[SI-LIST] Re: Difference betweem single stripline & dual stripline
- From: andrew.c.byers@xxxxxxxxxxxxxx
- To: jeff.loyer@xxxxxxxxx, whizplayer@xxxxxxxxx
- Date: Fri, 2 May 2003 09:45:10 -0700
Hello Bob and all,
Two comments on broadside-coupled striplines vs. edge-coupled striplines:
1) As frequencies increase and you enter the skin effect region, your
conductive loss will increase with the square root of frequency. But with
coupled lines, there is another mechanism that influences the loss curve,
which is called the "proximity effect". This is essentially currents
crowding to the edge of the stripline that is closer to the other line. This
effect can be difficult to quantify without an EM solver that solves for
currents in the conductors. There might be some approximate equations people
have developed in dissertations out there. My point is that the
broadside-coupled striplines will behave differently with regard to this
current crowding phenomenon. Of course, this won't really matter too much if
your loss is dominated by the dielectric loss tangent...
2) Depending on how flexible your via choices are, it might be more
difficult (and unbalanced) to hook into a broadside-coupled stripline
configuration. You will have to use a longer via or two microvias to get to
the lower trace, and maybe design some funny transition. Any difference in
path length or discontinuity will generate common mode, which you probably
don't want. Keeping the striplines on the same layer gives you a better shot
at keeping both lines absolutely symmetric.
Andrew Byers
-----Original Message-----
From: Loyer, Jeff [mailto:jeff.loyer@xxxxxxxxx]
Sent: Friday, May 02, 2003 7:14 AM
To: whizplayer@xxxxxxxxx
Cc: si_list
Subject: [SI-LIST] Re: Difference betweem single stripline & dual stripline
Hi Bob,
Here's a quick synopsis of a short study I did using XFX to model 4 =
topologies:
1) True "stripline"
2) Asymmetric stripline
3) Embedded (or I'm coining the phrase "homogenous") microstrip - =
meaning there's only a single reference plane, and the trace is embedded =
far enough in the dielectric that it sees only a homogenous dielectric
4) Microstrip, with only a thin dielectric covering the trace.
Findings:
True Stripline has conductor(s) sandwiched equidistant from 2 reference =
planes (Symmetric), and has homogenous dielectric. This is contrary to =
a previous understanding that I had - I thought stripline implied only a =
homogenous dielectric.
Of the 4 topologies studied - Symmetric and Asymmetric Striplines, =
Homogenous (embedded) Microstrip, and Microstrip -=20
FEXT is 0 for Symmetric Stripline
FEXT is very low (0.002) for traces on the same layer of Asymmetric =
Stripline and Homogenous Microstrip
FEXT is fairly high (0.016) for traces on different layers of =
Asymmetric Stripline
FEXT is highest (0.04) for traces on Microstrip
NEXT is lowest (~0.03) for Symmetric Stripline and traces on =
different layers of Asymmetric Stripline
NEXT is highest (-.13) for both Microstrips and traces on the same =
layer of Asymmetric Stripline
Losses are approximately equal, except in case where trace widths are =
reduced for symmetric stripline (to obtain correct Z0) - in that case =
losses are higher
Conclusions:
Given choice between Asymmetric Stripline or Microstrip:
Microstrip Traces will have significantly more FEXT
Microstrip Traces will have the same NEXT as Asymmetric Stripline =
traces on the same layer
Losses are similar
Traces on different layers of Asymmetric Stripline are susceptible to =
FEXT - should be routed orthogonally
Symmetric Stripline (once trace widths are reduced to obtain the same =
Z0) are more lossy than other topologies
Another interesting note: XFX calculates forward and reverse crosstalk =
as:
Fwdx =3D (Se - So)/(0.5(Se + So))=20
Rvsx =3D (Ze - Zo)/(Ze + Zo); not sure why units are given as v/v=20
I'd love to hear where these equations were derived. The XFX manual I =
have doesn't say.
This was a quick study, and I wouldn't deem it conclusive. I'll be =
interested to hear what others have found...
My study was also only for single-ended traces. The same concern about =
increased loss for symmetrical (edge-coupled) stripline would hold for =
differential, if you had to narrow the traces to attain the correct Z0, =
versus broadside-coupled.
Jeff Loyer
-----Original Message-----
From: Bob Patel [mailto:whizplayer@xxxxxxxxx]
Sent: Thursday, May 01, 2003 2:19 PM
To: si_list
Subject: [SI-LIST] Difference betweem single stripline & dual stripline
Hi!
Is there a difference in terms of electrical
parameters i.e. say loss, etc when using single
symmetric stripline layer and dual stripline layer.
I guess I am asking beside the cost & geometry is
there a harm if I use dual stripline layer over single
stripline layer where adding more routing layer is
necessary for me.
I ran cases to look at differences if I route edge
coupled differential traces on single symmetric
stripline or if I use dual stripline. I did not any
difference at least in loss or impedance etc.
Any input is appreciated.
THanks
Bob
__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at: =20
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages=20
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
=20
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
