Electric Shower tripping MCB

Page 3 of 3  
In uk.d-i-y, Chris Oates <none> wrote:

Piggin 'eck, this thread has generated - to use a painfully appropriate saying - lots of heat and no light: a ridiculous misapplication of a misnamed law, and crap advice. Sorry to be blunter than usual, but there it is.
To deal with the practical aspects first: there is something *wrong* with at least one of: the shower; the cable feeding it; the MCB; the incoming supply - and that's more or less in order of likelihood (with change in the supply being massively less likely than the other three). The original poster's brother could do a lot worse than spend 50 quid on getting a competent local electrician in: the mixture of lots of incoming amperage, wet skin, and well-earthed metalwork is *not* to be treated lightly. Yes, the RCD - *if* it's working - should protect your brother and any other members of the household from the worst consequences. How lucky are you feeling? And replacing the 40A MCB with a higher-rated one, even the slightly higher rating of 45A, could be
D A N G E R O U S
leading in a quite plausible case to a
F I R E.
Again I ask: how lucky are you feeling?
Now, a little more reasoning. To all intents and purposes, the supply voltage across the UK *is* 240V. The FAQ explains that for product design purposes, we've 'harmonised' across Europe on a wider range of voltages for which kit should behave, but the *actual* supply voltages in the different European countries is unchanged, for now and for a number of years to come. So let's stop dicking about with "well if the voltage was X, the current would be Y, but if it was Xprime, the current would be Yprime", 'cos in just about all reasonable foreseeable circumstances, the voltage will be within a gnat's whisker of 240V. (And another reason for dropping it is that one of the incessant responders got it all arse about face, as Bob tried to explain: more on this below).
So, as it's sensible to say "V = 240", what's I for a shower rated at 9.5kW? Well, P = IV, so I = P/V (/ meaning "divided by") = P/240 9500/240 = 39.6A near as damn it. Follow? So a "40A" MCB is *NOT*, *NOT*, N O T too low (or even 'to low' [sic]) for a 9.5kW shower. Thus we conclude that either the whole circuit - shower+supply cables - is pulling *more* than the 9.5kW it ought to, or the MCB has lost its marbles, and "become oversensitive". But "becoming oversensitive" is not a particularly well-known failure mode of MCBs. It could be that the 40A MCB is sandwiched between a couple of other MCBs also running close to, or a little over, their rated currents, so the mutual warming among these is making the 40A one trip: but that's not likely.
What else could be causing a fault? Well, maybe the heater element has somehow developed some internal shorts, so its resistance is now lower, making it pull more current. Possible - happens in electric blankets - but not likely in a leccy shower, where there's not much chance of the heating-element wire touching itself inappropriately. (Pauses for the giggles from the back of the class; looks severely over glasses; continues, nary a glimmer of a smile passing across his face). More likely is some other internal fault in the shower, allowing substantial extra current to pass from L to N, which after a while makes the MCB trip. (The marking of 40A on the MCB is a 'nominal' one; it'll pass 42A or 43A for hours and hours - it won't suddenly cut in at 40.001A! You can look up the detailed "overload versus time to break circuit" curves in any MCB supplier's technical data sheet.) Worse, the fault could be a Live to Earth fault, pulling 5A or more (that's about what you'd need for the MCB to trip not at once but after a little while). Surely the RCD would cut in if it was a L to E fault, wouldn't it? Why yes, of course it would... *if* it's still working. Is it?
What else? Well, maybe the cable was put in a while ago, and was sized for the kind of shower more common 10-15 years back - a 7.2kW model, say; with a 30A MCB. Then someone put in a new, more powerful shower - the current 9.6kW one; and read the instruction about "needs a 40A MCB", and put one in. After all, the cable looked pretty beefy, right? What would happen then? Contrary to Hollywood special-effects pictures, the somewhat overloaded cable will *not* instantly turn into a molten plasma, vapourising all within a 10-mile radius. No, it just runs warm; a bit too warm. Hot, in fact; hot enough - maybe somewhere out of sight, with a bit of thermal insulation - to make the PVC insulation start to soften. Somewhere where this is happening, the cable is bent; and the softening insulation allows the live and neutral conductors to come closer together. If they do that suddenly - bang, the trip will short. If they do it gradually, though, you could get just the behaviour you're seeing: L and N nearly touching when cold (shower off), bugger-all insulation between them now, just a small air gap. Then you turn the shower on; the conductors heat up gradually, expand, and now they touch. Off pops the MCB.
