachments: |
|
|
We have a requirement to make a quantity of circuit card assemblies for a
50 Ohm RF application which requires a uniform phase delay over the
entire production quantity, to be tested at 32MHz using a network
analyser.
The circuit tracking has been designed around FR4 substrate, with a
nominal dielectric constant of 4.7 and 0.060'' thick. However, the
standard range of 4.6 to 4.8, ie 4.7 +/- 2%, is too wide and we are
looking for stock with a tolerance of 1% or better. The nominal value
does not have to be exactly 4.7 but should be in that range.
One way of achieving this is to select the sheets from standard stock by
checking capacitance, area and thickness of each sheet and calculating
the dielectric constant. However, we are told that a typical sheet is
thicker in the middle and thinner around the edges, and we may need to
use only a centre portion of each sheet to obtain better uniformity.
Does anyone have experience in this area, or suggestions as to how to go
about this?
[log in to unmask]
Lothar Thole
Senior Manager - Manufacturing Planning
QPSX Communications Ltd
33 Richardson Street
West Perth 6005
Western Australia
Tel. +61-9-262-2000
Fax. +61-9-324-1642
|
|
|