Well, Casa Zaba has been at the mercy of my tender d-i-y ministrations.
Compared to two weeks ago, it now has:
* 14 more RJ45/Cat5 wiring ports from loft-patch-panel to old and new studies
* two adjacent holes in the turns-out-to-be-wooden coving in the room
next door to the new study (oops)
* a battle-proven 30A MCB
* slightly less dust and mayhap glass-fibre shards in the loft, moved to
my lungs and working clothes instead
* a stage dressing-room-style row of golfball lights either side of
the bathroom mirror
* three sheets of 9.5mm plasterboard which need to go back to Wickes

So, we're moving my wife's study from the downstairs was-a-garage is-now-a-
den-with-ceiling-sloping-from-3.5m-to-2.2 (conversion done by previous
owners, reasonable but could do with a couple of Veluxen for more light
and better insulation), allowing that to become the teens' den&music-creation
room, to the upstairs bedroom next to my existing study (hers gets the
double window, meaning nicer view). Den has 4 RJ45 ports, and is termination
point for Home Highway line. So, as the new office will need all 3 HH lines
(ISDN for net access, two analogue lines for fax and voice-out) plus the
conventional line for "incoming" calls, and at least 1 honest Ethernet
link needed to have that office's hub/switch talk to the rest of the house,
we'll need 5 right now (unless we use port doublers for the phone lines,
ugh) and we're only mostly sure where the desk and fax machine will be,
so we'll put 6 ports at that end of the wall and 4 the other. That's 10
of the 14: the other 4 being in the old office/den, where there are "only"
4 right now which would be all used carrying the telephony up to the loft
patch panel and down to 4 in the new office, and since I left the draw wire
in from the last time it'll be easy enough to add another 4. After all,
we can't leave the teens without Ethernet, can we...

No drama with the 4 new cables to the den (except being tired resulting
in attaching them to the *end* rather than the middle of the deliberately
double-length draw wire, so no draw wire left for any later pull now, bugrit.
Still, it wasn't *that* hard putting it in in the first place...). No
drama with the 4 to the inner corner of the new study: route is down through
the airing cupboard with the hot water tank, so visble just about all the
way. Cut them too short to run with the others, mind, so had to put in new
cable-tie bases to let them take a more direct route through the loft.

Dropping the other 6 to the corner near the outside wall produced the
fetching holes in the coving of the room next door, though. Y'see, it's
a 20-yr old timber-frame house, pretty well built. I needed a new power
point in that corner, and there's a power point already on the other side
of that wall in my office, so we carefully cut a power-point-back-box-sized
hole in the plasterboard in the new office, find the back of the existing
power point where we expected it to be, sitting on a little noggin. Having
expected to put all 10 data cables down the airing cupboard and to then
run 6 in minitrunking along the top of the skirting to the far corner, with
the ugliness and handcut almost-mitres and possible over-sharp bends in the
route, I thought "Ah! Cavity! If there are no noggins in the way I could
drop the 6 data cables straight down from the loft and avoid the ugly
surface-run trunking!". Probe up with wide steel tape measure shows no
obstructions. Drill small hole up through edge of ceiling, put bit of
stiff wire through to show position in loft. Worm/crawl over to edge of
loft (note: newish house, shallow roof pitch, about 6ins of fibreglass
insulation with waxed paper overlay - 4in batts laid between joists, then
2in thicknesses run in long lengths crossways); so working at edge of
loft is a bitch (especially when you're shaped more like David Brent -
http://www.bbc.co.uk/comedy/theoffic...le_david.shtml -
than Gareth Keenan -
http://www.bbc.co.uk/comedy/theoffic...e_gareth.shtml ).
Consequently, working with the auger to put a couple of 20mm holes through
the plasterboard and expected timber-at-top-of-cavity-whatever-that-bit's-called
with the fibreglass insulation not pulled quite all the way back and the
length-of-extended-forefinger calculated to hit the center of the
cavity... didn't. Oh, it felt quite plausible going down: grind through
plasterboard (cheap auger bit, and resharpenable by hand anyway), hit wood,
good, drill, drill, hmm, pretty thick bit of wood here, ah, now we're
through: will 20mm take 6 Cat5 cables? no, put another hole just next door,
pb, wood, again thickish, aah, now we're through, great. Crawl back out.
Check in new office: good, haven't come through on this side! Check in
room next door (my office). Bugrit. Never knew that coving stuff was
wooden - assumed it was plaster. My, what fetching ragged holes I've
made as we burst through. That'll be fun to patch. Ho hum, crawl back to
edge of roof, pull insulation back properly so I can *see* what I'm
doing - all is clear now, the cramped and unsighted working conditions
meant I was fully *two* fingerlengths along from the trial hole, rather
than just one.

