On 21/11/2017 11:29, whisky-dave wrote:
On Monday, 20 November 2017 19:03:54 UTC, John Rumm wrote:
Its much the same situation with a fuse. Both will permit small
overloads for a long duration. In some cases (much depending on
the installation method used for the cable) even that may result in
cable damage, or at the very least premature ageing.
Yes I know, but what I don't know is whether a 32A MCB would be used
with cable that the MCB is not up to protecting.
For a general purpose ring circuit, the short answer is it should offer
both fault and overload protection. That means the cable in each leg (as
installed) has at least 21A of current carrying capacity, that loads are
not disproportionately bundled right at the end of the ring, and that
the overall loop impedance at the furthest reach of the circuit is low
enough to ensure "instant" disconnection in the event of a fault.
Isn't this why you
shouldn't but a 13amp fuse in a lead that has 3amp flex ?
If all the fuse is doing is providing fault protection (i.e. any
portable appliance made in the last few decades) then it will usually
remain "safe" even with the wrong fuse. (flexes on modern appliances are
usually sized / length limited to ensure they are fault protected on
your typical Euro style 16A circuit with no additional fusing).
In a simialr way that a basic fuse is meant to protect the cable
NOT the equipment.
For the one at the origin of a circuit, or in a plug yes.
(equipment may have additional internal fuses for self protection
though)
for the equipment not the cable. MCBs aren't there to protect the
equipment, so what are they for. well a short circuit is the most
likely reason for a trip or rather a 'faulty' mains adapted which has
tripped it in the past.
For a general purpose socket circuit, then the MCB will serve two
purposes. Protecting against faults, and protecting against excessive
overload.
The one it *must* provide is the fault protection. In some circumstances
the overload protection can be provided elsewhere and/or by other means
(the most obvious example of which is a spur from a ring - where the
maximum current carrying capacity of the spur cable is less than the
nominal trip of the MCB - the overload protection then comes from the
limitation to only a single socket (unfused spur). It will still
adequately offer fault protection though)
Fault currents however are a different class of fault. Here
you have something causing a short circuit, and the current
that flows can be 100s or 1000s of amps. This results in very
rapid heating of the circuit wires (in some cases even
explosive heating)
called a fuse yes.
I think its generally accepted that if you circuit wires vaporise
in the event of a fault you can consider your circuit protective
devices were inadequate (or at least the operating characteristics
of the circuit was so far from ideal, that the CPDs are were
operating out of spec)
Yes that is what I would have thought, if the wires get damaged then
the MCB isn't doing it's job maybe that's why they come in differnt
amp ratings.
Indeed.
Yes I know I buy then, Quick blow, anti surge, time delay,
semi-delay, 'normal' I'm just glad I don't have to worry about male
and female and LGBTQ versions. Don't seem to nhave those options with
MCBs
You have a similar choice (at least for the larger loads): Common
nominal ratings of 3, 6, 10, 16, 20, 32, 40, 45, 50, 63 (and possibly
others depending on range and brand)
Three different fault / inrush characteristics: Types B, C, & D
And often a range of maximum breaking currents, typically 6kA, and 10KA,
but again there are others. (those plug in wylex 3036 rewireable
replacements often only do 3kA)
Is there any likelihood that your combination of loads will have
exceeded 100A?
No, we no longer have a power lab, labs here. well they are all done
on the ELVIS system now or which we have about 30 in use.
http://www.ni.com/en-gb/shop/select/...ab-workstation
Yup, cute, but not really "power electrics" is it? ;-)
(I recall an ex Marconi college engineer lamenting the lack of exposure
to things over 5V by most of the current generation of new engineers -
he used to like demoing drawing an arc a couple of metres long from the
output of a high power transmitter!)
If the answer is no, then you did not trip the fault current
detection mechanism of the MCB - since that is the minimum required
current for that to happen (and indeed it would still be in spec if
it required 160A to trip using its fault current mechanism)
I don't know what tripped it, I think switching tripped it as the
'2KW' heaters went from 2KW down to 700W then back up to 1.6KW there
were 5 of them. Perhaps it was the 60W soldering iron that was the
'last straw for the camel' But at least we know the MCB actually
tripped. Previously we had RCD and I used a 10K resistor between
earth and live in a plug and used to go around testing the ciruits
once a month. The H&S got involved and I had to stop testing and if
the trips did trip, we had to call maintaince who would then come and
inspect & corect a few hours later, so we don't tell them now, unless
something really weird starts happening.
Measure what ?
You made the claim that you were not drawing anything close to
140A. I agree with you, you weren't. I suggest that if however
you were to drive a nail through one of the circuit cables
(i.e. to introduce a fault) and then measure the current draw,
you will see a *significantly* larger current - hopefully only
briefly.
along with sonme sparks perhaps, but I don't see how this would
make a differnce because if we had ZERO currunt draw and then put
the nail through to make a short circuit the MCB would have most
likely tripped irrespective of the current already flowing. If
anyhting it;s make tripping slightly faster NOT slower.
