electrical safety on LAN

I have a requirement to introduce a pc into a LAN that is in a seperate
building to the rest of the network and server. I am aware that this is a
candidate for a wireless link but I would nevertheless like to know whether
it is a viable option to carry the network cable to the other building for
this purpose. As I recall there may be an issue with earthing that makes
this unsafe however I do know that both buildings share the same substation
if this makes a difference

Many thanks


Mike Saunders

On Wed, 21 Apr 2004 11:37:21 +0100, in uk.comp.home-networking "Mike
Saunders" <(E-Mail Removed)> strung together this:

>I have a requirement to introduce a pc into a LAN that is in a seperate
>building to the rest of the network and server. I am aware that this is a
>candidate for a wireless link but I would nevertheless like to know whether
>it is a viable option to carry the network cable to the other building for
>this purpose. As I recall there may be an issue with earthing that makes
>this unsafe however I do know that both buildings share the same substation
>if this makes a difference
>
You want a optoisolator to break the earth link between the buildings,
these basically convert the copper wire signal into IR and back again
so there is no physical electrical connection between the two
buildings.
The other thing you probably want to look at is a lightning arrestor,
which basically earths the external run of cable to prevent any spikes
getting as far as the NIC.
--

SJW
A.C.S. Ltd.

Must say optoisolator is a new one for me. This sounds like a job for the
experts

Many thanks


"Lurch" <(E-Mail Removed)> wrote in message
news:4086b4fc.169200877@195.129.110.67...
> On Wed, 21 Apr 2004 11:37:21 +0100, in uk.comp.home-networking "Mike
> Saunders" <(E-Mail Removed)> strung together this:
>
> >I have a requirement to introduce a pc into a LAN that is in a seperate
> >building to the rest of the network and server. I am aware that this is
a
> >candidate for a wireless link but I would nevertheless like to know
whether
> >it is a viable option to carry the network cable to the other building
for
> >this purpose. As I recall there may be an issue with earthing that
makes
> >this unsafe however I do know that both buildings share the same
substation
> >if this makes a difference
> >
> You want a optoisolator to break the earth link between the buildings,
> these basically convert the copper wire signal into IR and back again
> so there is no physical electrical connection between the two
> buildings.
> The other thing you probably want to look at is a lightning arrestor,
> which basically earths the external run of cable to prevent any spikes
> getting as far as the NIC.
> --
>
> SJW
> A.C.S. Ltd.

On Wed, 21 Apr 2004 19:36:51 +0100, in uk.comp.home-networking "Mike
Saunders" <(E-Mail Removed)> strung together this:

>Must say optoisolator is a new one for me. This sounds like a job for the
>experts
>
Yep, it is a fairly simple job, not too difficult to get right but
very easy to get it wrong and render the entire system useless!
As it happens we're not too far away from you, just the sort of thing
we do if you hadn't anyone in mind.
--

SJW
A.C.S. Ltd.

Lurch wrote:
> On Wed, 21 Apr 2004 11:37:21 +0100, in uk.comp.home-networking "Mike
> Saunders" <(E-Mail Removed)> strung together this:
>
>> I have a requirement to introduce a pc into a LAN that is in a
>> seperate building to the rest of the network and server. I am aware
>> that this is a candidate for a wireless link but I would
>> nevertheless like to know whether it is a viable option to carry the
>> network cable to the other building for this purpose. As I recall
>> there may be an issue with earthing that makes this unsafe however
>> I do know that both buildings share the same substation if this
>> makes a difference
>>
> You want a optoisolator to break the earth link between the buildings,
> these basically convert the copper wire signal into IR and back again
> so there is no physical electrical connection between the two
> buildings.

Or a wifi bridge would do the same task...

On Wed, 21 Apr 2004 19:56:11 +0100, in uk.comp.home-networking "Steve
Walker" <spam-(E-Mail Removed)> strung together this:

>Lurch wrote:
>> On Wed, 21 Apr 2004 11:37:21 +0100, in uk.comp.home-networking "Mike
>> Saunders" <(E-Mail Removed)> strung together this:
>>
>>> I have a requirement to introduce a pc into a LAN that is in a
>>> seperate building to the rest of the network and server. I am aware
>>> that this is a candidate for a wireless link but I would
>>> nevertheless like to know whether it is a viable option to carry the
>>> network cable to the other building for this purpose. As I recall
>>> there may be an issue with earthing that makes this unsafe however
>>> I do know that both buildings share the same substation if this
>>> makes a difference
>>>
>> You want a optoisolator to break the earth link between the buildings,
>> these basically convert the copper wire signal into IR and back again
>> so there is no physical electrical connection between the two
>> buildings.
>
>Or a wifi bridge would do the same task...
>
I know the OP said that but wondered _if_ it could be done with a
wired option. Try reading before posting.
--

SJW
A.C.S. Ltd.

On Wed, 21 Apr 2004 19:13:44 GMT, (E-Mail Removed) (Lurch)
wrote:

>On Wed, 21 Apr 2004 19:56:11 +0100, in uk.comp.home-networking "Steve
>Walker" <spam-(E-Mail Removed)> strung together this:
>
>>Lurch wrote:
>>> On Wed, 21 Apr 2004 11:37:21 +0100, in uk.comp.home-networking "Mike
>>> Saunders" <(E-Mail Removed)> strung together this:
>>>
>>>> I have a requirement to introduce a pc into a LAN that is in a
>>>> seperate building to the rest of the network and server. I am aware
>>>> that this is a candidate for a wireless link but I would
>>>> nevertheless like to know whether it is a viable option to carry the
>>>> network cable to the other building for this purpose. As I recall
>>>> there may be an issue with earthing that makes this unsafe however
>>>> I do know that both buildings share the same substation if this
>>>> makes a difference
>>>>
>>> You want a optoisolator to break the earth link between the buildings,
>>> these basically convert the copper wire signal into IR and back again
>>> so there is no physical electrical connection between the two
>>> buildings.

