Jeff,
You are just fine on what you understand about STP wiring. The problem is in
the TERMINATION of those shields. Two truths must exist in order for the
shield to do its job correctly:
1) It MUST have a full circumferential grounding contact to the chassis of the
equipment into which the cable gets connected. Except for direct hard wiring,
this does not happen with the conventional RJ-45 jack, "shielded" or not; 2)
The net path to chassis at each end MUST be low enough in impedance to
effectively route the currents induced into the shield to the chassis. The
"rule of thumb" for effectively low Z is 1/8th to 1/20th wavelength at the
highest frequency of interest, with shorter path length preferred.
Without these two conditions in effect, STP is not much better that UTP with
respect to net shielding effectiveness ("SE"). There are other factors,
notably the transfer impedance of the shield itself (referred to as "Zt"), and
that is where the type of shield construction comes into play. (For example,
braided shields are better than spiral shields, and 100% coverage shields are
best of all.) In aerospace systems that must meet the requirements of
MIL-STD-461, network cables operating at 100 Base-T speeds, or faster, can not
meet the requirements using standard RJ-45 connectors. Instead, STP is used,
but the terminations are changed from RJ-45 to MIL-spec circular, or sometimes
sub-D, connectors that have metallic back shells and full circumferential
termination of the shield. Note: the "drain" wire that is often used to ground
a cable shield makes a fine antenna for radiating the currents it carries
unless kept very short and hard-grounded to chassis.
Jim, K9YC, talks about the shields going direct to chassis. That's the problem
with standard RJ-45 plugs and jacks. The shield current path is neither short
nor direct, and as a result, re-radiation of currents on the shield occurs with
great gusto. See my next few paragraphs for information on ways to correct the
problem, but not easily. However, Jim has made many postings on this reflector
about the use of ferrites applied to cables as a means to reduce or limit
radiated interference from the associated equipment. Each user experiencing
interference from a LAN cable needs to determine the amount of reduction
necessary to solve his or her problem. If a reduction of interference by about
10 dB appears to be sufficient, judicious use of ferrite chokes may very well
be enough to solve the problem. Otherwise, if faced with needing a reduction
of 30 dB, or more, plan to use STP and perform hardware hack mods as discussed
below.
I recently saw an advertisement from a company that is now offering something
that should be better than the "shielded" RJ-45s seen to date. Their RJ-45
jack is fully encased in a metal shroud that mounts to the chassis or main pc
board of the equipment. They have a matching RJ-45 plug which, I do believe,
does offer the full circumferential termination of the cable shield, with a
back shell. However, this is not a "mod" to hang onto a computer or router.
The equipment must be designed from Day 1 to use this special jack, as it
becomes part of the device design.
When my house was being built back in 2000, I pre-wired it for many
conveniences, including a wired LAN. I used regular UTP, since there was no
equipment offering proper shield terminations at that time. I have little
problem except for the spur that occurs right on 145.39 MHz, which happens to
be the output frequency of the local repeater of which I am the trustee. The
spur will get into an HT if I get within about a meter of the LAN cable. I do
not hear the spur when using a base radio and an outside antenna, and I hear no
evidence of the LAN when operating HF or 6 meters.
You might wonder why I just didn't go with a wireless LAN. Back then, I think
most wireless LAN hardware used the 2.4GHz band. I am an avid VHF/UHF op, and
I did not want a non-ham signal in that band present at my location should the
day come that I actually get active on the 12cm band.
I hope that my explanation of the problems associated with correct termination
of STP shields helps you understand that the issue is not the cable - it's the
means of terminating that shield and the real lack of a correct way to do so
with about 99% of the equipment that is currently available. Enterprising hams
who have real significant issues with radiation from LAN cables may wish to
"bite the bullet" and make some serious "hack mods" to their system equipment
so that the shields of STP cables will be effective. However, it is not
something for the faint of heart or those who worry about warranty coverage to
contemplate. Adding true shielded jacks or removing jack/plug combos and going
to direct hard wiring, plus providing full shielding of the equipment
when/where necessary, will be a challenge.
73, Dale
WA9ENA
Sr EMC Engineer
E-N-A Systems, LLC
-----Original Message-----
>From: Jeff Stevens <jeff@mossycup.com>
>Sent: Mar 1, 2013 3:18 PM
>To: rfi@contesting.com
>Subject: Re: [RFI] RFI suppression cores for 147 MHz
>
>On Fri, Mar 1, 2013 at 12:02 PM, Jim Brown <jim@audiosystemsgroup.com>wrote:
>
>>
>> I don't work in the computer networking world, but I have seen shielded
>> CAT5/6/7 patch cables. They are terminated with connectors that have a
>> metal surround to mate with similar female connectors.
>>
>
>And that seems to be the easy part. I've done runs with STP terminated
>with metalized RJ-45s. They are installed exactly the same way as
>standard plastic connectors with the exception that the foil shield and
>drain wire are placed in positive contact with the metal on the RJ-45
>before crimping.
>
>
>> Like ALL cable shields, it must go straight to the shielding enclosure at
>> each end to be effective.
>
>
>This seems to be the difficult part. Core networking equipment (switches,
>routers, etc.) is readily available which includes contact points to
>connect the shielded RJ-45 to the device chassis ground. On the other
>hand, terminal equipment such as computers and networked printers virtually
>never come with the appropriate contact points to accommodate a shielded
>RJ-45. Simply examine the NIC on any computer -- the grounding contact
>points are missing. This is what confuses me when people suggest the use
>of STP. What does one do to resolve the issue that there is no contact
>between the STP shield and the terminal equipment chassis/ground? I
>suppose that the drain wire could be connected to the computer chassis
>using a drilled and tapped screw but there has to be less labor intensive
>off the shelf solution, no? It's left me thinking that there is some
>aspect of STP wiring that I don't understand.
>
>-Jeff
>W7WWA
>_______________________________________________
>RFI mailing list
>RFI@contesting.com
>http://lists.contesting.com/mailman/listinfo/rfi
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|