IET logo
 
IET
Decrease font size
Increase font size
Topic Title: "Electricians" that don't have a clue about RCD's !
Topic Summary:
Created On: 21 September 2015 05:43 PM
Status: Read Only
Linear : Threading : Single : Branch
1 2 Next Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 21 September 2015 05:43 PM
User is offline View Users Profile Print this message



daveparry1

Posts: 7728
Joined: 04 July 2007

Some people I know have just taken over a building which has been converted to offices. They say their power goes off randomly when they have their computers running (about 12 of them) and usually the power goes off when they run the hot water and the boiler fires up. Upon looking around I found the modern twin rcd c/unit is fed by a 16mm t/e from the meter on the floor below, this sub-main has a 30m/a rcd and a 40 amp mcb feeding it. The people there said it's always that rcd that trips out, not surprising of course. My point is that they tell me they had an electrician in the other day who changed that rcd because it was only a 40 amp one so he replaced it with an 80 amp one, still only 30 m/a though!
I'm going back there tomorrow and was going to fit a 100m/a rcd but thinking about it there's no need for the sub-main to be rcd protected anyway because it's not concealed in the wall anywhere.
 21 September 2015 06:19 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 15838
Joined: 13 August 2003

Not TT is it? (just asking the silly questions first)

Even if it was, a 100mA S-type would seem more appropriate.

Or particularly flammable (stored or processed materials etc) - maybe then 300mA max. could be called for.

or running through a bathroom, or similar special location?

Or unusually high Zs?

Otherwise I can't think of a need for an RCD either.

- Andy.
 21 September 2015 07:00 PM
User is offline View Users Profile Print this message



davezawadi

Posts: 3847
Joined: 26 June 2002

Assuming its not TT (even then its pretty poor design)

Ah, you've just found a screwfix man, shower CUs are cheap and 16mm T&E is obviously free!
I'm not keen on the MCB either.
And then the guy who changed the RCD for an 80A one? Well I wonder where that idea came from (perhaps the 16mm)?

I cannot say that this is at all surprising, and these guys probably belonged to a scheme too.... so this must be the "approved" design although the approver (QS / scheme?) should be ashamed.
None of it is dangerous of course, so perhaps everyone should be congratulated on a good job done!

Then again perhaps not, it obviously doesn't work very well.

-------------------------
David
BSc CEng MIET
david@ZawadiSoundAndLighting.co.uk
 21 September 2015 08:44 PM
User is offline View Users Profile Print this message



Zs

Posts: 3814
Joined: 20 July 2006

Dave P,

You touch on something I oft mention to clients who perhaps find me a bit too, well, explanatory when advising them;

Our profession, in my opinion, is the point at which being a tradesman rubs shoulders with knowing a bit about physics. So we are somewhere between the bloke getting a sandwich box and a load of cable out of the back of his van and the scientist with his teflon head and white lab coat.

I blame straightforward curiosity for my own incessant googling and learning but for those who don't have that degree of wanting to take the back off things and look inside them, I reckon the kind of scenario which you mention is quite normal. If in doubt, increase the A? He probably said the word overload? That's the kind of thing we talk about in the wholesaler after all. The fount of all knowledge is the counter at the wholesaler isn't it? He might also have been right of course and it might need both things to be attended to. I'm playing devils advocate of course.

You don't know what you don't know. I can't give you a list of mine because I don't know what's on it.

In passing, I have a few mentees. One or two of them are quite senior and are very experienced installers. They are scared to ask about what they don't know, on site or even on here. I am musing now on whether that's where the ladies have a role to play because they are not scared to ask me. Hmmm, not sure I like that at all but you have started a train of thought Dave...I'll develop that.

Zs
 21 September 2015 09:20 PM
User is offline View Users Profile Print this message



daveparry1

Posts: 7728
Joined: 04 July 2007

Yes Zs, they did say the other chap mentioned "overload" causing the RCD to trip!
 21 September 2015 11:12 PM
User is offline View Users Profile Print this message



mapj1

Posts: 9371
Joined: 22 July 2004

Still, when you have fixed it properly, leave them your card ;-)
If the Zs and R1&2 will stand it, if I understand you correctly it could have been a fuse at the submain origin and be fine, and might as well be.