Not a networking issue as such, but I ran a TV aerial coax extension
round the outside of the house from a booster/splitter in one room to
a computer in another. Probably about 70ft in all. There's a divider
half way round -- when both ends are plugged in, and the circuit is
broken at the divider, you can feel the voltage beween the "earthed"
outers. I suspect that between buildings, possibly on different mains
phases (even from the same substation), trouble is likely.

--
Please use the corrected version of the address below for replies.
Replies to the header address will be junked, as will mail from
various domains listed at www.scottsonline.org.uk
regards. Mike Scott Harlow Essex England.(unet -a-t- scottsonline.org.uk)

On Wed, 21 Apr 2004 21:07:19 +0100, in uk.comp.home-networking Mike
Scott <(E-Mail Removed) k> strung
together this:

>Not a networking issue as such, but I ran a TV aerial coax extension
>round the outside of the house from a booster/splitter in one room to
>a computer in another. Probably about 70ft in all. There's a divider
>half way round -- when both ends are plugged in, and the circuit is
>broken at the divider, you can feel the voltage beween the "earthed"
>outers. I suspect that between buildings, possibly on different mains
>phases (even from the same substation), trouble is likely.
>
Yep, that's pretty much the same as with the network cabling. Those
little voltages can be extremely destructive if\when they appear in
the wrong place. They can come from external cabling inducing small
currents into the cable or from the appliances they are connected to
at each end.
You can get a ground loop isolator for co-ax as well, I would
recommend fitting one as your co-ax is connected to the PC.
--

SJW
A.C.S. Ltd.

"Lurch" <(E-Mail Removed)> wrote in message
news:4086d605.177657087@195.129.110.67...
> On Wed, 21 Apr 2004 21:07:19 +0100, in uk.comp.home-networking Mike
> Scott <(E-Mail Removed) k> strung
> together this:
>
> >Not a networking issue as such, but I ran a TV aerial coax extension
> >round the outside of the house from a booster/splitter in one room to
> >a computer in another. Probably about 70ft in all. There's a divider
> >half way round -- when both ends are plugged in, and the circuit is
> >broken at the divider, you can feel the voltage beween the "earthed"
> >outers. I suspect that between buildings, possibly on different mains
> >phases (even from the same substation), trouble is likely.
> >
> Yep, that's pretty much the same as with the network cabling. Those
> little voltages can be extremely destructive if\when they appear in
> the wrong place. They can come from external cabling inducing small
> currents into the cable or from the appliances they are connected to
> at each end.
> You can get a ground loop isolator for co-ax as well, I would
> recommend fitting one as your co-ax is connected to the PC.

Also, the "ground" on many modern equipment is not actually connected to
mains earth. I discovered this with my new TV: when I was connecting the
aerial from the other aerial socket in my house to my computer, I got a
noticeable shock. I measured about 200 VAC between the aerial cable earth
and the case of the PC (which *is* earthed) - which shocked me in both
senses of the word!

By unplugging items in turn, it turned out that the outer connectors of my
TV's and VCR's aerial sockets are not connected to mains earth (neither
appliance even has an earth wire in the mains cable) and the aerial cable
runs within the house were not earthed either.

Panasonic confirmed that this was "standard modern practice" and that there
was a very high resistance between mains and the outside, to limit the
current and drop most of the voltage across it, but across a resistance of a
few hundred kilohms (typical across-the-body resistance) there was still
about 50 V present which is enough to give you a definite tingle.


What about connecting equipment that's in the same room but on different
mains phases? I remember our lab at work had several benches of computers,
with adjacent ones being on different phases - and huge signs warning us not
to connect equipment together from different benches. This really applied to
serial and parallel connections - strangely LAN connections weren't deemed
to be a problem! Spot the logic in that.

On Thu, 22 Apr 2004 11:53:53 GMT, "Martin Underwood" <(E-Mail Removed)>
wrote:
....
>Panasonic confirmed that this was "standard modern practice" and that there
>was a very high resistance between mains and the outside, to limit the
>current and drop most of the voltage across it, but across a resistance of a
>few hundred kilohms (typical across-the-body resistance) there was still
>about 50 V present which is enough to give you a definite tingle.

I'd guess then that's about what I'm seeing. Come to think of it, I
guess the TV does have a twin-wire mains lead. I suppose it's safe to
assume the PC is earthed?

>What about connecting equipment that's in the same room but on different
>mains phases? I remember our lab at work had several benches of computers,
>with adjacent ones being on different phases - and huge signs warning us not
>to connect equipment together from different benches. This really applied to
>serial and parallel connections - strangely LAN connections weren't deemed
>to be a problem! Spot the logic in that.

Even on the same phase you can get interesting effects. In our lab,
being in the digital speech business, we'd taken care to get the mains
"right" (to avoid hum loops) in spite of Site Services' electricians'
efforts to the contrary :-) Even so, we could draw, iirc, a couple of
milliamps of AC through an AVO between earths on different benches.

I'm not up to speed on current h/ware (I *still* think of ethernet as
amazingly thick coax, vampire taps and bulky drop cables :-) ), but
don't rj45 ethernet sockets connect via an isolating inductive coupler
anyway? In which case earth issues would be more or less moot.

--
Please use the corrected version of the address below for replies.
Replies to the header address will be junked, as will mail from
various domains listed at www.scottsonline.org.uk
regards. Mike Scott Harlow Essex England.(unet -a-t- scottsonline.org.uk)