Maybe I missed something? Exacly how did you make this model?
You begin to have problems (fun if your an rf person) when the wavelength is on
the order of your features. At 800 Mhz assuming K=4 the wavelength in the dielectric
is about 7 inches, so your design is close to the challenging range.
As far as fabrication of boards at frequnecy the fundamental problem is dielectric loss.
At low frequencies electrical loss is almost totally conductor loss. However, at high
frequencies the signals begin to behave as electromagnetic waves which are propagating
largely in the dielectric media itself. So in short, dielectric loss dominates and
great attention must be paid to the board material.
Also note that conductor roughness can also be a problem for very high frequncy designs
since the skin effect causes the fields to be crowded to the outer skin of the conductors.
If conductors have a rough finish loss will result.
At virginia tech we have studied a few laminate materials (see ref 1 and 2 below),
by now means a comprehensive
study of all board materials, which are used in high density multichip modules. A
comparison was done with two laminates and a mulitlayer ceramic board. What we found
was that the two particular laminiates that we studied are of no use above 1-2 Ghz since
the loss becomes very large. By contrast Rogers makes some very nice laminates which are
useful well beyond 5 - 10 GHz. The point of all this discussion is that you need to evaluate
the particular board material you are using as to its ability to handle this frequnecy range.
I would also be very intrested to see what type of data is available for common board
materials? In my experince we are lucky to get accurate electrical charaterization at
1 MHz, and data above 500 MHz is very rare. Does any group or organization provide a
comprhensive database of this sort?
ref:
1: ISHM 1995 Proceedings, Wideband Electrical Charaterization of Materials used in MCM
applications, p430 - 435.
2. Wideband Electrical Characterization of MCM materials, to be published Electronic Packaging
and Production, June 1996.
NOTE: Both of these papers focus more on how to measure the properties than providing a database
of properties. However, this information may be of value to designers stretching the bandwidth
of laminates.
Mr. F. Barlow, Research Associate
The Microelectronics Laboratories
Virginia Polytechnic Institute and State University
Blacksburg, Virginia, 24061
(540)-231-6560
[log in to unmask]
> > > >From [log in to unmask] Tue May 7 19:03:51 1996
> > > Resent-Date: Tue, 7 May 1996 21:00:21 -0700
> > > Resent-Sender: [log in to unmask]
> > > Old-Return-Path: <[log in to unmask]>
> > > From: [log in to unmask]
> > > Date: Tue, 07 May 96 18:55:11 CST
> > > Encoding: 16 Text
> > > To: [log in to unmask], [log in to unmask]
> > > Subject: PC Board Frequency Limits
> > > Resent-Message-ID: <"U3ffi.0.8DH.W1_Zn"@ipc>
> > > Resent-From: [log in to unmask]
> > > X-Mailing-List: <[log in to unmask]> archive/latest/1686
> > > X-Loop: [log in to unmask]
> > > Resent-Sender: [log in to unmask]
> > > Content-Length: 659
> > >
> > >
> > > I would like any responses to be off-line for the following
> > > discussion.
> > >
> > > Please use the following email address: [log in to unmask]
> > >
> > > I made a simple mathematical model of a stripline, 1 inch trace, on
> > > FR-4, nothing out of the ordinary as far as dimensions and found to my
> > > surprise a resonant freq at about 800 MHz with a 20 to 40 dB roll-off
> > > above 800 MHz
> > >
> > > Could someone in the RF/Microwave world of nuts and bolts pc board
> > > design tell me what I'm supposed to do at say 10 GHz with regular pc
> > > board materials/construction/fabrication???
> > >
> > > Thanks in advance, Doug McKean
|