Mind you I work with a lot of very bright chaps some of whom also move electrons around for a living, some who have other areas of expertise, and I am often sharply reminded how easy it is for someone to drop off the end of their knowledge and not realise.
Then one does the engineering equivalent of that thing where bugs bunny runs off a cliff and keeps running, and does not fall until he looks down.
By which I mean that having made the wrong assumption, it is very easy to keep going trying to do more of the same, still using the wrong assumption, often for sometime, and by the time it becomes really obvious, getting back is quite hard, and sometimes expensive.

-------------------------
regards Mike
 22 September 2015 12:31 AM
User is offline View Users Profile Print this message



spinlondon

Posts: 5494
Joined: 10 December 2004

Is it safer to provide RCD protection to the sub main?
If yes, are we permitted to make an installation less safe?
 22 September 2015 12:49 AM
User is offline View Users Profile Print this message



davezawadi

Posts: 3847
Joined: 26 June 2002

Interesting...

Hello Zs, you see where I was going which is good.

And Mike, the problem is not the knowledge or the people, it is being alone with a problem. A team makes all the difference. Its worth asking here, which is pretty safe, if you are not really sure. At worst someone will say BOO, but after a few posts the answer will be there. and free too.

The bit I do not understand is why the fundamental concepts are not understood better. It is always why? that is missing. Ask, enquire, learn, that is the basis of all education and training. There is more CPD on this site than all the courses one could attend, but it does take a bit of effort. Probably worthwhile though.

-------------------------
David
BSc CEng MIET
david@ZawadiSoundAndLighting.co.uk
 22 September 2015 06:51 AM
User is offline View Users Profile Print this message



normcall

Posts: 8514
Joined: 15 January 2005

When you have a device that is marked 40A and 30mA, isn't it reasonable to assume it will switch off at either 30mA earth fault or 40A overload?
If it looks like a circuit breaker, marked like a circuit beaker, it must be a circuit breaker - mustn't it?
Like those old black things marked 60A 240V and voltage operated. When the volts went over 240v they tripped off (along with half the street, so it must be true!).
Must go and take me tablets or I might be worse.

-------------------------
Norman
 22 September 2015 08:51 AM
User is offline View Users Profile Print this message



Jaymack

Posts: 5357
Joined: 07 April 2004

Originally posted by: daveparry1
they tell me they had an electrician in the other day

Should the criticism be directed at "they" or the "electrician" Did they name the firm or person involved and did registration apply?

Regards
 22 September 2015 09:07 AM
User is offline View Users Profile Print this message



mapj1

Posts: 9371
Joined: 22 July 2004

Is it safer to provide RCD protection to the sub main? If yes, are we permitted to make an installation less safe?

Now there is perhaps an interesting one for the legal team

Actually, with my RA hat on, I'd suggest it only makes things safer if there is a credible fault that would only operate a 30mA RCD, but not an MCB, or a fuse, or an RCD of higher rating.
It might be possible in some installations (high Ze for example, or water damage to pyrotenax cable) that a partial cable damage would give a fault that may not remove the supply under MCB or fuse as rapidly as an RCD would. But is it likley ? I can't see the set up - maybe if there is a cable running outdoors or where it can be scuffed or eaten by mice without being noticed. Indoors, not concealed but well out of the way of normal traffic, less likley.

Can we leave the installation less safe ?

In one sense we do every time we add a new socket or light, as there is now more stuff that is permanently live waiting to go wrong...
What we must not do however is break the "double fault to danger /defence in depth" principles, by removing the CPC or the outer insulation, or any equivalent action that may cause the ADS not to operate, or to operate too slowly (for example a circuit with too high a Zs and no RCD.).

So I reckon "yes" and "yes", but with some caveats on both.

-------------------------
regards Mike
 22 September 2015 09:21 AM
User is offline View Users Profile Print this message



daveparry1

Posts: 7728
Joined: 04 July 2007

Thanks for all the replies everyone. I have decided to go there this morning and replace the "front end" RCD with a 100 amp DP switch and the 40 amp mcb with a 50 amp one. I cannot see any reason for the 16mm t/e sub-main to be rcd protected and it's feeding a twin rcd c/unit so all circuits are 30m/a rcd protected.
 22 September 2015 01:03 PM
User is offline View Users Profile Print this message