The deities of d-i-y, M*rphy and S*d by name, having had their bit of
fun for now, relent and allow the length of PVC flex dropped from the
now-correctly positioned hole above the cavity to come neatly through and
be immediately fishable through the hole for the new socket. Who says
they're unrelenting? As that nice Mr Cleese has it in Clockwise (?),
"It's not the despair. I can stand the despair. It's the hope"... If they
made *every* job a nightmare, we'd all of us just give up; no, they have
to string us along, lulling us into hope, and *then* hit us with the
slapstick plank of pride and the banana skin of hubris.

Oh, the "battle-proving" of the 30A MCB? Well, you see, the socket in
my office was, as I'd hoped, on the ring; so extending the ring to include
the new socket just meant running the two existing feeds to each of the
old and new sockets, and joining them with a short new bit of cable. What
could be simpler? Feeding that short new bit of cable was a tiny bit more
awkward than planned, because of the noggin on which the existing socket
sat, so I'd pulled it through with a little stiff drawwire. Got impatient
unwrapping the stretched PVC insulating tape joining draw wire to new
cable and sliced through it rather than untaping it. Stupidly, in doing
this I nicked the live conductor (but didn't see that), further along than
the ends I'd strip and wire in to the socket. So, when I folded the
conductors back into the metal back box, where more natural for the
live conductor to choose to bend than at the nick in the insulation?
Now it's resting firmly against the earthed backbox (well-earthed, since
you ask: I've an explicit earthing strap here, not relying merely on
conduct through the mounting screws). So, downstairs to the fusebox,
plug the MCB back in, push the little button - ooh, pretty little blue
flash. Embarassment. Examine back box: hmm, one area's a little darker,
greyer, even (you might say) smokier than the rest. And so's this here
new-cable's bit of Red wire. Aha, an 'ole in its insulation. Bugrit, must've
nicked it when cutting away that temporary securing tape. Wot a silly sod.
And I *had* considered quickly measuring the resistance between the three
terminals on the newly-extended ring, but didn't. 'Course, I did *after*
fixing the fault: stable door, horse, ...

'Course, fiddling about in the backbox was made all the more fun by the
*git* who invented the Adjustable Lug. Y'know, those galvanised boxes
often have one fixed lug to screw the faceplate screws into, and one which
slides up and down, "to allow fine adjustment to the horizontality of the
faceplate". More like, to cause endless frustration in fitting the plate
in the first place. You can *guarantee* that it'll be on the unsighted
side of the work, if there's any awkwardness in access at all; that the
faceplate screws provided will be just long enough to reach the lug, so
they can push it *back* just far enough to stop the threads engaging; and
they'll corrode away in time leading in at least one illustrated case where
a fire resulted from the heating through a not-low-enough-resistance-to-blow-
the-fuse from a live-to-earth fault just like the one I'd caused on this
occasion. All so that first-fix can be a little more sloppy in horizontal
alignmnet. Sigh. One of the circles of hell needs reserving for the genius
who brought us the Adjustable Lug, I feel... Along with the mutton-head
who, when disconnecting the second lighting point in our bedroom wot I also
revived during this session of loft work, decided not to just push the
tape-wrapped cable back into the void, but to cut it off: leaving it about
2cm too short to reach the ceiling rose it used to pick up a switched feed
from, making it necessary to put in a new junction box to take the place
of the wiring in the back of the rose. Deep joy.

