silly question - UTP or STP cables ???

silly question perhaps
I always assumed 100-Base-T with UTP - ie Unshielded twisted pair

I now see that cat 5e patch cables are available in STP for the same price

is this a case of:
"won't make any real difference but 'shielded' is *theoretically* better
than 'unshielded' so get the shielded version..."

OR

have I made a mistake and they aren't interchangeable ???

"no-name" <no-name@false-address.net> wrote:
>silly question perhaps
>I always assumed 100-Base-T with UTP - ie Unshielded twisted pair
>
>I now see that cat 5e patch cables are available in STP for the same price
>
>is this a case of:
>"won't make any real difference but 'shielded' is *theoretically* better
>than 'unshielded' so get the shielded version..."
>
>OR
>
>have I made a mistake and they aren't interchangeable ???

Shielding has certain advantages... and even more disadvantages.

The capacitance between the shield and the conductors has a
significant effect on the length of cable you can use at any
given frequency. Hence, you'll do fine if you have no long
cable runs.

With patch cords, which are typically short, it would appear
that STP would be insignificantly different that UTP.

--
FloydL. Davidson <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@barrow.com

Floyd L. Davidson <floyd@barrow.com> wrote:

> With patch cords, which are typically short, it would appear that STP
> would be insignificantly different that UTP.

Plus don't you have to earth them [STP] at one end?

--
Cameron Kerr
cameron.kerr@paradise.net.nz : http://nzgeeks.org/cameron/
Empowered by Perl!

> > With patch cords, which are typically short, it would appear that STP
> > would be insignificantly different that UTP.
>
> Plus don't you have to earth them [STP] at one end?

And *only* one end. Grounding (Earthing) at both ends creates a path for
current to flow to ground OTHER than the mains wiring, and if there are a
few amps (or more) going to ground, you do *not* want it going through your
shielding ("Hey! Why did the network just die?"). You also don't
necessarily need the hum from having the ground loop, either.

steve

Steve Wolfe wrote:
>>>With patch cords, which are typically short, it would appear that STP
>>>would be insignificantly different that UTP.
>>
>>Plus don't you have to earth them [STP] at one end?
>
>
> And *only* one end. Grounding (Earthing) at both ends creates a path for
> current to flow to ground OTHER than the mains wiring, and if there are a
> few amps (or more) going to ground, you do *not* want it going through your
> shielding ("Hey! Why did the network just die?"). You also don't
> necessarily need the hum from having the ground loop, either.
>
> steve
>
>

Just an anecdote about grounding. A couple of years ago, working on a
seismic crew in the desert, where offices are setup in trailers for a
few weeks, I happened to find out that the gounding of some trailer was
badly setup. We had coaxial cable on the network at the time. Working
on a pc, connecting a (metallic) BNC to a card, I happen to notice a
tiny spark. By curiosity, I redisconnected, took a multimeter and
checked. The reading was 90 volts... But nobody was complaining about
the network. (!)

Gaetan

Steve Wolfe wrote:

>> Plus don't you have to earth them [STP] at one end?
>
> And only one end. Grounding (Earthing) at both ends creates a path for
> current to flow to ground OTHER than the mains wiring, and if there are a
> few amps (or more) going to ground, you do not want it going through your
> shielding ("Hey! Why did the network just die?"). You also don't
> necessarily need the hum from having the ground loop, either.

Actually, that should be at only one point. You can ground the shield
anywhere along it's length, so long as there is only one grounding point.
However, with most cables, it's easiest to ground at the ends.

--

Fundamentalism is fundamentally wrong.

To reply to this message, replace everything to the left of "@" with
james.knott.

Gaétan Martineau wrote:

> By curiosity, I redisconnected, took a multimeter and
> checked. The reading was 90 volts... But nobody was complaining about
> the network. (!)

Just measuring voltage, may not give you an accurate indication of a
problem. You have to see if it can support a significant amount of
current. Get a proper leakage detector.

--

Fundamentalism is fundamentally wrong.

To reply to this message, replace everything to the left of "@" with
james.knott.

James Knott <bit_bucket@rogers.com> wrote:
>Steve Wolfe wrote:
>
>>> Plus don't you have to earth them [STP] at one end?
>>
>> And only one end. Grounding (Earthing) at both ends creates a path for
>> current to flow to ground OTHER than the mains wiring, and if there are a
>> few amps (or more) going to ground, you do not want it going through your
>> shielding ("Hey! Why did the network just die?"). You also don't
>> necessarily need the hum from having the ground loop, either.
>
>Actually, that should be at only one point. You can ground the shield
>anywhere along it's length, so long as there is only one grounding point.
>However, with most cables, it's easiest to ground at the ends.

Virtually *all* outside plant telecom cables are grounded at
each splice box. That is, *both* ends of the cable are
grounded.

It reduces noise induction into the pairs. Likewise, if it is a
large cable, grounding both ends of all spare pairs will also
reduce noise.

