Sheilded F/UTP grounding

Does anybody have any experience with the theory and practice behind shielded STP or F/UTP cabling.

My understanding is that the shield should only be grounded at one point to avoid ground-loops from occurring. In isolation that statement seems to lend itself to having your structural wiring connected up with standard plastic UTP sockets at the wall - but to terminate the shield at the central patch panel.

And in terms of terminating at the patch panel, you can still use UTP panels, providing you attach the shielding to a metal part and run that to a ground (e.g. the earth pin on a standard socket).

You then have only one ground, the bulk of your network cabling (the structural bit) is shielded, and the devices connected use UTP or standard plastic sockets to avoid ground loops.

Is this right?

http://www.ctrlink.com/2006/04/to-shield-or-not-to-s... seems to agree

Sure I posted this in home networking and not fibre - nothing to do with fibre, could a mod move it?

It is now in home networking

Whereas https://www.siemon.com/us/standards/Screened_and_Shi... doesn't

It is a fact that screens and shields offer substantially improved noise immunity compared to unshielded constructions above 30 MHz... even when improperly grounded.

so not far off

The fact is that, left ungrounded, a screen/shield will still substantially attenuate higher frequency signals because of the low-pass filter formed by its resistance, distributed shunt capacitance, and series inductance. The effects of leaving both ends of a foil twisted-pair cable ungrounded can also be verified using the previous experimental method. As shown in figure 8, the coupling between two UTP cables (shown in black) is still a minimum of 20 dB worse than the interaction between two ungrounded F/UTP cables (shown in blue). It should be noted that 20 dB of margin corresponds to 10 times less voltage coupling. Even under worst-case, ungrounded conditions, the UTP cable behaves more like an antenna than the F/UTP cable!

I don't think grounding one end of a shield is especially difficult if you have a centralised location, but if you screw it up, that article seems to suggest it'll still be better.

The method you describe in the opening post sounds like it should work fine in principle (in that you are eliminating any even hypothetical problem in the last patch to the end device).

The follow up suggests it's not as much of a problem (unlike audio / video connections when they become smothered 50Hz hum).

The scenario I do worry about is linking 2 adjacent buildings via FTP cable in case the the buildings grounds end up a different potential - I'd imagine there are cases where a significant shield current could be flowing.

Naturally one would likely avoid that entirely as a by-product linking in fibre anyway.

In reply to a post by prlzx:

The scenario I do worry about is linking 2 adjacent buildings via FTP cable in case the the buildings grounds end up a different potential - I'd imagine there are cases where a significant shield current could be flowing.

Naturally one would likely avoid that entirely as a by-product linking in fibre anyway.

This brings back bad memories of being at school in the 1980s. We had a bunch of BBCs running on Econet - bus topology, two twisted pairs (or a twisted quad, depending on the generation of the cable) with a foil shield and drain wire. Econet has a clock pair and a data pair using CSMA/CD.

It was far from unusual for a thunderstorm to knock out all the line driver chips on a segment, especially the one where the cable was strung between two buildings on a catenary wire. I could change a 75159 line driver chip in about 90 seconds per station - less if the case screws weren't fitted (as was common - we got so used to changing these chips). In one extreme case - possibly a direct hit - we had to replace some melted connectors on the backbone cable.

The real pain with blown Econet line drivers was that the fault was usually manifest when a station was plugged in and switched off - it failed to disconnect from the bus and held the data pair such that no other traffic could pass. Suddenly there'd be a howl of users reporting "Line jammed" errors until the faulty station was disconnected from the network.


You can certainly get all sorts of weird effects from differences in ground potential between buildings. These days, galvanic isolation - and therefore immunity to ground potential differences and electrical storms - is easy to achieve using fibre. Unlike the 1980s, fibre is affordable and easy to deploy.

It's no accident that there's fibre linking our house to the outbuilding where we keep our servers in an air-conditioned room. We may have got away with stringing some external grade Category 5e UTP cable between the buildings, but it's not worth it. One fried switch would more than pay for the cost of the fibre. We did it the easy way, installing a pre-terminated fibre prepared off-site - deploy the fibre, pull the PVC tubing off the cable gland and click the individual LC terminated fibres into an LC patch panel.


The only thing we use external grade Category 5e cable for is telephone.