Oh, a third group of residents for that circle of hell: whichever bunch
of ninnies who couldn't agree what guage of bolts to settle on for 19inch
rack nuts. The best trick of all is to supply the caged nuts in sizes which
look similar, but which you only realise as you hold the retired-from-HP
hub with one hand and the bought-a-while-ago patch panel with the other,
while screwing the bolts in place with the driver held in the mouth (or
related contortions) that they're in fact *different* gaugues: a moment's
Google suggests at least 4 "standard" sizes - M5, M6 in Metrick land, and
10-32 and 12-24 over in the Land Of The Free. Ah, standards: "the great
thing is there are so many to choose from"... Still, the 14 * 8 * 2 = 224
punchdown connections all seemed to work fine - nothing cross-wired or
unconnected, one two-wires-in-one-slot spotted by visual inspection after
subconcious prompting of the "did I do that one quite right" variety. Must
try to borrow a Pentascanner or moral equivalent from work to check them
over with more than the simple continuity-and-pairing my own tester does;
with some lines being intended for telephony rather than 100BaseT I should
be able to cope if a couple got kinked, bent, folded, mutilated or stapled
by some mysterious force into being marginal as data-grade carriers (though
I was careful to be nice and gentle, smoothing out incipient tangles or
loops, no oversharp turns and the like, only gentle restraint/guidance from
cable ties rather than pull'em tight as hell...)

The bathroom lights? More fun with cables-in-cavities. The right-hand side
one went fine, as I'd cut a hole for the new pushbutton dimmer switch, so
could put a small mirror in to see what went on there (extra uses for the
sighting mirror on a Silva compass ;-) and could fish the new lighting cables
through it and across a little way to the small hole for the RH fitting.
The LH side of the mirror was, of course, two studs further along, and with
no handy large hole it was a case of fishing the new cable through the
2cm? diameter hole I'd just cut for this cable. Previously, Mungo's FAQ
technique of feeding a tape measure through the hole to make a loop in
t'cavity, drop the cable, pull back the tape had worked a treat. This
time - with my son as the tape operative, with me callling instructions
from the loft above - it was tantalisingly close to working, but repeatedly
banjaxed by the tape tangling or other misfortunes. Finally managed to
make enough contact with a partly-bent coathanger to get the jaws of
fine-nosed pliers around the little git (the cable, not the son, who'd
tried his patient best!). The supplementary bonding wire was fun, too:
down by the floor there's an existing bonding point round the sink's
cold water feed, so we'll use that, right? Drop a 2.5mmsq single-core
flex (it'll be Mechanically Protected in the cavity) and away we go. Drop.
Hmm, mirror shows a noggin in the way. But it has a cutaway we could get
through! Strong magnet on a string in the cavity, strong magnet on the
outside, guide it, guide it, bugrit. In retrospect, if you try this, make
the thing inside the cavity be a lump of *unmagnetised* steel: there are
too many attractive nails, plaster-beadings, and the like lurking within
yer average cavity which your inside magnet will find much more worth
clinging to than the weak (inverse-sqaure law applies!) pull of its
complementary magnet some 15-20mm of pboard+skim away. Resolution? Well,
t'other side of this bottom section is only a built-in wardrobe, so we'll
go through the pboard on the other side too, then come in again at the
height of the dimmer switch hole where our friendly draw wire awaits.

'Course, the lights I'd put in - IKEA MUSIK - had no provision for earthing,
being labelled as "Suitable For Bathroom Use" on account of their style and
an unimpressive IP21 rating. The back of these is plastic, and the markings
include the square-in-square logo to signify "double insulated" or Class II
as we're supposed to call it now. But the front and sides are dominated by
a wraparound chromed pressed steel section, which I cannot see as anything
other than an Extraneous Conductive Part. So (voiding any right to take
them back) I drilled a little 'ole towards the top of each, crimped a
ring connector onto some silicone-insulated green-and-yellow single-core
I Just Happen To Have, attached it with a bolt, nut, and shakeproof washer,
and connected the new supply cable's CPC (i.e. "earth wire") to that, while
bonding the CPCs of the two fittings and the supply to the switch up at the
new junction box in the loft just above the bathroom. My friend the OSG tells
me it's OK both to use the CPCs as bonding conductors, and to do the
bonding "close" to the bathroom, so I reckon I'm on the side of the angels
here. (And don't ask how Mechanically Protected the back of the built-in
wardrobe is...)

