Hi folks,

in a rushed piece of diy, i wired a lamp cord switch incorrectly which
resulted in a short between live and neutral. i plugged the lamp (it
had a 2 pin fuseless plug) into a mk twin wall socket. i had tested
it by flipping the switch on the wall socket. A big flash occurred
behind the switch and the house lights went out. Upon dissasembly of
the wall socket I found the switch contacts inside the socket fused
together.

Now what i found strange was that besides the 32A CB for the wall
sockets, my RLCB and my main 50A CB tripped too. Why should more than
1 CB trip for a given incident? Isnt the wall socket circuit CB
supposed to trip alone and keep the rest of the house from plunging
into darkness? Thanks

Sam

In article ,
(Sam) writes:
Hi folks,

in a rushed piece of diy, i wired a lamp cord switch incorrectly which
resulted in a short between live and neutral. i plugged the lamp (it
had a 2 pin fuseless plug) into a mk twin wall socket. i had tested
it by flipping the switch on the wall socket. A big flash occurred
behind the switch and the house lights went out. Upon dissasembly of
the wall socket I found the switch contacts inside the socket fused
together.

Now what i found strange was that besides the 32A CB for the wall
sockets, my RLCB and my main 50A CB tripped too. Why should more than
1 CB trip for a given incident? Isnt the wall socket circuit CB
supposed to trip alone and keep the rest of the house from plunging
into darkness? Thanks

There are potentially several issues here.

First, it's worth describing the difference between a fault current
and an overload current. A fault current is what happens where there
is a short circuit, and until a protective device (fuse/MCB) trips,
the current is limited only by the impedance (~resistance) of the
supply conductors. The current flowing during a fault current can
reach 100's or even 1000's of Amps in domestic premises. An overload
current is a situation where there's nothing wrong with the circuit,
it is simply being used to carry more current than it was designed
for or can safely carry. The incident you created was a fault current.

Now the next issue is that MCB's and fuses can only break the current
flow at two points in a mains cycle, that's where the current passes
through zero. Current doesn't stop flowing just because two metal
contacts come apart, particularly at high current flow. An arc will
form which will extinguish when the current flow momentarily stops
as it passes through zero. (There are other ways to extinguish the
arc, but not used at normal mains voltage in domestic situations.)

So, let's suppose when you shorted out the live and neutral that the
supply impedance was, say, 1/4 ohm. This means 1000A flowed. Now
this exceeds the fault current trip rating of all your breakers
most likely, so they will all try and break the current flow. None
will actually manage to do so until the next zero crossing point,
but this is why they all tripped -- they tripped when the fault
current happened, but continue to conduct until the next zero
crossing point.

Generally, for fault currents, it's difficult to get discrimination
between daisy-chained MCBs because they all trip instantly but
continue conducting until the next zero crossing point. Fuses take
longer to trip, so it's easier to get better fault current
discrimination, although still possible to get multiple fuses
blowing. Unfortunately, it is difficult to design MCB's to take just
a little bit longer to trip on a fault current (unless you go for
purely thermal breakers like the US use). For overload currents,
this is not generally a problem and discrimination works -- the
lowest current breaker trips saving the others.

--
Andrew Gabriel

Thanks Andrew.i didnt know anything about the zero crossing point.
thanks for taking the time to reply.

Sam

(Andrew Gabriel) wrote in message ...
In article ,
(Sam) writes:
Hi folks,

in a rushed piece of diy, i wired a lamp cord switch incorrectly which
resulted in a short between live and neutral. i plugged the lamp (it
had a 2 pin fuseless plug) into a mk twin wall socket. i had tested
it by flipping the switch on the wall socket. A big flash occurred
behind the switch and the house lights went out. Upon dissasembly of
the wall socket I found the switch contacts inside the socket fused
together.

Now what i found strange was that besides the 32A CB for the wall
sockets, my RLCB and my main 50A CB tripped too. Why should more than
1 CB trip for a given incident? Isnt the wall socket circuit CB
supposed to trip alone and keep the rest of the house from plunging
into darkness? Thanks

There are potentially several issues here.

First, it's worth describing the difference between a fault current
and an overload current. A fault current is what happens where there
is a short circuit, and until a protective device (fuse/MCB) trips,
the current is limited only by the impedance (~resistance) of the
supply conductors. The current flowing during a fault current can
reach 100's or even 1000's of Amps in domestic premises. An overload
current is a situation where there's nothing wrong with the circuit,
it is simply being used to carry more current than it was designed
for or can safely carry. The incident you created was a fault current.

Now the next issue is that MCB's and fuses can only break the current
flow at two points in a mains cycle, that's where the current passes
through zero. Current doesn't stop flowing just because two metal
contacts come apart, particularly at high current flow. An arc will
form which will extinguish when the current flow momentarily stops
as it passes through zero. (There are other ways to extinguish the
arc, but not used at normal mains voltage in domestic situations.)

So, let's suppose when you shorted out the live and neutral that the
supply impedance was, say, 1/4 ohm. This means 1000A flowed. Now
this exceeds the fault current trip rating of all your breakers
most likely, so they will all try and break the current flow. None
will actually manage to do so until the next zero crossing point,
but this is why they all tripped -- they tripped when the fault
current happened, but continue to conduct until the next zero
crossing point.

