On Sat, 02 Jun 2018 21:19:47 -0700, OFWW
wrote:
On Sat, 02 Jun 2018 21:09:48 -0400, J. Clarke
wrote:
On Sat, 02 Jun 2018 17:37:45 -0700, OFWW
wrote:
On Sat, 2 Jun 2018 16:26:12 -0500, -MIKE-
wrote:
On 6/2/18 10:11 AM, Puckdropper wrote:
-MIKE- wrote in news
Funnier part... finding this pipe under the pad distracted me so much
that I forgot that I had already figured all this out. :-)
I can't use that path because it I would have to take a hard right
turn and then another left to go around the end of the septic system's
leach field. I don't like the prospect of pulling #6 through those
bends on a 100'+ run.
On the far edge of the driveway is a row of 80ft Poplars. Because I
don't want to dig through their major roots, I can't use a ditch-witch
to trench that path.
At the corner of the garage where I intended to start the underground
conduit, there is an expansion joint in the parking area pad with
asphalt expansion joint filler. The path along that joint is far
enough away from the leach bed and far enough from the tree roots that
I can go straight back along that expansion joint.
All I have to do is rent a concrete saw and make one cut a few inches
from the existing expansion joint and then fill it back in with
Quickcrete when I'm done laying the conduit.
That will be easier than hand digging a trench around 3-4" tree roots
and trying to weave the conduit over and under them.
My most recent wiring project was running outdoor rated CAT6 out to the
garage. You're already digging for one set of cable, might it be worth
digging for another? (Cat6 is easy to terminate, just use a punch down
connector and a decent punch tool.)
You can't run network cable close to power cable, though, unless you take
certain precautions. Parallel runs are a bad thing, but if you must go
close to power cables you can enclose the cable in a grounded pipe. I
didn't run in to these problems with my cable run, so I didn't research
them further.
Puckdropper
My buddy is an IT guru and he told me to run CAT10 with the AC and I'd
be fine.
I am Cisco certified, it is not recommended.
The important issue is what code says and I believe that code says
that data cable must be in a separate conduit from power cable.
The crosstalk issue, while real, is much overblown--AC is 60Hz,
Ethernet today is 100-250 Mhz. Any 250 Mhz transcierver that can't
reject 60Hz is crap. The overvoltage issue is also overblown with
Ethernet--modern Ethernet is transformer-coupled and in any case the
actual risk if you blow both ends is under a hundred bucks worth of
hardware.
Have you ever seen someone zapped with high voltage when they touched
their computer case? Yes, I know the in most cases the NIC will act as
a fuse, but not always.
If the NIC acts as a fuse, the NIC was a piece of **** to begin with.
What part of "transformer coupled" are you having trouble with? Unless
someone has screwed up there is NO contact between a twisted pair
Ethernet cable and any part of the computer case, directly or
indirectly.
Getting zapped by a floating ground with Token Ring or Arcnet or
coaxial Ethernet are more likely occurrances--all of those use
shielded cable with the shields grounded to the chassis. But those
have all been dead for a couple of decades now.
It was more of an issue with phones--nobody wanted to answer the phone
and get a load of high voltage for his trouble.
The big issue though is that somewhere down the road some moron is
going to assume that there is only data cable in the conduit, run a
fish tape down it with the power on, zap himself, and possibly burn
down the building.
In the early 1980s the Hamilton-Standard factory in Windsor Locks CT
(roughly a million square feet, roughly 10,000 employees) was shut
down for a day because somebody tried to pull a phone line through a
power conduit. He got himself electrocuted (he survived mostly due to
the fast action of one of the engineers) and set the building on fire.
Part of the factory was shut down for a week due to no power while the
conduit was replaced and the wiring pulled and tested, this time sans
data cables.
This is the big reason to keep them isolated.
Yes, it is the biggest reason, but when you get network speeds running
really slow because of the crap interference, and this being a daily
issue it is quite costly.
Quite costly to some guy running cable from his house to his garage?
OK, tell us exactly how much it costs him, in dollars, to be getting
500 Mb/sec instead 1000 Mb/sec.
I've seen T1 data lines running at 2mbits because of it, and a db
backup that would take multiple hours as opposed to minutes because of
it, add that to normal network traffic and it is ridiculous.
You just blew your cred. T1 is rated for 1.544 Mb/sec. If you were
getting 2 then it was exceeding its rated performance and you have
nothing to complain about.
In any case, nobody is going to be running a T1 from their house to
their garage on a CAT6 cable. What's going to be running on it, in
2018, is gigabit Ethernet.
In one job I did a log of the bit rates over a period of time to show
consistency of the loads and loss of data speed verses what a properly
installed data line would do, and they ended up not only paying the
money to replace the entire line but upgraded the speed so that there
was no negative impact whatever was happening.
That's nice. What does it have to do with some guy running a wire
from his house to his garage?
In that case they thought they could not afford the cost of
replacement, and over a period of years they just got used to it and
considered in "normal" for that site. So I basically invested my own
time to do this and show them what their real costs were. After that
their eyes were opened and they started looking at all their remote
sites to see if they were getting what they were paying for.
I got nothing out of it, but thanks from the employee's that were
effected. That was good enough for me.
Well that's all well and good. So how much measured degradtation have
you experiences in wires run from people's houses to people's garages
carrying gigabit?