Ethernet being differential means (among other things) that there shouldn't be interference in signal reception due to ground potential difference. But when you look at ethernet transformer circuits, the transformer center-tap - which is in the middle of the differential and therefore connected / at similar potential at both ends of the cable - is only one small capacitor away from circuit ground. So if that cap shorts out at both ends (surge?), then the circuit grounds of the devices at both ends become connected. If those devices' grounds are bonded to their buildings which have different ground potential, interesting times happen.
Your conclusion is still the best solution. I suspect that VDSL circuits are better designed to survive cases such as the above. On Sat, Feb 4, 2017 at 2:45 PM, Bill Prince <[email protected]> wrote: > Depending on the distance the potential between two buildings is something > you want to avoid "tying" them together. Ethernet signals are differential, > so should be immune to the difference in ground potential. So if you're > using a shielded cable, ground at one end only (for noise suppression). > > Still, a fiber link would be preferable, or at least a VDSL link. > > > bp > <part15sbs{at}gmail{dot}com> > > > On 2/4/2017 12:39 PM, Colin Stanners wrote: > > I'm also looking at this type of problem for a customer, a few buildings > far apart with long ethernet runs between them, on that network ports and > devices sometimes die, I believe due to the varying voltage potential > difference between the buildings. There are some grounded surge supressors > there that may be helping - but not much. Most SSes only trigger at higher, > lightning-level voltages. > > Having the long ethernet cable between buildings be shielded (and > correctly grounded at both ends) can help too, but if it ends up equalizing > the potential between both buildings, you could have a few amps flowing > between the buildings over that shield, which would be very bad indeed. A > heavy-gauge ground cable between the buildings may be the best option but > would be expensive. > > Another option is creating an optoisolator at one end with two > back-to-back fiber-optical converters. As long as the one wired to the > further building doesn't couple Ethernet "ground" to power ground inside > itself. > > > On Fri, Feb 3, 2017 at 12:33 PM, Adam Moffett <[email protected]> wrote: > >> A grounded surge suppressor on each end will help with that. >> The ground can have a different electrical potential at two points that >> far apart. Then you get a current through your devices trying to equalize >> it. >> >> >> ------ Original Message ------ >> From: "That One Guy /sarcasm" <[email protected]> >> To: "[email protected]" <[email protected]> >> Sent: 2/3/2017 1:14:02 PM >> Subject: Re: [AFMUG] PoE Cat5/6 cable distances >> >> We have a customer who trenched 500 foot of cat5 to an outbuilding, >> routers pop for "some reason" >> I wonder if an extender would help with that >> >> On Fri, Feb 3, 2017 at 11:48 AM, Ken Hohhof <[email protected]> wrote: >> >>> Are you trying to exceed 100 meters? Is there a good reason not to use >>> fiber + DC power? >>> >>> -----Original Message----- >>> From: Af [mailto:[email protected]] On Behalf Of Adam Moffett >>> Sent: Friday, February 3, 2017 11:47 AM >>> To: [email protected] >>> Subject: Re: [AFMUG] PoE Cat5/6 cable distances >>> >>> Google for ethernet extenders with PoE. >>> They exist. Never used one. I think up to 1000' >>> >>> >>> ------ Original Message ------ >>> From: "Dev" <[email protected]> >>> To: [email protected] >>> Sent: 2/3/2017 12:45:40 PM >>> Subject: [AFMUG] PoE Cat5/6 cable distances >>> >>> >Is there any way to extend the cable length over a few hundred feet and >>> >still get reliable power/data? >>> >>> >>> >>> >> >> >> -- >> If you only see yourself as part of the team but you don't see your team >> as part of yourself you have already failed as part of the team. >> >> > >
