|
Posted by BJ on August 8, 2006, 12:34 pm
If you were Registered and logged in, you could reply and use other advanced thread options
We're having our office rewired; old Cat3 and Cat5 wiring is being replaced
with a shielded Cat6. (The reason for shielded is another subject
completely and is a done deal, so please, no "Why Shielded? You don't need
it" posts. I will say that having a shielded solution is of lesser
importance today, but will take on greater importance in the future.)
My concern has to do with the proper way to implement a shielded ethernet
solution. We got two quotes. One was very specific about materials, from
someone whose expertise I trust. We sent his itemized list of materials to
another company and asked them to match the materials and give us a price.
Their quote was less detailed, but since it was supposed to be based on the
first quote I assumed they quoted the same or equivalent stuff. The cost
was similar. For reasons of business politics too complicated to go into,
we selected the second company's "equivalent" quote.
Now stuff is starting to arrive. Where the first quote explicitly provided
a shielded solution from end-to-end, some of the materials arriving are
indeed not shielded products. Instead of shielded patch cables we're
getting unshielded. Instead of an empty multimedia panel into which
shielded jacks would be installed, we're simply getting a cat 6 patch panel,
which I haven't been able to verify supports a shielded solution (I'll check
it out when it arrives).
This leads to my questions:
- I am assuming that a shielded patch panel or jack provides the means of
connecting all the shields from the horizontal cabling and the patch cables,
effectively creating a single ground point for all cable segments from end
to end. Is this correct?
- If I'm only concerned about EMI in my horizontal cabling locations, and
think EMI will not be an issue in my rack or office locations at present, is
it appropriate to use unshielded patch cables, as long as the shielded
horizontal cabling is properly grounded to the patch panel (which is itself
grounded thru a properly-grounded rack)?
- If the cable shield is ungrounded, does the shield simply provide no
benefit, or does it create the potential to cause harm? I've read
speculation that an ungrounded shield could act as an antenna that actually
increases EMI exposure to the pairs inside.
Please understand, I'm not asking for an opinion about whether EMI is likely
to pose a problem in my environment; I'm asking for sound ethernet grounding
theory.
Thanks in advance!
Bryan
|
 
| |
Posted by glen herrmannsfeldt on August 8, 2006, 2:37 pm
If you were Registered and logged in, you could reply and use other advanced thread options
BJ wrote:
> We're having our office rewired; old Cat3 and Cat5 wiring is being replaced
> with a shielded Cat6. (The reason for shielded is another subject
> completely and is a done deal, so please, no "Why Shielded? You don't need
> it" posts. I will say that having a shielded solution is of lesser
> importance today, but will take on greater importance in the future.)
(big snip)
I first learned about the problems of grounding looking at a big
rack full of electronic equipment, all with the ground connection on
the power cords cut off...
> - If I'm only concerned about EMI in my horizontal cabling locations, and
> think EMI will not be an issue in my rack or office locations at present, is
> it appropriate to use unshielded patch cables, as long as the shielded
> horizontal cabling is properly grounded to the patch panel (which is itself
> grounded thru a properly-grounded rack)?
Are you asking about EMI from ethernet to other equipment, or EMI from
outside into ethernet?
Shielding and grounding is a complicated subject once you have more
than one piece of equipment that is line powered. Electrically,
the best way is to have only one point connected to earth ground,
with everything else grounded through the shield. This often doesn't
agree with electrical codes requiring equipment to be grounded through
the power line.
> - If the cable shield is ungrounded, does the shield simply provide no
> benefit, or does it create the potential to cause harm? I've read
> speculation that an ungrounded shield could act as an antenna that actually
> increases EMI exposure to the pairs inside.
Ethernet is transformer coupled, and is pretty immune to outside EMI.
The possible exceptions are factories full of arc welders, or inside
a building with a high powered radio transmitter. There are possible
resonant situations where the shield length might be the appropriate
multiple of a half wavelength for a nearby EMI source, or for the
ethernet signal itself. In some cases a resistor between the shield
and the rest of the system might be needed.
> Please understand, I'm not asking for an opinion about whether EMI is likely
> to pose a problem in my environment; I'm asking for sound ethernet grounding
> theory.
Without knowing the possible EMI sources, it is hard to say. Sound
grounding theory says ground the cable to earth at only one point,
everything else grounds through the cable shield. There can only
be one grounding path between any two pieces of equipment, and that
path is through the shield. RJ45 connectors are not very good at
making ground contact, so you should connect from the shield to
a screw connection on each end of the cable.
-- glen
|
|
Posted by BJ on August 8, 2006, 3:23 pm
If you were Registered and logged in, you could reply and use other advanced thread options
Glen replied:
> Are you asking about EMI from ethernet to other equipment, or EMI from
> outside into ethernet?
I was asking about EMI from outside into ethernet, although the reverse
might also be a concern. Many of my runs will be supplying PoE to end
devices, so I suppose the potential exists for EMI between bundles of wires
running parallel for some distance.
>> - If the cable shield is ungrounded, does the shield simply provide no
>> benefit, or does it create the potential to cause harm? I've read
>> speculation that an ungrounded shield could act as an antenna that
>> actually increases EMI exposure to the pairs inside.
>
> Ethernet is transformer coupled, and is pretty immune to outside EMI.