Generally, for fault currents, it's difficult to get discrimination
between daisy-chained MCBs because they all trip instantly but
continue conducting until the next zero crossing point. Fuses take
longer to trip, so it's easier to get better fault current
discrimination, although still possible to get multiple fuses
blowing. Unfortunately, it is difficult to design MCB's to take just
a little bit longer to trip on a fault current (unless you go for
purely thermal breakers like the US use). For overload currents,
this is not generally a problem and discrimination works -- the
lowest current breaker trips saving the others.

"Andrew Gabriel" wrote in message ...

Now the next issue is that MCB's and fuses can only break the current
flow at two points in a mains cycle, that's where the current passes
through zero. Current doesn't stop flowing just because two metal
contacts come apart, particularly at high current flow. An arc will
form which will extinguish when the current flow momentarily stops
as it passes through zero. (There are other ways to extinguish the
arc, but not used at normal mains voltage in domestic situations.)

Andrew, I think you'll find that all modern MCBs and most HBC fuses are
so-called 'current-limiting' designs which do interrupt arc current well
before it reaches its prospective peak. Compare the manufacturer's
published I^2*t let-through figures with the I^2 integral over a full
half-cycle at the full rated breaking capacity.

MK technical supplement: "The arc drawn beteen the contacts is moved by
magnetic forces into the multiple plate arc chamber where the arc is split,
rapidly cooled and extinguished, the total operating time of the MCB is
between 3 and 5 milliseconds."

Electrium-Wylex catalogue: "The high speed current limiting action ensures
that the MCB operates before the full prospective fault current is allowed
to develop. Under fault conditions, damage can be sustained to the
installation and associated equipment due to the amount of energy that
passes before the current is completely interrupted. The total energy
let-through depends on the value of current and the time for which it flows,
and is denoted by the symbol I2t. The high speed current limiting action of
MCBs ensures that the energy let-through and any subsequent damage is
minimised. This reduced energy let-through assists greatly with both back-up
and discrimination considerations."

But you still don't get discrimination on a high level s/c fault of course -
the trip mechanisms all fire together, then it's a race to see which
arc-quench chamber cuts off the current first.

--
Andy

In message ,
(Sam) wrote:

Hi folks,


Mr. Gabriel has already given a comprehensive reply, but I was a little
worried by this bit:

i plugged the lamp (it
had a 2 pin fuseless plug) into a mk twin wall socket.

Do you mean the sort of plug found on shavers? Did you use an adapter
because if so, this should have been fused too - at 1 amp. If not, how
did you fit the plug in? Is the lamp even suitable for UK use?

Hwyl!

M.

--
Martin Angove: http://www.tridwr.demon.co.uk/
Two free issues: http://www.livtech.co.uk/ Living With Technology
.... I'm not schizophrenic. It's this guy beside me!

In article ,
"Andy Wade" writes:
"Andrew Gabriel" wrote in message ...

Now the next issue is that MCB's and fuses can only break the current
flow at two points in a mains cycle, that's where the current passes
through zero. Current doesn't stop flowing just because two metal
contacts come apart, particularly at high current flow. An arc will
form which will extinguish when the current flow momentarily stops
as it passes through zero. (There are other ways to extinguish the
arc, but not used at normal mains voltage in domestic situations.)

Andrew, I think you'll find that all modern MCBs and most HBC fuses are
so-called 'current-limiting' designs which do interrupt arc current well
before it reaches its prospective peak. Compare the manufacturer's
published I^2*t let-through figures with the I^2 integral over a full
half-cycle at the full rated breaking capacity.

Oh, OK. Looking at the tripping times for MCB's they are allowed
to be up to 1/2 a mains cycle regardless of fault current flowing.
I guess there's certainly no harm if they do better.

I dissected a larger CB some years ago (I think it was 200A, and
I don't know what it came out of). That had an air chamber and
plunger to extinguish the arc with an air blast. Initially I
thought it was a damper, but it didn't damp the trip mechanism.
Then I realised it generated a blast of air through the contacts
and through a mica honeycomb, and the mica did look like it might
have had an arc blown into it -- some vapourised metal in it.

--
Andrew Gabriel

"Andrew Gabriel" wrote in message ...

I dissected a larger CB some years ago (I think it was 200A, and
I don't know what it came out of). That had an air chamber and
plunger to extinguish the arc with an air blast. Initially I
thought it was a damper, but it didn't damp the trip mechanism.
Then I realised it generated a blast of air through the contacts
and through a mica honeycomb, and the mica did look like it might
have had an arc blown into it -- some vapourised metal in it.

The designers must have a lot of fun testing things like that during
development...

--
Andy

On 22 Mar 2004 22:35:36 GMT, (Andrew
Gabriel) wrote:

I dissected a larger CB some years ago (I think it was 200A, and
I don't know what it came out of). That had an air chamber and
plunger to extinguish the arc with an air blast. Initially I
thought it was a damper, but it didn't damp the trip mechanism.
Then I realised it generated a blast of air through the contacts
and through a mica honeycomb, and the mica did look like it might
have had an arc blown into it -- some vapourised metal in it.

