Good morning Hal,
TDR testing is indeed performed at 10 to 20 MHz and the storage scope
measuring the response has a band width of typically at least 1.5 GHz.
Depending on the instrumentation and it's setup, higher frequencies can be
used from the frequency generator, a fast step output pulse generator,
built into the storage scope or in some instances, an external pulse generator
can be connected to the scope. Either way, higher frequencies can be used for
TDR measurement.
Materials usually specify relative permittivity, Er, at a frequency. It is
typically 1Mhz, but the material data sheet will specify the frequency. Why is
the frequency important? Because materials are frequency sensitive....FR4's
Er changes from 4.85 at 1MHz to 4.2 at 1000MHz (1Ghz). Since Er determines the
value of the impedance of the etched line, the frequency of the measurement
should match the frequency of system. Actually the rise time/fall time of the
logic is used as it is faster than the clock frequency. A 100Mhz Pentium chip
will run with logic at least 5 times faster and more likely at least 10 times
faster.
Remember that the bandwidth of the measuring instrument, the TDR scope
should have a bandwidth of at least 10 times the frequency being measured.
To specify TDR measurement with customers who want different frequencies,
have them specify either the frequency of the step pulse generator or it's
rise time. I would think you could standardize at some higher frequency, lets
say something between 500MHz and 1000MHz.
Through Dana Korf, I know you are persuing a Polar Instruments TDR sysytem of
measurement. You can contact Andy Burkhardt at Polar Instruments, in the U.K at
011-44-0481-53081 as to their capability to program in a frequency change to
their instrumentation. Changing setups with a pulse generator might be a pain.
I have reference material that I can FAX or mail to you if you want it. They
are:
1. A paper on Er versus frequency for FR4 material
2. A tutorial, 35-40 pages, on High Speed/High Frequency Design including
a section on TDR testing
3. The IPC test method for TDR testing, IPC-TM-650, method 2.5.5.7.
Two other resources on this topic are Dana Korf, Zycon and Lee Ritchey, 3Com
408-764-5736.
Please call me at the below number if you want to chat. I expect to be in my
office most of the day.
Cheers
Harry Parkinson
Tel 603-884-6760
FAX 603-884-1036
From: US2RMC::"[log in to unmask]" "MAIL-11 Daemon" 24-JUL-1995 19:54:32.61
To: [log in to unmask]
CC:
Subj: TDR test frequency
Recently several of our customers have began asking for impedance
testing to be performed at various frequencies, from 200 Megahertz to
1 Gigahertz.
From what I have been told about TDR operation, the device works with
a carrier frequency of between 10 Megahertz and 20 Megahertz, and a
response frequency in the "X-band" of approx. 1.5 to 2 Gigahertz.
Can anyone shed some light on what TDR test frequencies are and how
they should be specified?
Hal Brown
Zycon Corp.
[log in to unmask]
(408)261-4019 voice
(408)244-6349 fax.
% ====== Internet headers and postmarks (see DECWRL::GATEWAY.DOC) ======
% Received: from mail1.digital.com by us2rmc.zko.dec.com (5.65/rmc-22feb94) id AA02806; Mon, 24 Jul 95 19:36:16 -040
% Received: from miso.wwa.com by mail1.digital.com; (5.65 EXP 4/12/95 for V3.2/1.0/WV) id AA02073; Mon, 24 Jul 1995 16:30:18 -070
% Resent-Date: Mon, 24 Jul 95 13:18:43 PST
% Received: from ipc by gagme.wwa.com with uucp (Smail3.1.28.1 #8) id m0saWvp-000FF0C; Mon, 24 Jul 95 18:29 CD
% Received: by ipchq.com (Smail3.1.28.1 #2) id m0saWBi-0000HXC; Mon, 24 Jul 95 17:41 CD
% Old-Return-Path: <miso!zycon.com!hbrown>
% Date: Mon, 24 Jul 95 13:18:43 PST
% From: "hbrown" <[log in to unmask]>
% Message-Id: <[log in to unmask]>
% To: [log in to unmask]
% Subject: TDR test frequency
% Resent-Message-Id: <"1xTqu.0.zU8.K625m"@ipc>
% Resent-From: [log in to unmask]
% X-Mailing-List: <[log in to unmask]> archive/latest/722
% X-Loop: [log in to unmask]
% Precedence: list
% Resent-Sender: [log in to unmask]
|