Oh, and the plasterboard? *Major* contribution from my wife was to find a
local plasterer willing to do one Saturday's work PVA-and-skimming the
relatively subtle but still unwelcome Artex in her new office and the
lad's bedroom next door (to make a day's work of it; Later we'll get the
rest of the house de-Artexed this way too). So, the "screw some plasterboard
over the Artex and fill in the gaps between the boards and at the edges"
job I'd just bought the three sheets of board for has Gone Away, in favour
of a neater and smoother-finish job than I'd've managed. Hallelujah! See,
there are many advantages to living out in yokel-land rather'n in or around
the Bigge Smoke; and finding tradesmen at relatively short notice (fingers
crossed, hasn't actually done the job yet!) is one of them. No, you *can't*
have his contact details; I want him to have time free at relatively short
notice again, thank you very much!)

Aah, the joys of time off work. Lessons learnt? Well, measure-twice, cut-once,
and *use* the measurement; don't rush; experience still allows dumb mistakes
to slip through; knives and mains cables don't mix well; S*d and M*rphy will
always win. Oh, and IKEA *do* stock 25W SES golfball lamps as well as 40W; and
for extra fun, they're in two different opalescences (part-pearl, which is
naff for the fittings I got, and fully-whited-inside, which looks just right)
with no indication on the packaging that I noticed as to which is which. Just
as well it's on the route between home and work; still have to face the
hordes at Sale time, though. Now it's back to saving the Internet for 2004
(with one hand, while painting and fitting-out the new office for my wife
with the other).

Happy new year to all - Stefek

On 5 Jan 2004 15:04:13 GMT, wrote:

Well, Casa Zaba has been at the mercy of my tender d-i-y ministrations.
Compared to two weeks ago, it now has:

excellent story snippage

Thanks for that, Stefek. All of a sudden my trials and tribulations
with the alcove cupboards and related doors in the living room seem
lightweight by comparison )) I was with you all the way on the
cabling and the wiring!
--
cheers,

witchy/binarydinosaurs

snip

Bloody 'ell, thought *I* was having fun ;p

Happy new year to all - Stefek

Same to you and everyone else in D(estroy)-I-Y land

--
Jet
(Five rooms completed in 2003, whoopee )
(3 more + 1 x hall, 3 x landing, loads of stairs, hall and conservatory to
renovate in 2004)

In uk.d-i-y, I wrote:

............................................. Still, the 14 * 8 * 2 = 224
punchdown connections all seemed to work fine - nothing cross-wired or
unconnected, one two-wires-in-one-slot spotted by visual inspection after
subconcious prompting of the "did I do that one quite right" variety. Must
try to borrow a Pentascanner or moral equivalent from work to check them
over with more than the simple continuity-and-pairing my own tester does;
with some lines being intended for telephony rather than 100BaseT I should
be able to cope if a couple got kinked, bent, folded, mutilated or stapled
by some mysterious force into being marginal as data-grade carriers (though
I was careful to be nice and gentle, smoothing out incipient tangles or
loops, no oversharp turns and the like, only gentle restraint/guidance from
cable ties rather than pull'em tight as hell...)

So I managed to prise the Pentascanner from the enveloping clutches of
our resident LAN wiring guru, after promising to take better care of it
than my own most delicate bodyparts, etc. etc. I'm happy (to the point of
smugness) to report that all 28 of the Cat5/5e runs rate a decent PASS
according to the Pentascanner. Because it's only a house, all the runs
are quite short in datacom-cabling terms - all are around the 15m mark.
So, although the powersum NEXT measurements are not massively inspiring,
just "decent enough" (an average of 37 dB, worst one 33.4dB), and the
return loss averages at 16dB with a couple as poor as 12dB, the shortness
of the runs means that there's so little attenuation that the ACR (moral
equivalent of signal-to-noise ratio) is massively overspec. Dunno if
we'll ever run Gigabit (1000BaseT) over these links, but they're certainly
good for 100BaseT.