And *that* is one of the possible reasons for the MCB going off.
And *that* is the situation in which some posters are suggesting 'whack a (slightly) bigger MCB in'. *Not* all that sound a suggestion, that. The cabling fault - if such it be - need not relate to an undersized cable, as the detailed picture drawn above had it: similar effects can occur from connections which have worked loose, so they're now getting too hot, maybe sparking a bit, again softening/weakening/charring insulation so it doesn't do its job anymore. No, I don't *know* if a worsening cable fault is causing the MCB to blow more often than before; but it *could* be; and you'd be a fool to just cover up the symptom by whacking in a bigger MCB. If you don't have the knowledge to work out what's going on, for pity's sake pay for an hour's time of someone who does.
Oh, the "mains supply" possibility? Well, it's remotely possible that the incoming supply voltage has been increased recently: maybe someone further away from the substation has complained of too low a voltage, and the supply company's shifted up to the next tap on the transformer; or some local heavy user's no longer heavy-using (any cannabis-in-the-loft growers been busted round your way recently? or a little engineering company's gone bust and no longer fires up its heavyish loads at the times the shower's in use?) A higher incoming voltage will (as we work out in remedial-level detail below) cause increased current to flow. But any change in incoming voltage won't be dramatic - otherwise you'd be seeing lightbulbs glowing extra-bright and burning out noticeably more often; so as an explanation for the MCB behaviour it's the least plausible.
Finally, the electrickle theory bit. The two relevant equations are "V = IR" - that's the one we most usually call Ohm's Law - and "P = IV". Each one tells you how *three* quantites are interrelated, so to ask 'how does one quantity vary as the other one varies' is unanswerable, *unless* you can *justify* the assumption that the third one will stay *constant*. Geddit? The guy rabbiting on about the greater current being drawn as voltage fell in order (by some unexplained, and in this case totally non-present) to magically maintain a constant power was talking utter tosh. As Bob tried to explain, the most nearly constant quantity here - as a matter of physical truth, given the way a shower element is built - is the resistance of the element when it's up to its operating temperature. So, knowing the resistance is fixed, we can do some Really Complicated Algebra. Ready?
V = IR
We can actually stop the analysis right here. Given that R is constant, this says that as V goes up, so does I, 'linearly' - i.e. double V, and you double I; knock V down by 5% and I comes down by 5%. End of story.
If we want to go doing Deep Mathematics, we can go on to work out how the power developed across the fixed resistance will change as the voltage changes. Coo, that's going to be advanced mathematical reasoning, since we haven't been given a single formula which has both P and R in it. But we're going to *derive* it, i.e. work it out. You see, we know
P = IV
And we're going to monkey with V. Can we assume I will stay constant? Can we hell - we just reasoned a couple of paras ago that I was linearly dependent on V: that's what "V = IR" means, remember? Now, given that V = IR, we can write V/R = I (/ meaning Division), by dividing both sides of the Ohm's Law equation by R. (It's mathematically and physically valid to do so, since we know R isn't 0: no superconductors in sight ;-). So, armed with the manipulated flavour of Ohm's Law saying "I = V/R", we can replace the I in "P = IV" with "V/R", to get - wait for it -
P = (V/R)V
or as we more usually write it,
P = V*V*R or P = V-squared times R or P = V^2*R
Thus we see that, *across a constant resistance*, the power developed is proportional to the *square* of the voltage. So, if the voltage sags by 10%, so it's only 0.9 of the previous value, the power across constant-resistance elements will be 0.9*0.9 = 0.81 times its previous value, i.e. nearly 20% down. Fascinating? Maybe. Relevant? Not to the original poster's brother, really; to the unteachable who kept on hoping the constant-power fairy was in the room? yes, but will they ever realise its relevance?
OK? Are we settled on what to do? - For the original poster's brother: get someone genuinely competent in to find the fault. There is a fault. Find it before it has more serious consequences than it has till now. - For the wannabe electrickle theorists: - get your theory straight in your head; - align it with physical reality; - consider the consequence of shitty advice when it concerns heavy-duty volts, amps, and wet skin.
Stefek, shirtier than his normal self...
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@hp.com wrote:

<snip>
Well reasoned * well said = well done
--
Toby.

'One day son, all this will be finished'
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

<blush>thanks, Toby</blush> Another possibility is that there's more than the shower running off the single MCB! Of course, there *shouldn't* be; just like the immersion in this place *shouldn't* have been put onto the upstairs ring back at the CU when someone needed a couple of new ways in the only-one-spare-way-left CU. If there was another load sharing the shower feed - some recent hack-it-about alterations - then that too would account for the behaviour reported. And the MCB would be *right* to trip to protect the cable...
Stefek
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Blimey! Thanks for the explanation Stefek.
Could you please explain why a motor which runs at a lower that nominated voltage, draws more current to keep the power output the same?
Mike.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

very good explanation here http://www.ecmweb.com/ar/electric_highs_lows_motor /
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 18 Jan 2004 19:30:51 UTC, "Chris Oates" <none> wrote:

It doesn't really explain why a motor doesn't behave like a resistive load; it states 'it does this' but glosses over 'why'.
--
Bob Eager
rde at tavi.co.uk
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 18 Jan 2004 21:03:26 +0000, Bob Eager wrote:

To be fair, I think the article assumes a backrground knowledge of electrical engineering. Where term like "power factor" and "magnetic saturation" would be concepts the reader is expected to know about beforehand.
--
Ed Sirett - Property maintainer and registered gas fitter.
The FAQ for uk.diy is at www.diyfaq.org.uk
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
<massive snip>
Thanks Stefek, for your comprehensive reply.
Just to fill in a couple of blanks, the cable (10mm2), the RCD and the 40amp MCB were all installed at the same time as the shower. There was no previous electric shower. There is definately nothing else on the same circuit. The total length of the cable run is about 6 metres.
I will get my brother to test his RCD. Is the built in test button a sufficiently good test to rule out a problem with the RCD?
I believe the shower is still within guarantee, so if the RCD works and all the connections look sound, we'll get someone out to have a look at it.
Cheers Simon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I recently went around this exact loop when installing a shower.
It was a similarly rated Triton model to that whose spec was mentioned by Ano poster(s) - i.e. 8.7KW at 230V, 9.5 KW at 240 V.
Anyway, when rating the cables etc. I thought I'd better measure the local voltage. Turned out to be 248 Volts under moderate load, which implied a draw of a shade under 41 Amps, IIRC, for the shower. (Making the assumption of a linear resistance change with increasing voltage - not 100% correct, but a reasonable approximation here).
Should also say the load of the shower would pull the actual supply voltage down a bit, so reducing the current actually drawn... never checked how much.
Anyway, as the Triton installation guide said "40/45 Amp MCB" for this nominal 9.5 KW unit, I fitted the latter.
Given my problem finding a **competent** electrician, I'd think it would be worth you checking the local voltage **but only if you feel you can do this safely**, just to rule this in/out as a possibility.
BUT, as has been correctly stated, MCB's don't just trip out as soon as the rated current is exceeded - typically a so-called type B will allow double the current for 60 seconds or more. How fast does it trip ??
"coherers"
wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I've got a multimeter that can handle upto 700 volts ac, so I'll take it round there and get a read for the voltage.
My brother gets up first in the morning and usually has a shower OK without it tripping. When his wife get in the shower a bit later on, it often trips after 5 min or so. Not always, but increasingly often.
The RCD has been tested, using the built in test button, and appears to work correctly.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Best to test the voltage when the shower is on if you can ( But your test points not nearby!)
Reading graphs for various MCBs, I figure it would need to be drawing *at least* 50A for 5 minutes to trip a "B" or "C", which seems unlikely. However, really depends on the exact characteristics of the type in use. And it could have wandered off spec, hence it starting to trip where it never did before.
If it is drawing over 40 A, and the shower manufacturer allows for it, replacing the MCB (6 or 7 IIRC) could be a fast and cheap diagnostic/fix. However, as I recall not all consumer units brands take the 45A flavour and the 50 A is a definite no-no....
However, following up the other posters' comments on safety, I would seriously recommend no-one uses the shower for its intended purpose until you have got to the bottom of this one.....

no
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

And
diagnostic/fix.
and
'wandering' is what mine did - I brought home the clamp meter and found it tripping after 4-5 minutes at 36A

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

That means tha the MCB is tripping thermally, due to a sustained slight overload, rather than magnetically, due to a heavy overload.
Serious question: How long does your brother take in the shower? If he is one of those people who takes quite a while, this could explain the problem.
And how soon after him does his wife shower?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I don't know their showering habits well enought to give an answer. I will have another conversation with them to find out. Thing is, they have had the shower for several months, and their morning routine has not changed recently, so why should it start tripping and why should it trip more often as time goes on.
Simon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I can only suggest that the thermal part of the MCB is becoming 'tired' due to repeated tripping and/or lengthy periods of being heated to an 'almost trip' state.
If the usage cycle of the shower is such that it normally has almost but not quit reached a trip state, it wouldn't take much cahnge t cause the problems that you have described.
A few minutes extra use each day, your sister-in-law's shower following your brother's more closely, a slight change in line voltage - any ro all of these could affect the situation.
One simple test I'd suggest - turn on the shower, leave it running, and see how long it takes to trip.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Cheers. Radio silence from me is because I'm in the US for a week; just chewing through uk.d-i-y backlog while considering wording of major document I need to start on tonight ;-)

OK, so the more worrying possibilities become less likely (phew!) It's possible then that the positioning of the MCB between other loaded MCBs is the problem - they're warming one another up and making the shower one trip. But that's not nearly as likely as a fault in the shower, methinks...

Sadly not: if the test button doesn't make the RCD pop, the RCD *is* faulty for sure. But if it does pop, the RCD could still be out of spec: too sensitive, not sensitive enough, take too long to pop (not as if you can reliably press the test button for exactly 20 milliseconds, is it!). But in your brother's case it's an individual MCB which is popping, not the RCD, right? To "properly" test an RCD needs a "proper" RCD tester, which full-time electricians can spread the 700-quid cost of over years of work, but which thee and me aren't going to shell out on! It'd make a good school *design* project, though, using suitable resistors to make a 10mA, 15mA, and 30mA load, and a gravity-dependant timing element such as a steel ball running down an inclined channel with a conductive section: lots of good simple applied maths working out how to do various short timing intervals. Course, Health-n-Safety being wot it is, you couldn't ever *build* one with the kids :-(

Hope you/he get it all sorted... Stefek
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Site Timeline

Related Threads

    HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.