Fitting new MCB

[davidclifton]

Since moving to our current address we have been changing the lights and we now find the the MCB (5amp) for one circuit trips out occasionally. I suspected we had exceeded the limit on the MCB and when I checked yesterday we have approx 1.8kw of lights in that circuit (about 8.7A I think) so I need to fit a 10A MCB. As I've never done this before do I have to get the same make as the old one (mini-trip)? Also do they just pull out of the CU. I don't want to damage anything. Can someone give me some advice please?

[Steve]

Sadly its not just a case of swapping the MCB. tests need to be carried out to ensure the wiring on the circuit is suitable for the upgrade.

Out of interest, when does the MCB trip? Is it when you turn on certain lights? (I'm thinking toroidal transformers with high inrush current here) or is it when the circuit is definitely overloaded? It may be sufficient if this is the case, to fit a C6 MCB.

I would seek professional advice from a registered electrician - he will be able to assess your wiring better than we can. I believe this is also legally notifiable to your local building control under part p of the building regs.

And if you post a picture fo the CU using a image host such as www.imageshack.us we will be able to tell you a bit more about the difficulty of job this is - some CUs are easier to swap MCBs.

[sm1thson]

you could buy some energy saving bulbs! -easyer than changing the MCB+ cables! and will save on your elecrtricity bill!

[kevnurse]

You seem to have some knowledge of electricity. As a fellow DIYer, I think it would be reasonable for you to so some careful investigation on your own. Obviously, choose the right moment, for the family, to switch off the main isolator and have a look inside the CU. Be aware that the large grey phase wire going into the main switch assembly can carry a lethal current and if you can see any exposed red sheath under the grey cover or even, God forbid, any exposed conductor, STOP, and put the CU cover back on. Call out a lecky, because the CU is unsafe and the Main Isolator Fuse (as I call it) on the supply cable going into the CU needs to be removed by an authorised person to properly refit that phase wire.

Assuming the CU is safe, you can check the size of the wire to the lighting circuit as a a basic start. You might know that MCBs are categorised into 3 types: B,C and D, whereby the letter indicates how quickly the MCB reacts to overloads. B is the quickest, C is medium and D is slow (too slow for domestic use) Wiring regs require lighting circuits to be isolated within 5 secs of a phase to earth fault. See the Wiki for more details. I guess the 5 secs is there to allow for continuous switching of lights (current surges) and the fact that no-one is expected to be handling lamp holders on the ceiling, etc, so injuries are not expected if the circuit is properly operated (ie not handled).

MCBs are easily removed. They are usually clipped onto a DIN rail. It should be obvious how to do it once they are exposed. Even though the main isolator will be off (because you did that to begin with) I strongly suggest you check throughout the CU for any stray voltages with a multimeter before you put your hands in there.

It would be sensible to change the MCB like for like as far as manufacturer is concerned, but you could simply take the old MCB out and take it to the wholesaler and get one of the same design. The rating of the replacement needs some thought. Here's my thinking: despite the 5 secs time limit, you should aim for a quicker disconnect (B type) if any lamps on that circuit are wall mounted and easily touched or in areas of heavy people traffic. ie if there is any chance that someone might knock it of the wall and create a short to earth it must disconnect ASAP and a B type is the better one for the same current rating. If all the lamps are out of normal reach, a C type is good enough. Given that a C is more resilient to overload, you could fit an MCB that exactly matches the max potential load. If you need a quick disconnect you will need to get a B type that will be resilient to surges at max potential load, so a B 10A will probably work with your 1.8kW max load.

I don't think this is notifiable, but it is a minefield.

Doubtless a qualified spark will comment on my DIY approach.......

[ricicle]

The circuit still needs testing to see if fitting a 10A MCB is acceptable due to EFLI and installation method of the cable itself.
The EFLI needs checking to ensure that at the circuits furthest point it is low enough to ensure that sufficient earth fault current will flow to operate the breaker in 5 secs or less.Also although the cable (1.0mm²/1.5mm² usually for lighting) maybe able to carry over 10 amps, it might not due to its install method (eg ambient temp, thermal insulation, grouping)
I think you should seek the help of a qualified electrician.

[Spark123]

The current flowing in a fault needs to disconnect a breaker in 0.1s regardless of wether it is a 5s disconnection or not. If you look in the tables in appendix 3 of the regs it only gives one level of current per MCB for 0.1s-5s disconnection. Types B, C and D have different magnetic current settings, for type B the min current to operate the magnetic trip is its rating X5, type C is X10 or type D is X20.
240/(current setting * type) = max efli @ max design temp.