If your fault current is not high enough to trip the magnetic
response of the MCB,
What current is that then ? Is 32 amp OK and 33 a fault. ?
For a B32 MCB the minimum fault current to be *sure* of getting an
instant trip would be the 5x In rating, or 160A
Its the one in the table on the RHS of the graph:
http://wiki.diyfaq.org.uk/images/d/d...e-MCBTypeB.png
then it will still trip, but it will have to do so using the
thermal mechanism, and this may react *significantly* more slowly
- especially if it was not already right on the boundary of
tripping (which you can't assume - a fault can happen at any
time).
One can use the adiabatic equation to assess the effects on the
cables.
Only if you know the cables used.
You can use the equation to indicate the minimum size of cable required.
If those used are equal or greater, then you are happy. If they are
smaller, you have a problem.
(I would be very surprised if anyone would have installed undersized
cables initially - generally if you install one of the standard circuits
(say 2.5mm^2 ring for a 32A protected ring) all will be fine unless you
have excessive de-rating factors to take into account.
Let's say you have a circuit wired in 2.5mm^2 T&E - that means
your smallest conductor is the pair of CPCs, totalling 3mm^2.
Let's say you have a fault current of 200A, and we can assume the
MCB will open the circuit within 0.1 secs. We have a minimum
conductor CSA of:
MinCSA = sqrt( I^2 x t ) / k
(K will be 115 for PVC insulated cable)
So you get sqrt( 200^2 x 0.1 ) / 115 which means you need a
conductor CSA of at least 1.1mm^2 to survive the fault and not be
damaged.
Now compare with a case when you can only muster say 130A of fault
current. That may take 25 secs to open the MCB. So run the sum
again:
MinCSA = sqrt( 130^2 x 25 ) / 115
and you now need circuit conductors of at least 5.7 mm^2 to
survive without damage.
Is this how say the wiring of a home is wired ? i.e that the cable
has to survive the fault current ?
Yes. The "On Site guide" has a table that gives a maximum length of
cable permitted for each of the standard circuit types to save needing
to do the sums. In some cases the limitation is that of voltage drop,
and in others its maximum earth loop impedance. Where the limitation is
the latter, the level is set so that the cable should always have proper
fault protection.
And I believe that if a MCB 32 amp is installed the wiring in
that ciruits would be designed to take the current that a MCB of
32a could pass.
You are misunderstanding what a MCB (or fuse) does.
MCBs have absolutely *no ability* to limit the current that passes
though them (save for a tiny internal resistance). Under fault
conditions, the current limitation is mostly[1] down the the fault
loop impedance. If you have (say) a loop impedance of 0.05 ohms,
then you could see a 4600A fault current irrespective of the MCB's
nominal trip current.
All a MCB can do is limit the time during which the fault current
is allowed to pass.
Very similar to most fuses then.
Yup.
[1] The inductance of the supply transformer at the sub station
will slow the rise time somewhat for very high fault currents.
But the fault curretn is a bit obsure because you;re factoring in
time. What is the fault current of a 32amp MCB.
Nominally 160A for a type B device. 320A for a type C, and 640A for a D
type D.
They produced an overload. However the voltage drop you
witnessed during this episode does cast doubt on the ability of
the circuit to correctly deal with faults.
I'm not sure where you get that idea from.
By doing sums with the data you provided.
Then perhaps we have a faulty instalation.
Possibly.
Yes and will this happen at 32A 40 A 50 A and how long will
it take ?
No it won't happen at 40A or 50A. Those are not fault
currents, those are overloads.
I never said they were fault currents, the whole idea behind
this excercise was to find the overload current NOT the fault
current which is pretty much irrelivant to us.
Your exercise highlighted to Adam and I that there may be problem
with the circuit. We have simply tried to explain to you, why this
*could* be dangerous in some circumstances. Should our fears be
correct, then a genuine fault may not be cleared in time to prevent
bad things happening.
True but I've no idea what a genuine fault might be.
The most likely case is when a cable gets physically damaged - say
penetrated, crushed, burnt etc.
Hopefully you are now aware of this and may choose to do with this
information whatever you like.
Nothing I can do with it.
Probably nothing you can do directly...
For those wishing to don the tin knickers, a memo to "the powers that
be" would be about all you could do.
i.e. When we did this, we observed that. It has been drawn to my
attention this may indicate a problem, but I am no expert and can't say
for certain. You might want to consider having the wiring checked etc.
If it then blows up later, or someone gets electrocuted, you can put on
the "I told you so" tee shirt and wash your hands of it.
--
Cheers,
John.
/================================================== ===============\
| Internode Ltd -
http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/