Results saved away in spreadsheet (the Pentascanner is kind enough to
dump the test results in comma-separated value format; sadly, unless
you load up the proprietary Windows-only software, the columns are not
handily identified - had to decode what-was-what from the nicely-formatted
output of the same data!) for a time when I'm bored enough to want to
analyse them...

Happy days - Stefek

On 15 Jan 2004 21:03:12 GMT, wrote:



So I managed to prise the Pentascanner from the enveloping clutches of
our resident LAN wiring guru, after promising to take better care of it
than my own most delicate bodyparts, etc. etc. I'm happy (to the point of
smugness) to report that all 28 of the Cat5/5e runs rate a decent PASS
according to the Pentascanner. Because it's only a house, all the runs
are quite short in datacom-cabling terms - all are around the 15m mark.
So, although the powersum NEXT measurements are not massively inspiring,
just "decent enough" (an average of 37 dB, worst one 33.4dB), and the
return loss averages at 16dB with a couple as poor as 12dB, the shortness
of the runs means that there's so little attenuation that the ACR (moral
equivalent of signal-to-noise ratio) is massively overspec. Dunno if
we'll ever run Gigabit (1000BaseT) over these links, but they're certainly
good for 100BaseT.

Personally I think it was a fluke.......


Results saved away in spreadsheet (the Pentascanner is kind enough to
dump the test results in comma-separated value format; sadly, unless
you load up the proprietary Windows-only software, the columns are not
handily identified - had to decode what-was-what from the nicely-formatted
output of the same data!) for a time when I'm bored enough to want to
analyse them...

Happy days - Stefek

..andy

To email, substitute .nospam with .gl

In uk.d-i-y, Andy Hall wrote:

Personally I think it was a fluke.......

Hmm. At least I was agilent enough to reach into the corners of the loft.

On 16 Jan 2004 09:23:21 GMT, wrote:

In uk.d-i-y, Andy Hall wrote:

Personally I think it was a fluke.......

Hmm. At least I was agilent enough to reach into the corners of the loft.

very good :-)

..andy

To email, substitute .nospam with .gl

In uk.d-i-y, Andy Hall wrote:

Hmm. At least I was agilent enough to reach into the corners of the loft.

very good :-)

ITYM "very gould". Oh, I think of it as an avocation. Using anything less
would be robin myself. I'd have to scope it out, and detect chronic subtle
problems. Still, at least in a loft you've a good chance of hearing the first
martindale of spring. Thurlby trouble if I don't. And I might hear Ray call
too.

Andy Hall wrote:
On 15 Jan 2004 21:03:12 GMT, wrote:



So I managed to prise the Pentascanner from the enveloping clutches
of our resident LAN wiring guru, after promising to take better care
of it than my own most delicate bodyparts, etc. etc. I'm happy (to
the point of smugness) to report that all 28 of the Cat5/5e runs
rate a decent PASS according to the Pentascanner. Because it's only
a house, all the runs are quite short in datacom-cabling terms - all
are around the 15m mark. So, although the powersum NEXT measurements
are not massively inspiring, just "decent enough" (an average of 37
dB, worst one 33.4dB), and the return loss averages at 16dB with a
couple as poor as 12dB, the shortness of the runs means that there's
so little attenuation that the ACR (moral equivalent of
signal-to-noise ratio) is massively overspec. Dunno if we'll ever
run Gigabit (1000BaseT) over these links, but they're certainly good
for 100BaseT.

Personally I think it was a fluke.......

Yes, a Fluke Pentascanner....

D

wrote in message
...

In uk.d-i-y, Andy Hall wrote:

ITYM "very gould". Oh, I think of it as an avocation. Using anything
less would be robin myself. I'd have to scope it out, and detect
chronic subtle problems. Still, at least in a loft you've a good
chance of hearing the first martindale of spring. Thurlby trouble
if I don't. And I might hear Ray call too.

ObDIY: ... and if you can't make any of this telly 'quipment work, you can
consider yourself a bit of a wayne kerr.

--
Andy