Faulty consumer unit

A new consumer unit with RCBO's would be one answer. it would not necessarily prevent nuisance tripping caused by faulty appliances (including electric showers) but would be safer in that event.

Is the cable to the shower correct in the days of 7Kw showers 6mm2 was the usual connection. More modern showers of 10 - 11Kw really need 10mm2. This aspect is often overlooked by DIYers when replacing electric showers. Also is the MCB up to providing the current drawn something in the/region of

The wiring was all done professionally.

I served an apprenticeship in electrical engineering, but would always stop at replacing sockets, switches, etc at home. Certainly no digging out the plaster and re-wiring!

But still, I'll have that checked out too. Thanks.

Thanks for that.

I have no reason to suspect that the shower is faulty. We have occasionally had trouble with the ring main trip when toasters or kettles have become U/S. A friend observed that it's always a kettle or a toaster.

If you can tell us:

1. The model of the shower and its kW rating
2. The rating of the MCB and its curve (letter and number will be printed on the front like B40, C32)
3. The size of the cable (4/6/10mm2)

Then we can make guesses as to how marginal things are.

Most showers have four heat positions: 'off', 'cold', 'I' and 'II' which correspond to inlet valve closed ('off') and open with 0, 1 and 2 heating coils in circuit. The knob for fine temperature control actually controls the water flow rate.

When in 'I' it's only taking a fraction of the current that it does in 'II' (not necessarily half, I think it might be more like 60% in some showers, the coil resistance may be stated somewhere) so unsurprising it doesn't trip in 'I'.

MCBs have trip curves which aren't perfectly specified - they're supposed to trip at multiples of the fault current within a specified time. If you take them a small amount over the rated current they may take hours to trip, but that amount could change as they age.

The length of the shower could also affect things - eg if you take longer than you once did, you'd only notice now. Maybe you had a 5 minute shower before so you didn't notice it tripping after 10 mins.

It's no guarantee that just swapping the MCB will make it better (could make it worse), but we can tell you if it's marginal and what upgrades might be needed to fix it.

Theo (whose 10.8kW shower failed EICR because it was on 6mm2 cable, now thankfully gone)

No. It sounds much more like you have an issue with the *wiring* on that circuit.

Possibly an earth neutral leak if its happening to trip the earth leakage, or if its tripping the overcurrent MCB, earth neutral.

I assume this is a spur, not on a ring main? In either case get a megger on to the circuit when its isolated from the CU AND all loads

And have a good look at the wiring wherever you can to ensure that its not your RCD (Rodent current detector ) refrying a rat...

Forgot to add:

1. Where the cable goes and roughly how long it is.

Specifically I'm interested in which 'reference methods' apply:

(these are used to calculate the thermal overload parameters for the cable. Not strictly relevant for the trip, but will indicate whether the cable is correctly specified for the shower you have)

Theo

I can surmise a strange situation which /might/ result in what occurs, and involves neither a fault in the shower nor the CU. I assume that what you were originally referring to is the RCD tripping rather than the MCB (all power lost), but the MCB only is tripping now (only power to the shower is lost).

If TD is right, and the cabling is running near to its limit, consider a mouse or rat chewing through the insulation of both live and earth. Running at full power the cabling got hot, and expanded and/or sagged enough to touch the earth, so tripping the RCD. However, if not running at full power for several minutes the cabling never got that hot and did not touch the earth, so no tripping occurred. Perhaps further insulation has been removed in a slightly different place and live and neutral are now touching rather than earth, so tripping the MCB rather than the RCD.

Crazy speculation I admit, but we've all heard of Sod's Law...

Thanks for all the responses.

While I can easily quote the shower model (Triton Madrid II, about

Many thanks to all who have offered advice here. I am grateful.

With high current circuits, it's worth checking the connections periodically, as the thermal cycling tends to cause them to go bad faster, which causes larger thermal cycling, heading in to burnout. Depending what form they take, and slight slackening and retightening would do this (with circuits isolated, obviously). In commercial premises, they are sometimes inspected with infra-red cameras to check for hot terminals.

Heating of the MCB terminals will cause the MCB to run hotter, and hence trip at a lower current too. Heating of an adjacent MCB can have the same effect, as can having two adjacent MCBs running near full load.

That wording suggests to me that it was tripping out multiple circuits and not just the shower one. If so, it suggests that there is a whole house RCD and that the shower has developed an electrical leak to earth. Possibly there are two heating elements (one for low and both for high) and the high element is failing.

If that is the case then replacing the shower should solve the problem. However, we kept having trips and figured that the total leakage of all our (lots) of electronic devices, with filters that permanently leak to earth, was too high. We removed the RCD and replaced all the MCBs with RCBOs, thus splitting the leakage across multiple circuits ... none have tripped since.

If used on the higher heating selection, that is what happened.

It happened on the previous unit too. It was not Triton branded, but made by Triton under another name and the equivalent of the Madrid II model. I remember ringing Triton to get that information when the previous one failed.

That sounds like something else which needs to be explored - thanks.

That was how I read it, until the part where the fault had persisted across several shower replacements.

This does sound to me one of the most likely reasons for the fault. A

Well that's the problem! A 10kW shower needs a 45A circuit breaker and wiring.

Can you confirm only the RCD is being tripped and not the MCB as well?

Some RCDs can be over sensitive. It might be worth testing tripping current, or if you don't have access to making the measurement, simply change the RCD.

The OP hasn't told as what circuit breaker or wiring they have, AFAICS. That's why I asked. Otherwise we're just guessing.

(even just telling us the number written on the MCB would be enough to work out if it's suitable or not)

Theo

One thing you should be aware of, with giant resistive loads on house wiring... is your *line voltage*.

Say a particular gadget you bought, lists "3kW" on the rating plate. That's 3kW at "nominal line voltage". The voltage at my house, is always too high. This increases the current flow through resistive loads. Suddenly, it's not a 3kW load any more.

This prevents me from running electric fires on "high".

Resistive heater loads, when the element is cold, they draw even more current. It could be the initial presentation of an ice cold shower load, which instead of 3kW, draws 3.5kW until it starts to heat up. If the house line voltage is too high, the initial power draw is now 4kW, and it settles down to about 3.5kW. It would not drop to 3kW, unless the line voltage dropped to the specified nominal value.

The protection device, may just be too close for the size of load, and there is not enough margin. Several decades ago, a standard electric water heater here, would have used a 15 amp breaker. Today, they use a 20 amp breaker (just to avoid nuisance trips), and the wire from the panel is gauged for 20 amps).

While your consumer unit could be faulty, a confluence of factors could also account for nuisance trips.

Paul