daveparry1

Posts: 7728
Joined: 04 July 2007

Well it's all up and running now, all the computers, lights, heating and hot water, they are very happy! I would rather have used a cartridge fuse than the 50 amp mcb for discrimination reasons but I was using the existing enclosure so that wasn't really an option. I have told them that they might get some tripping in the consumer unit but hopefully because all the circuits are now shared between two rcd's it will be ok. As there are quite a few computers would it still be ok to re-configure the sockets circuits so that those feeding the computers are not on rcd? I will need to read up a bit on AMD3 notes won't I !
Thanks again for all replies.
 22 September 2015 01:08 PM
User is offline View Users Profile Print this message



ectophile

Posts: 746
Joined: 17 September 2001

Originally posted by: spinlondon

Is it safer to provide RCD protection to the sub main?

If yes, are we permitted to make an installation less safe?


The safest installation is one that isn't even connected to the supply. Does that mean you should never connect any new installation?

-------------------------
S P Barker BSc PhD IEng MIET
 22 September 2015 01:11 PM
User is offline View Users Profile Print this message



mikejumper

Posts: 2416
Joined: 14 December 2006

Originally posted by: normcall
When you have a device that is marked 40A and 30mA, isn't it reasonable to assume it will switch off at either 30mA earth fault or 40A overload?

If it looks like a circuit breaker, marked like a circuit beaker, it must be a circuit breaker - mustn't it?

Indeed.

There did seem to be some confusion as to the exact function of RCD's and MCB's when they started to be more widely used.

That was quite a long time now ago so you'd expect someone working professionally within the industry to have caught up by now.
 22 September 2015 01:16 PM
User is offline View Users Profile Print this message



daveparry1

Posts: 7728
Joined: 04 July 2007

My thinking exactly Mike !
 22 September 2015 01:21 PM
User is offline View Users Profile Print this message



mapj1

Posts: 9371
Joined: 22 July 2004

As there are quite a few computers would it still be ok to re-configure the sockets circuits so that those feeding the computers are not on rcd? I will need to read up a bit on AMD3 notes won't I !


if it is either surface wired or very deep in the wall, (or armoured ..)then the simplest might be RCD sockets on a per machine (or per cluster) basis, and then just MCB at the board. (Or a even cheaper, a risk assessment that permits the computer sockets not to have RCD, - contentious option that. )

Or if the cable installation method precludes this then re-group it into a few smaller separate final circuits on an RCBO each.

If they really do have more than 10mA of leakage, it may be worth asking if it should be using some form of high integrity earthing if the CPC is <4mm. (rings are fine )

certainly if they add more IT equipment then their problems may recurr.

-------------------------
regards Mike
 22 September 2015 05:17 PM
User is offline View Users Profile Print this message



cptspalding

Posts: 1
Joined: 25 July 2008

Originally posted by: daveparry1

As there are quite a few computers would it still be ok to re-configure the sockets circuits so that those feeding the computers are not on rcd? I will need to read up a bit on AMD3 notes won't I !


As stamapj1 says, if you want to take the computer circuits off the RCD then you'd need to perform a risk assessment. IMHO these are simple office workers ups of coffee on the desk, usual crap under the desks, cables probably being kicked around accidentally below desks, I can't see you getting away with taking them off the RCD.

I have my own rule of thumb that I don't put more than 10 PC's on the same RCD. If there is continued tripping you may have to consider individual RCBO's on your final circuits.
 22 September 2015 05:28 PM
User is offline View Users Profile Print this message



daveparry1

Posts: 7728
Joined: 04 July 2007

Yes that might be a way around it if there is still a tripping problem. I have told them to try it out for a week or two just to see if there is still a problem. At least they won't be losing everything now that the "front end" rcd has gone !
 04 October 2015 10:18 AM
User is offline View Users Profile Print this message



daveparry1

Posts: 7728
Joined: 04 July 2007

Hi all,
Just an update on this one. After having a good look at the run of the sub-main I decided it didn't need rcd protection so I fitted a 100 amp DP switch (consumer unit type) and 50 amp 60898 mcb in place of the 30m/a rcd. That was about 10 days ago and they say they've had no tripping problems at all, in fact they're over the moon that they now have all 12 computers, lights, and hot water/heating on at the same time! All final circuits 30m/a rcd protected at the c/unit but now spread across two rcd's of course
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2017 FuseTalk Inc. All rights reserved.

..