--
FloydL. Davidson <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@barrow.com

> > Plus don't you have to earth them [STP] at one end?
>
> And *only* one end. Grounding (Earthing) at both ends creates a path
for
> current to flow to ground OTHER than the mains wiring, and if there are a
> few amps (or more) going to ground, you do *not* want it going through
your
> shielding ("Hey! Why did the network just die?"). You also don't
> necessarily need the hum from having the ground loop, either.



However this is wrong.

I know what I'm talking about. It is my job



Normally you only ground on one end for low frequency signals.

Otherwise you get hum.

For high frequency signals (Ethernet) if you ground only one side you create
an FM antenna.

So therefore you design a system where you invite trouble.

The hum is no issue because the signals are symmetric.



Due to the fact that it is hard to maintain a network to be completely
shielded, with every patch cord shielded cables, and al computers/switches
connected to earth, you basically are asking for random troubles with
shielded cables. This is why UTP is almost always preferred.

In my jobonly once I was forced to use ftp cable in a discotheque.



FTP cable has a thin foiled shield, and a small earth wire. However and the
foil can be vaporized by lightning, changing your network from FTP to UTP.
Changes are not very likely that this will happen however. STP (however very
hard to get), has a real copper shield like coax. It is sometimes used for
military situation or on high transmission towers.



The shielding is also used to prevent other for listening on your network. I
have never seen this type of listing equipment though.



Almost no situation FTP or STP is useful.



Jan-Willem Michels

"news.xs4all.nl" <jwillem@stads.net> wrote:
>> > Plus don't you have to earth them [STP] at one end?
>>
>> And *only* one end. Grounding (Earthing) at both ends creates a path
>for
>> current to flow to ground OTHER than the mains wiring, and if there are a
>> few amps (or more) going to ground, you do *not* want it going through
>your
>> shielding ("Hey! Why did the network just die?"). You also don't
>> necessarily need the hum from having the ground loop, either.
>
>However this is wrong.
>
>I know what I'm talking about. It is my job

Hmmm...

>Normally you only ground on one end for low frequency signals.
>
>Otherwise you get hum.

"Hum" might be defined as "low frequency signals", however the
results of incorrect grounding are not limited to low
frequencies. That just happens to be common because power
distribution systems are 1) ubiquitous and 2) low frequency.

Higher frequency signals can also be significant problems.
Examples are AM radio transmissions, FM radio transmissions, arc
welding equipment, fluorescent light fixtures, any large power
load being switched on or off, and of course other data
communications signals (T1 lines, for example).

>For high frequency signals (Ethernet) if you ground only one side you create
>an FM antenna.
>
>So therefore you design a system where you invite trouble.
>
>The hum is no issue because the signals are symmetric.

Eh? I'm not sure what you mean by "symmetric", but low
frequency power line interference is and issue in *all* cases.

>Due to the fact that it is hard to maintain a network to be completely
>shielded, with every patch cord shielded cables, and al computers/switches
>connected to earth, you basically are asking for random troubles with
>shielded cables. This is why UTP is almost always preferred.

While it is true that UTP is preferred, the reason you give is
not correct. In fact it is *essential* to maintain a network
where all computers/switch/etc are in fact connected to an earth
ground. Maintaining shielding grounds is not difficult, but it
is generally unnecessary and more expensive.

In some specific cases, it is necessary regardless of the expense.

>In my jobonly once I was forced to use ftp cable in a discotheque.

Before I retired, in my job... we used shielded cabling for
virtually *everything*, with certain exceptions. DS1
interconnects within the same row that were no more than 5
racks apart can be UTP, and all other inter-bay cabling is
shielded. Plus cross-connect numbers on distribution frames
are not shielded.

The reason for using shielded cabling (in telephone central
office wiring) is as much to reduce induction of noise from the
particular cable being run as it is prevent induction of noise
from the environment into the cable! When you start getting
cable racks that have 6 to 12 inches (or even more) of cables,
those signals themselves are a *very hostile* environment.

>FTP cable has a thin foiled shield, and a small earth wire. However and the
>foil can be vaporized by lightning, changing your network from FTP to UTP.

Funny.

>Changes are not very likely that this will happen however. STP (however very
>hard to get), has a real copper shield like coax. It is sometimes used for
>military situation or on high transmission towers.
>
>The shielding is also used to prevent other for listening on your network. I
>have never seen this type of listing equipment though.
>
>Almost no situation FTP or STP is useful.

That is simply *not* true. There are *many* situations where
shielding in essential. Running one or two Ethernet systems
around a typical home or office does not require STP. On the
other hand, if we are talking 50 or more cables, or any other
condition that amounts to a hostile enough environment (e.g., a
place where arc welding is being done), then clearly STP (which
is almost always a foil wrapped cable) is appropriate. That
would include distribution cables for any large building (office
or apartments), but need not include the cabling within each
office or apartment, for example.

--
FloydL. Davidson <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@barrow.com