5A not enough on lighting circuit?

Does it trip when they have been running for a bit, or the instant when you switch stuff on?

I think some tests on the circuit would be advised. Even a simple current measurement would tell you plenty. If you have access to one, a clamp meter placed round the live wire at the MCB in the CU would let you see the actual current drawn with no apparent load (ought to be 0) an then as you switch in each light.

From there you should have a good indication as to the type of fault you are looking for, which could be a simple as a faulty MCB to a wiring or appliance problem.

It sounds like you have a definite problem that needs investigating and fixing. IMO uprating the protection device is *not* the way to go (even if the cable can nominally take it in ideal conditions).

It can run quite happily for anything up to 30 mins before it trips, it seems almost random.

Second time in as many weeks I've been told I need a clamp meter. Shall add one to Mondays shopping list!

Thanks

Loose wires in a socket or to one of the fans. Could be the result of vibration and causing an intermittent load that trips the breaker

Does it only trip when the fan motors are running/starting?

Not always... there are 4x58w inline fans plus 3 x shower extractors, 6x35W

You can run everything for a few minutes or a combination of rooms for half an hour, after than it gives up.

It was perfectly fine for 2 years up until about 3 weeks ago.

That would suggest a sustained overload then. Remember that a MCB will pass current significantly above its marked value for some time. Assuming we are talking about a 6A type B MCB, then table 3.4 of BS7671 would apply. That would indicate a sustained load of some 9A to result in a trip in that time frame.

It is not essential, but it is a quicker and safer way to make these sorts of measurements. Some multimeters will measure AC current (although some that do only go up to 5A or so), but, they need to be placed in circuit to make a reading. This obviously requires disconnecting wires and finding a safe way to connect them to the meter (a good set of shielded probes with insulated croc clips - or failing that a chockie block to join the circuit wire to either the probe end, or better still a suitable shrouded banana plug).

With a clamp meter you can place it round any suitable individual conductor[1] (i.e. you can't just clamp round a whole flex or cable since the current flowing back through the neutral will exactly cancel the reading you would get from the current in the live). So for example placing it round a meter tail will give you a base reading for the whole houses consumption at that moment without even needing to open the CU. Assuming this is reasonably static (i.e. no big loads switching in and out on thermostats etc) you can probably conduct useful tests with just this.

An alternative is to connect it across the circuit's switch - with the switch "off" the meter will be in circuit.

If you have two 5 amp MCBs on the CU then swap them over to see if the MCB is faulty.

Adam

Yes, good point! handy for narrowing down likely culprits if it turns out to not be the wiring itself.

What else has happened in the last two weeks? Any DIY going on that required holes in walls? Changed any bulbs or fittings?

shaver/lamps

Jeuss, thats some hall way / bathroom! Are you sure you dont live in a hotel?!

Tim..

Not far from the truth! A Youth Hostel. Bear in mind this circuit only covers about 40% of one of three floors.

Now you begin to understand my problem!

AFAIK nothing has changed. :-?

In message , TonyK writes

Something may have changed, but I think you've got a lot of low power factor loads, particularly all the 2D fittings if they're magnetic ballasts.

Before these calculations get torn apart, I've made a guesstimate at the typical PF of these loads. Particularly the motor loads I'm not sure of. Anyway, Qty x Wattage / PF = VA

Total VA = 1387, or 5.78A.

So the circuit _may_ be on marginal overload anyway, although as pointed out previously a typical B breaker would take that lot pretty indefinitely. I think some actual current measurements are in order, and check any magnetic ballasts for signs of over heating (although you'd have smelt them by now if they were on their death bed and cooking).

Diversity should see that lot run indefinitely.

FWIW the halogen transformers and CFLs will have much worse pf than that, they will only take current at v peaks. But diversity still applies.

NT

I agree, I'd say you are close to the limit, and possibly the circuit breaker has started to go 'sensitive'.

Given the cost, I'd go for a new one of the correct rating for the wiring, before doing anything else.

Its cheaper than a meter.

I'd make it a type C as well

NT

Just what I was thinking ...

I suspect you may have overestimated some of the PFs.

I'd guess 0.65, so 450 VA.

Efficiency would be about 95% if they're iron-cored transformers (so 220 VA), but electronic xfmrs could have a much lower PF (due to waveform distortion), so in worst case the loading might be as high as ~400 VA.

PF of these, unless corrected, is close to 0.5, so 410 VA, say.

Ditto 0.55, so 130 VA.

They might be fluorescents, but let's stick to worst-case.

CFLs

PF closer to 0.5, say 130 VA.

Worst case total VA = 1760, or 7.6 A - enough to trip an old 5 A MCB (if that's what it is), but probably not a 6 A one.

The circuit is certainly capable of being overloaded and should either be spit into two, or uprated to 10 A, if the cable installation conditions allow. As has been said, test and inspect first before, uprating the breaker. And make sure that the clamp meter reads true RMS amps, otherwise the results will be fairly meaningless (even for the iron-cored loads harmonic currents will contribute significantly to the lower power factors).

TonyK said the following on 31/12/2005 14:39:

BS7671 Regulation 553-03-01 (Table 55B) applies. Overcurrent protection of lampholders (under certain circumstances) is limited to 6 Amp for SBC and SES lampholders. This is why domestic lighting circuits are always rated at 5/6A. Even if you don't have any light fittings with SBC (small bayonet) or SES (small edison screw) lamps IMHO it would still be considered bad practice to upgrade the MCB to 10A.

If it turns out that you have an overloaded circuit, you should split the load onto two or more separate 5/6A MCBs.

HTH