Hi,
I had some RFI from Ethernet cable. I don’t remember all the details, when
Covid hit I ran a 50’ CAT 5e the length of the house to a new access point and
some RFI started showing up that I could not kill with toroids. I bought a new
50’ CAT 8 cable to replace it, which has shielding for each twisted pair I
believe, and the RFI disappeared. I still have cat 6 for short jumper cables,
but if I was going to do any long runs between rooms, etc., I would use CAT 8.
Just my 2 cents worth.
Joe K1VW
> On Sep 25, 2024, at 11:40 AM, Greg Troxel <gdt@lexort.com> wrote:
>
> I'm about to run Ethernet cable, some new runs and perhaps replacing
> existing old cable, which is Cat 5 CM (not 5E!) that was free to good
> home over 20 years ago (pulls from a commercial data center that decided
> they needed CMP). There will be some outdoor (on the house, not
> planning interbuilding) and some attic. The root switch is AC powered,
> and remote switches, APs, and a few other things are POE-powered. Most
> things are Ubiquiti UniFi.
>
> Before I get to RFI concerns, for outdoors and highly likely attic the
> cable needs to be outdoor rated (UV and water), and code says anything
> inside needs to be rated CMX and <= 0.25", or else CM or better. (CMX
> is a fire rating, and does not mean outdoors, but typically CMX and
> outdoor go together. However some outdoor cable is not even CMX rated.)
> If other than a 1/2-unit dwelling, code would require CMP in plenums,
> CMR for vertical runs, and >= CM for the rest (except for outdoor CMX
> coming in < 50'), and I'm inclined to follow that as a good idea though
> not required. (I wonder if NFPA will change the code with houses being
> much more wired; it seems the "CMX ok" is perhaps about "but you don't
> have much of it".)
>
> Generally the world recommends shielded outdoors, with that being
> omittable if all runs are along the building and perhaps not so long
> outside. This is about lightning (distant more than direct)) and static
> buildup, as I understand it.
>
> With unshielded indoors, I don't really expect a ground reference at the
> remote switches etc. With everything shielded, all the networking
> equipment is basically star tied back to the root switch. So it doesn't
> make a lot of sense to me to run shielded outdoors and unshielded
> indoors, unless I have ground connections at each switch connected to a
> cable that goes out. Most of my switches are intended to be POE powered
> and do connected the shields of shielded cables.
>
> So it seems reasonable approaches would be:
> everything shielded
> everything unshielded not worrying about short outdoor runs
>
>
> Before posting I reviewed RFI-HAM :-) and my summary of that is that one
> needs choking impedance for common-mode, and that WiFI has less
> interfence. There wasn't really any support for shielded Ethernet cable
> being better. But also no support for it being worse.
>
>
> It makes intutive sense that shielded cable would be less trouble, but
> there is the point that all the signals are differential on high-quality
> twisted pair, arguing that it won't matter much. But, having a DC
> ground even if not low impedance at RF is appealing.
>
> It seems shielded costs a bit more. A lifetime buy of 1000' each of CMP
> and outdoor 5e is $123 more, and then connectors are a bit more. It's a
> bit bulkier, and bit more work to terminate. It doesn't seem to have
> any major issues.
>
> I would like to hear
>
> - opinions about the value of shielded vs unshielded Ethernet
> (assuming chokes held constant).
>
> - opinions if sihelded outdoor to non-grounded switch with unshielded
> back to grounded switch makes sense
>
> - experience of anyone who has faced this choice especially their
> rationale before and their hindsight opinion later
>
> - experience of anyone who has done an unshielded to shielded
> conversion and their hindsight opinion.
>
>
> 73 de n1dam
>
> _______________________________________________
> 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
|