Hi Doug,
thanks a lot for your reply. I take this chance to express my appreciation
for your web page with so many useful articles and suggestions for people
in this field (especially the Technical Tidbits).
Yes I know about the potential problems related to the shield, and I also
know your article.
I used to work with printers (in my previous life as EMC engineer) and I
discovered many times that the shield of common interface cables like
Centronics and RS232 worked more as antennas than as shields.
The reason why we are using a shielded cable for 10bT is that in this way
we are exempted from GR1089 Intrabuilding Surge requirements. If the cable
is not shielded we have to test and we should put extra protections on the
interface, but that would be a problem because one of the affected units is
bought from an outside vendor and we can no longer get it changed.
In any case, we already tested radiated emissions and immunity and found no
difference between UTP and FTP (that means the ground to which the shield
is connected is clean enough RF-wise).
So, being stuck with shielded cables we are considering STP vs FTP, and I
already told the rest of the story...
Paolo
------------------------------------------------------------------------------------------------
At 11:37 03/09/2001 -0700, you wrote:
Hi Paolo,
I worked for a few years in the part of Bell Labs that was involved with
networking cable systems several years ago. I am suspicious of foil
shielded cables as it is difficult to properly terminate the foil shield
(360 degrees). I have had a foil shield crack and split if the cable is
flexed very much.
On a related topic, a good UTP (unshielded twisted pair) system can have
lower emissions and higher immunity than most shielded systems. The
interface design is not hard to do to make this happen. I have seen
cases where the presense of the shield causes immunity problems because
a large noise current lands on the chassis of the data equipment. It
there are slits and holes in the chassis (back of a PC?) problems
result. Take a peek at the paper:
http://www.dsmith.org/pdf/roma94.pdf
which I authored several years ago and presented at the EMC Roma
conference. In that paper a couple of commercial STP systems were
compared with a good UTP system at a test lab in Switzerland. The UTP
system won hands down on both emissions and immunity! The presense of a
shield, by itself, does not guarantee good performance. I seem to
remember also that the shielded systems had higher losses for signal
transmission than UTP which resulted in shorter permissable cable
lengths.
Paolo, we can go offline if you want to discuss more details. Either
call or email me.
Doug
Paolo Roncone wrote:
>
> Hi all,
>
> anybody out there can explain the difference between FTP (Foiled
> Twisted Pair) and STP (Shielded Twisted Pair) cables used for Ethernet
> links.
> In Ethernet standard IEEE 802.3 (2000) I find only a definition of STP
> (herebelow in copied and pasted from the standard - page 28):
> In a vendor's catalog, I found a definition of FTP as
>
> <<1.4.249 shielded twisted-pair (STP)cable: An electrically conducting
> cable,comprising one or more ele-
> ments,each of which is individually shielded.There may be an o erall
> shield,in which case the cable is
> referred to as shielded twisted-pair cable with an o erall shield
> (from ISO/IEC 11801:1995).Speci .cally
> for IEEE 802.3 100BASE-TX,150 . balanced inside cable with performance
> characteristics speci .ed to
> 100 MHz (i.e.,performance to Class D link standards as per ISO/IEC
> 11801:1995).In addition to the
> requirements speci .ed in ISO/IEC 11801:1995,IEEE 802.3 Clauses 23 and
> 25 provide additional perfor-
> mance requirements for 100BASE-T operation o er STP.>>
>
> The reason of my inquiry is that we bought samples of "STP" and "FTP"
> cat.5 cables for 10bT ethernet applications from different vendors and
> to our surprise we discovered that both "STP" and "FTP" types have an
> overall (external) shield made of aluminum foil, but no shields on
> individual wires or wire couples (as per 802.3 definition above).
>
> Any inputs, suggestions etc. would be appreciated.
>
> Paolo
>
>
-------------------------------------------------------------------------------
> Paolo Roncone
> EMC Compliance Engineer - Cisco Photonics Italy
> via Philips 12 - Monza (MI) 20052
> mailto:[email protected]
> phone: +39 039209 1538
> fax: +39 039209 2036
--
-------------------------------------------------------
___ _ Doug Smith
\ / ) P.O. Box 1457
========= Los Gatos, CA 95031-1457
_ / \ / \ _ TEL/FAX: 408-356-4186/358-3799
/ /\ \ ] / /\ \ Mobile: 408-858-4528
| q-----( ) | o | Email: [email protected]
\ _ / ] \ _ / Website: http://www.dsmith.org
-------------------------------------------------------
-------------------------------------------------------------------------------
Paolo Roncone
EMC Compliance Engineer - Cisco Photonics Italy
via Philips 12 - Monza (MI) 20052
mailto:[email protected]
phone: +39 039209 1538
fax: +39 039209 2036