Interesting - I always wondered how very large currents were switched
on the power grid, even an electric kettle can produce a spark when
switched on - so multiplying that by the equivalent of several tens of
thousands electric kettles must mean the capacity for a very large
spark as contacts meet and separate.

I'd be interested in hearing more about this topic

PoP

---
If you need to contact me please submit your comments
via the web form at http://www.anyoldtripe.co.uk. I'll
probably still ignore you but at least I'll get the
message.....

In article ,
PoP writes:
On 22 Mar 2004 22:35:36 GMT, (Andrew
Gabriel) wrote:

Interesting - I always wondered how very large currents were switched
on the power grid, even an electric kettle can produce a spark when
switched on - so multiplying that by the equivalent of several tens of
thousands electric kettles must mean the capacity for a very large
spark as contacts meet and separate.

With difficulty, see:

http://phil.ipal.org/electric/switch-arc-1.mpg
http://phil.ipal.org/electric/switch-arc-2.mpg

Actually, for dealing with fault currents (which isn't what these
switches are meant to do), air-blast and oil-blast breakers are
used. Air-blast is similar to what I dissected but much bigger.
Oil-blast is where the contacts are under oil -- the arc is quenched
by the oil but also any oil vapour pressure generated by the arc is
used to speed up the separation of the contacts. In both cases, the
effect is quite explosive. I may be somewhat out of date on current
technology though.

--
Andrew Gabriel

In article ,
PoP wrote:

I'd be interested in hearing more about this topic

There's also 'magnetic blowout', where a strong
local magnet is used to deflect the arc sideways
across the opening contacts, and so increase the
length of the arc until it cannot sustain itself.
Used more often when switching DC voltages.

--
Tony Williams.

On 23 Mar 2004 11:10:11 GMT, (Andrew
Gabriel) wrote:

With difficulty, see:

http://phil.ipal.org/electric/switch-arc-1.mpg
http://phil.ipal.org/electric/switch-arc-2.mpg

Fcuk me! Thanks for that - very interesting!

So whilst we are on the subject of switching power lines, how do
different generating stations get themselves syncronised? I presume
two genny's are pumping power into the national grid at the same time,
and thus must be properly syncronised or else.....

PoP

---
If you need to contact me please submit your comments
via the web form at http://www.anyoldtripe.co.uk. I'll
probably still ignore you but at least I'll get the
message.....

In article ,
PoP writes:
On 23 Mar 2004 11:10:11 GMT, (Andrew
Gabriel) wrote:

With difficulty, see:

http://phil.ipal.org/electric/switch-arc-1.mpg
http://phil.ipal.org/electric/switch-arc-2.mpg

Fcuk me!

I'm sure that's roughly what the guy walking away says just at the
end of the second clip, but I can't quite make out his exact words...

Thanks for that - very interesting!

There's another mpeg there of a substation transformer arcing away
and then blowing up...
http://phil.ipal.org/electric/transformer-explosion.mpg

So whilst we are on the subject of switching power lines, how do
different generating stations get themselves syncronised? I presume
two genny's are pumping power into the national grid at the same time,
and thus must be properly syncronised or else.....

I made some reference to it in this article in another newsgroup:
http://www.google.com/groups?selm=b7...net.uk.sun.com

--
Andrew Gabriel

So whilst we are on the subject of switching power lines, how do
different generating stations get themselves syncronised? I presume
two genny's are pumping power into the national grid at the same time,
and thus must be properly syncronised or else.....

IANAEE[*] but you are right that the whole grid must be synchronised or
Bad Things(tm) will happen. When a genset is brought online, it is first
spun up to speed and then synchronised with the grid by carefully
monitoring its output compared to the grid until there is no phase
variance. It is then switched into the grid. I believe this used to be
done manually but I am sure it all happens automatically now.

Generators are synchronous devices so once they are switched in to the
grid they stay synchronised. Any tendency for a particular genset to lose
synchronisation (perhaps caused by excessive load) would cause a breaker
to trip.

When one section of the grid becomes isolated from another then both parts
can lose synchronisation and that is why it can take some time to bring
everything back online.



* I Am Not An Electrical Engineer

--
Alistair Riddell - BOFH
Microsoft - because god hates us

Alistair Riddell wrote:
So whilst we are on the subject of switching power lines, how do
different generating stations get themselves syncronised? I presume
two genny's are pumping power into the national grid at the same time,
and thus must be properly syncronised or else.....

IANAEE[*] but you are right that the whole grid must be synchronised or
Bad Things(tm) will happen. When a genset is brought online, it is first
spun up to speed and then synchronised with the grid by carefully
monitoring its output compared to the grid until there is no phase
variance. It is then switched into the grid. I believe this used to be
done manually but I am sure it all happens automatically now.

I remember watching them actually do this many years ago when on a
school visit to a power station, there was a big meter indicating the
phase difference and when it reached zero the generator went on line.

(IAAEE)

--
Chris Green