[Jolberg]

In a previous message regarding lighting circuit tripping you said

"Out of interest, when does the MCB trip? Is it when you turn on certain lights? (I'm thinking toroidal transformers with high inrush current here) or is it when the circuit is definitely overloaded? It may be sufficient if this is the case, to fit a C6 MCB."

I'm having a problem of tripping of my ground floor lightnig circuit which has just been extended with a new kitchen. In the kitchen there are 6 LV spot lights each with their own transformer and a cable lighting set comprising a torroidal transormer and 3 12 V halogen lamps. This part of the circuit seems to cause the tripping which can happen on the first switch on or after 2 or three switch ons; but only when switching on. I have tested the incoming feed cable by putting a different light fitting on it and it doesn't trip. I have also disconnected the secondary side of the transformer and I still get the tripping. This is the second transformer I have tried since I assumed that the item was faulty in the first place. Can I get a trip from a torroidal transformer even if nothing is connected to the secondary side. The circuit is on a B6 MCB. Could changing to a C6 stop the problem? Do I have to test something else on the circuit before changing to a C6? IS there a risk to other parts of the circuit?

I'm pulling my hair out here. Can you offer any advice?

Jol

[bernardgreen]

Jolberg wrote:

Can I get a trip from a torroidal transformer even if nothing is connected to the secondary side.

Yes.

The primary winding is a low resistance to DC and a high impedance to AC. In simple terms for a few milliseconds the low resistance controls the current (and this can be very high) before the impedance takes over as the magnetic field builds up and reduces the current.

[Taylortwocities]

Kevnurse

He has mini trips so all mention of din rails etc do not apply.

Edit : As its mini trips then its reasonable to expect and oldish installation. Lighting circuit will without doubt be 1mm² so EFLI must be checked to see if trip time for 10A breaker can be achieved.

Ditto applies for C breaker but I've never seen them as mini trips

__________________
TTC

When you get to the end of your rope, tie a knot and hang on. - F.D.R.

[JohnD]

different poster. the new guy has B6

( i suppose this is what the hi-jacking rule is about

[Space cat]

Quote:

The primary winding is a low resistance to DC and a high impedance to AC. In simple terms for a few milliseconds the low resistance controls the current (and this can be very high) before the impedance takes over as the magnetic field builds up and reduces the current.

Almost right. The inrush is caused by two things: core saturation and cold filament resistance.

When a transformer is running correctly the core flux is peaking when the voltage is zero. During the next quarter cycle the voltage rises to maximum and the flux falls to zero. A quarter cycle later the voltage is zero once more and the flux has peaked in the opposite direction. The flux is zero when you first switch on and so, ideally, you would switch on when the voltage was maximum. In practice you can't do that. The worst possible time to switch on is at zero voltage. During the first quarter cycle the flux rises to its normal peak value but the voltage is not zero; it is maximum and it's still driving the flux up. Unless the transformer core has been designed to handle this doubled flux (which it won't have been) it will saturate. This is when the primary inductance falls to a very low level - less than one thousandth of its normal value - and so you get that huge current surge. Since you can't control at what point in the cycle you switch on, the effects of core saturation are random.

Now for those filaments. All incandescent filaments have a much lower cold resistance than their running value. Try this with a standard 60 watt bulb. Ohm's law predicts a resistance of just under a kilohm but your meter will tell you it's about 70 ohms. Halogens run hotter and so the ratio is greater. It was 16 : 1 on the last one I tried. As if that wasn't enough, low voltage filaments are shorter and thicker so they take longer to heat up.

What can you do about this? Electronic 'transformers' do not suffer from core saturation. They should also have built-in current limiting to deal with those cold filaments. Unfortunately many (maybe most) of them are inferior to real transformers in the reliability department. When a halogen bulb fails it can draw a massive current spike, much bigger than you get with a standard bulb. Your real transformers will survive this; all too often, electronic ones get fried!

__________________
Their base is so secret, even the space cats don't know where they are.

[bernardgreen]

Space Cat

I think you will find that switching power at the zero crossing point is the better option.

The reason is this ;-

At the instant of switch on the voltage is very close to zero across the low DC resistance so the V/R (voltage / resistance ) current is also very close to zero. As the voltage increases the V/R current increases and with it the core becomes magnetized. The increasing magnetic field induces an EMF ( voltage ) in the primary that opposes the applied voltage thus reducing the effective voltage and hence the current is reduced to less than the current expected from applied voltage over resistance.

[plugwash]

your understanding of inductive loads seems rather flawed! current will not suddenly rush through an inductor the instant it is turned on.

Space cats explanation of the core saturating during startup under certain startup conditions is a far more reasonable explanation for a surge soon after switch on.