> The possible exceptions are factories full of arc welders, or inside
> a building with a high powered radio transmitter. There are possible
> resonant situations where the shield length might be the appropriate
> multiple of a half wavelength for a nearby EMI source, or for the
> ethernet signal itself. In some cases a resistor between the shield
> and the rest of the system might be needed.
So, do I understand you to say that if my shield remains ungrounded for the
time being, the possibility *does* exist for the shield to "pick up" on
EMI -- but that the chance of just the "right" source of EMI being around is
unlikely?
> RJ45 connectors are not very good at
> making ground contact, so you should connect from the shield to
> a screw connection on each end of the cable.
So, if my EMI concerns are, at present, related to bundles of wires running
PoE, and to in-ceiling and in-wall locations, and NOT related to my rack and
office environments, should I still worry about having shielded, grounded
patch cables in my rack and in my offices?
--Bryan
|
|
Posted by glen herrmannsfeldt on August 8, 2006, 7:02 pm
If you were Registered and logged in, you could reply and use other advanced thread options
BJ wrote:
> Glen replied:
>>Are you asking about EMI from ethernet to other equipment, or EMI from
>>outside into ethernet?
> I was asking about EMI from outside into ethernet, although the reverse
> might also be a concern. Many of my runs will be supplying PoE to end
> devices, so I suppose the potential exists for EMI between bundles of wires
> running parallel for some distance.
Ethernet is transformer coupled. There is a transformer at each end of
each twisted pair, which guarantees no common mode signal. That is,
the sum of the voltage and current on the two wires of a pair is always
zero. The transformer also keeps common mode signals from outside,
such as a cable might pick up being near a power line, away from the
ethernet receiver. It has been suggested here, though I never got
around to testing it, that ethernet will work with a 240 volt AC common
mode signal. (I will be careful when testing it!)
If you want more detail, we need to know the source of the EMI that
you are trying to keep out. It makes a big difference. Note that
low frequency magnetic fields will go right through the shield,
for example. High frequencies can get in through small unshielded
gaps at the end.
Also, it is fairly easy to get a signal going down the outside of
the shield. If you want to keep that away, you need the ferrite
rings that you see on many cables these days. (A bulge in the
cable near the connector.)
(snip)
> So, do I understand you to say that if my shield remains ungrounded for the
> time being, the possibility *does* exist for the shield to "pick up" on
> EMI -- but that the chance of just the "right" source of EMI being around is
> unlikely?
I think that is true, but it depends on what sources you expect.
>>RJ45 connectors are not very good at
>>making ground contact, so you should connect from the shield to
>>a screw connection on each end of the cable.
> So, if my EMI concerns are, at present, related to bundles of wires running
> PoE, and to in-ceiling and in-wall locations, and NOT related to my rack and
> office environments, should I still worry about having shielded, grounded
> patch cables in my rack and in my offices?
PoE usually means that the device won't have a power line ground, which
is good. In that case, ground to the rack and don't worry about the
patch cables.
One I haven't heard about at all is the effect of 802.11 devices at the
end of an ethernet cable. You could easily couple that into your
shielded cable, even from inside the box, beside the fact that most
have a plastic box, anyway.
-- glen
|
|
Posted by BJ on August 8, 2006, 10:52 pm
If you were Registered and logged in, you could reply and use other advanced thread options
> If you want more detail, we need to know the source of the EMI that
> you are trying to keep out. It makes a big difference. Note that
> low frequency magnetic fields will go right through the shield,
> for example. High frequencies can get in through small unshielded
> gaps at the end.
I'm concerned mainly about EMI from power runs in the ceiling, and *lots* of
fluorescent lighting. Also, this cable *should* be capable of running
10gig, when it finally comes to that. Since (as I understand) alien
crosstalk in large bundles of cables may be one of the major obstacles to
10gig, we hoped a high-grade, shielded cable might increase our chances of
getting thru the 10gig generation of products before another rewire would be
needed. We briefly considered fiber, but realized early on that we wanted
to do PoE, not to mention the other challenges of fiber. So.
> Also, it is fairly easy to get a signal going down the outside of
> the shield. If you want to keep that away, you need the ferrite
> rings that you see on many cables these days. (A bulge in the
> cable near the connector.)
So that's what those are. Thanks, I'll remember that.
> PoE usually means that the device won't have a power line ground, which is
> good. In that case, ground to the rack and don't worry about the
> patch cables.
It sounds like having shielded patch cables isn't worth making a fuss about,
as long as the cable is grounded to the rack. If that's so, that'll make
things easier.
BJ
|
| Similar Threads | Posted | | How to make a shielded cat6 patch cable | July 8, 2005, 9:59 am |
| Newbie question: How many switches can I install 100m apart? | January 21, 2007, 12:27 pm |
| USB Ethernet controllers that use the ASIX AX88772 or AX88172 USB-Ethernet bridge devices | August 19, 2005, 4:04 pm |
| Difference between Ethernet 2 and 802.3 Frame per the Ethernet FAQ | July 28, 2006, 9:02 am |
| ethernet hub | April 27, 2006, 2:44 am |
| WAN Ethernet | May 2, 2006, 4:20 pm |
| Ethernet TAP | July 30, 2007, 12:28 am |
| Do I have an Ethernet card? | February 20, 2005, 10:48 am |
| Ethernet Repeaters | March 9, 2005, 8:10 am |
| Need Ethernet Hub - NOT Switch | March 15, 2005, 7:15 pm |
|
|
Yahoo! 
Google 
Windows Live 
del.icio.us 
digg 
Netscape 
XML site map