[SI-LIST] Re: Main attenuation effect of long cables
- From: "Michael Poimboeuf" <Michael_Poimboeuf@xxxxxxxxxxxxxx>
- To: <cchalmers@xxxxxxxxxxx>, "SI-List" <si-list@xxxxxxxxxxxxx>
- Date: Fri, 6 Aug 2004 10:59:31 -0700
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx=20
> [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Chris Chalmers
...
>=20
> With standard 100ohm twisted pair cable (such as cat 5e or cat 6)
> when they reach high frequencies (100MHz) at 10Metres, what is the
> main factor that causes the attenuation. Here are some options,
>=20
> 1. The AWG of the conductors
> 2. The dielectric properties (if the pairs are enclosed)
> 3. Transmission line effects, reflections
> 4. Crosstalk, NEXT, FEXT.
> 5. Noise pickup.
>=20
> The signal transmitted would be LVDS.
>=20
> Also I have read Doug Smiths excellent note on the comparison
> of the EMC performance of shielded versus unshielded cables.
> Does any one else have views on whether its worth using a shielded
> cable for 100ohm twisted pair data transmission?
>
=20
Any of the issues #1-#5 can cause signal integrity problems.
IEEE802.3ab-2002 (aka the 1GBT specification) includes analysis of the
magnitude of the various factors.
Yes, attenuation is due to #1-#3 (distributed RLC all matter vs.
frequency). Yes, transmission line effects, especially poor return loss,
leading to reflections cause signal integrity problems as in #3. Yes,
near and far end crosstalk in #4 is a big issue. Noise pickup, my
favorite, ALIEN crosstalk, in #5 is a big issue.=20
Regarding the question of "shielding."* I've read Doug's paper
http://emcesd.com/pdf/roma94.pdf too, and I've referenced its results
many times. Recently however, I've come across a case using 1GBT where
ISO/IEC 11801 STP Category 6 patch cables gave much better results for
short patch cables than the Category 6 UTP cables of the same length
(around 10 meters). We've found that for systems that are well grounded
on both ends, screen terminations have given us dramatically improved
immunity from alien crosstalk, ESD, and EMI (#5 in Chris Chalmers'
list).=20
The Cat6 STP cables we're using have much poorer NEXT than the
equivalent Cat6 UTP, but we have suffient SNR margin to mitigate that
loss in our application.=20
Ground paths are the biggest factor (aren't they always? ;-), so the STP
cables must have good RF chassis ground returns on both ends (STP cable
grounded on one end becomes an antenna). Grounding on both ends can
cause galvanic isolation issues and ground loops, but for short
distances it can be done in ways to mitigate the loss of galvanic
isolation. I suspect that the poor performance of STP that Doug measured
may have been due to screen termination issues and galvanic isolation
between the network components he tested. I have the luxury in my tests
of using the STP patch cable as a point-to-point connection, not a
network configuration.
*STP is "screened" cable, not the same as shielded. Screened cable has a
foil screen around the entire bundle of twisted pairs. Shieled cable
(such as Class F aka Cat 7 - but there is no Cat 7 spec yet), has
shields around each of the pairs and the entire bundle.
--
Michael Poimboeuf
------------------------------------------------------------------
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 FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
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:
