RCD with non sin wave

[rasklit]

Thats my problem , the inverter is floating

 

[ericmark]

Boat now sold so no longer my problem. However I seem to remember that neutral and earth were bonded.

We never tested the system as we were uncertain as to if the test instruments could damage the inverter.

However since the steel hull is sitting in water there was not really an issue with earthing it was the wave form which was an issue.

Seem to remember earth was connected through a diode while on shore power. This is common with boats the silicon diodes have a threshold voltage of 0.6 by using 4 it gives a 1.2 voltage barrier to stop electrolysis.

 

[securespark]

20A DP switches on a 32A circuit anyone?

That's no different to 2A sockets on a 6A lighting circuit: perfectly acceptable as long as the load drawn is 2A or less.

 

[SUNRAY]

It isn't ideal what he has done, it was just to prove a point between sparks. The bank of sockets is upstream from the RCD hence the N-E link was placed there as the N-E link.

Ideally the N-E link would be in the inverter as it is source of supply and ought to have an earth rod tying it down to mother earth. As you may find with small sources they are often left electrically floating or "separate".

123 your spot on.
The RCD tester will not work if the N & L were left isolated from E and the simplest way to achieve this for these tests was to use the shorting plug that we use when doing resistance tests. I must confess this was a long time ago and I had to think hard about it.
Interestingly I had to do some testing on a big 3 phase MSW UPS inverter last summer[I guess around 75 kw as it had 100A fuses] and we had to change all the 10 year old MEM RCD's to Merlin.. oops Schneider [about 20] as they would not trip within time.

 

[SUNRAY]

It isn't ideal what he has done, it was just to prove a point between sparks. The bank of sockets is upstream from the RCD hence the N-E link was placed there as the N-E link.

Ideally the N-E link would be in the inverter as it is source of supply and ought to have an earth rod tying it down to mother earth. As you may find with small sources they are often left electrically floating or "separate".

123 your spot on.
The RCD tester will not work if the N & L were left isolated from E and the simplest way to achieve this for these tests was to use the shorting plug that we use when doing resistance tests. I must confess this was a long time ago and I had to think hard about it.
Interestingly I had to do some testing on a big 3 phase MSW UPS inverter last summer[I guess around 75 kw as it had 100A fuses] and we had to change all the 10 year old MEM RCD's to Merlin.. oops Schneider [about 20] as they would not trip within time.

 

[ericmark]

20A DP switches on a 32A circuit anyone?

That's no different to 2A sockets on a 6A lighting circuit: perfectly acceptable as long as the load drawn is 2A or less.

Not quite the same. All lamps/bulbs should have internal fuses so if there is ionisation when the bulb blows it will take out the fuse in the bulb. I realise some bulbs from china don't comply but in theroy they should have fuses built into the bulb.
We all know 434.2.1 allows us 3 meters between the reduction in current carrying capacity and the protective device so spurs can be 3 meters long before we need to add a fused connection unit.

SUNRAY is correct. In the main the use of underside cable to feed the protective device is for where we take a tapping from a bus bar unless the bottle fuse system is used one does not really have an option. However the same rule does allow one to take a spur from a ring main. However the 3 meter rule still applies.

 

[SUNRAY]

In the main the use of underside cable to feed the protective device is for where we take a tapping from a bus bar unless the bottle fuse system is used one does not really have an option. However the same rule does allow one to take a spur from a ring main. However the 3 meter rule still applies.

But do the regs actually place a 3m limit on 2.5mm spurs on ring mains? I have certainly seen many which are well over 3m in length.

My big gripe is that a double socket has 26A worth of fuses which will happily work to around 35A or so.

A classic example is a spur installed for a washing machine and a dishwasher [of say German manufacturer] which come with moulded 13A plugs
Looking at the rating plate: 220V 3000W [quite typical]
Doing the sums:
1) 3000/220=13.6A which is a 4.6% overload of the plug and fuse
2) 220x220/3000=16.2 ohms
3) 240/16.2=15A which is a 15% overload of the plug and fuse
So with the pair of these runing on a 2.5mm spur the cable gets very warm.
Looking at real situations, my voltage now is 248V
248/16.2=15.3A which is a 17.7% overload of the plug and fuse and spur cable [depending on manuacturers spec] and could be a massive overload due to derating factors.
It gets worse, recently I was in an office welfare area and the double socket had two adapters and plugged in were two 3KW kettles, dishwasher, fridge and microwave, with a total possible load of around 9.5KW. This was running on a 2.5mm fp200 spur on a ring main which was also powering the remainder of the office, and had 'never caused a problem'

 

[ericmark]

I have argued in the past there is nothing in 433.2.2 which says this does not apply to a ring circuit and spur so I feel it does apply.

Long spurs will have a problem complying with the loop impedance requirements and I have also argued that the line - neutral impedance is also important both to ensure the magnetic part of a MCB can operate and also volt drop.

The other problem is where the impedance is below 1.44 ohms as with 16 amp radials is extension leads with normal 13A fuses would need to have less than zero resistance for it to rupture within the prescribed time.

The half conversion to Europe's lower standards can have some unexpected results. One items which I think we really need is a FCU able to take a 16A fuse so ovens, washing machines etc designed for Europe can be used in the UK without a part re-wire.

As to washing machine I can't see the difference in a machine which is fixed within a space left for it and building it into a cupboard they are to all intent and purpose fixed. So washing machine (with the exception of old top loading twin tub), tumble drier, dishwasher, oven, and immersion heater all come under the same rules. Appendix 15 refers to regulation 433.1 and points out items over 2kW should have their own dedicated radial circuit.

I have questioned many times the idea of having 20A switches connected to a ring and supplying fixed items in the kitchen area. Since we should really wire the fridge and freezer in ali-tube so RCD protection is not required one has to question if rather than two 2.5mm cables feeding the grid switch should each switch in the bank have it's own feed?

 

[Spark123]

It gets worse, recently I was in an office welfare area and the double socket had two adapters and plugged in were two 3KW kettles, dishwasher, fridge and microwave, with a total possible load of around 9.5KW. This was running on a 2.5mm fp200 spur on a ring main which was also powering the remainder of the office, and had 'never caused a problem'

I had to replace a spur which had a single to double adaptor double socket on the end of it. Plugged into it was a 2 way block. It was supplying a washing machine, dishwasher and a tumble drier
Got a little bit warm and melted the socket. Wire seemed fine tho.

In essence, a double socket isn't meant to have two 13A loads on it, as others will agree the washing machine tumble drier scenario finishes off a lot of double sockets!

I think 20A DP switches are built to a BS number which is not on the list of permitted BS numbers in the reg for ring final circuits.

 

[SUNRAY]

I have argued in the past there is nothing in 433.2.2 which says this does not apply to a ring circuit and spur so I feel it does apply.

Long spurs will have a problem complying with the loop impedance requirements and I have also argued that the line - neutral impedance is also important both to ensure the magnetic part of a MCB can operate and also volt drop.

The other problem is where the impedance is below 1.44 ohms as with 16 amp radials is extension leads with normal 13A fuses would need to have less than zero resistance for it to rupture within the prescribed time.

The half conversion to Europe's lower standards can have some unexpected results. One items which I think we really need is a FCU able to take a 16A fuse so ovens, washing machines etc designed for Europe can be used in the UK without a part re-wire.

As to washing machine I can't see the difference in a machine which is fixed within a space left for it and building it into a cupboard they are to all intent and purpose fixed. So washing machine (with the exception of old top loading twin tub), tumble drier, dishwasher, oven, and immersion heater all come under the same rules. Appendix 15 refers to regulation 433.1 and points out items over 2kW should have their own dedicated radial circuit.

I have questioned many times the idea of having 20A switches connected to a ring and supplying fixed items in the kitchen area. Since we should really wire the fridge and freezer in ali-tube so RCD protection is not required one has to question if rather than two 2.5mm cables feeding the grid switch should each switch in the bank have it's own feed?

There's a lot of good stuff in there, most of which I agree with.
A spur still comes under the circuit design rules but as a special case, so very long runs should not be permitted...
I do a lot of work with extension leads for public entertainment events and am constantly shocked by what I see others do. 3KW urn on several 10m 1mm leads, across walkways etc, equally they are amazed when they see a proper way to do the job.
Under 16th it was permissable to use a 13A fuse or 16A mcb off a ring but have not checked if thats still the case. Of course here's nothing to stop one wiring a house to a member states regs [as long as they equal or excede UK's regs] with 16A Shuko's then many of the problems will diminish.
My daughter has a new build [now 4 years, new?] and the kitchen has a ring for the sockets and a grid switch containing 6 DP switches and 6 13A fuse holders, two ar fed with a 20 mcb and 2.5mm for the oven and gas hob igniter and the other 4 are fed with a 2.5mm ring and 32A mcb to feed O/L plates for: 4 appliances, at times W/M, D/W, fridge freezer & tumble drier are all on so 9KW on a 32A breaker and thats a recent design [guess under 16th]. My arguement here is, as you say, they are fixed loads as O/L plates were fitted and should have had individual feeds alhough we did replace the plates with single sockets which sort of reverts back to portable.

It gets worse, recently I was in an office welfare area and the double socket had two adapters and plugged in were two 3KW kettles, dishwasher, fridge and microwave, with a total possible load of around 9.5KW. This was running on a 2.5mm fp200 spur on a ring main which was also powering the remainder of the office, and had 'never caused a problem'

I had to replace a spur which had a single to double adaptor double socket on the end of it. Plugged into it was a 2 way block. It was supplying a washing machine, dishwasher and a tumble drier
Got a little bit warm and melted the socket. Wire seemed fine tho.

In essence, a double socket isn't meant to have two 13A loads on it, as others will agree the washing machine tumble drier scenario finishes off a lot of double sockets!

I think 20A DP switches are built to a BS number which is not on the list of permitted BS numbers in the reg for ring final circuits.

Sadly I have had to replace possibly hundreds of 13A sockets due to overheating, and not all are due to overcurrent. Also had to replace too many 2.5mm spur cables [with 4mm, my standard] due to overheating
If a double socket cant cope with 2 loads of 13A then surely it is unfit for purpose. I don't care what the spec is or what's written on the back we can't expect the user to not plug 2 items in.
As long as the switch is protected by something 20A or less I do't see an issue.

 

[JohnW2]

But do the regs actually place a 3m limit on 2.5mm spurs on ring mains? I have certainly seen many which are well over 3m in length.

As I see it, whether or not 433.2.2 applies at all in this situation depends upon whether one regards the transition from the ring to the spur cable as "a reduction in CSA".

However, even if it does apply, 433.2.2 can be satisfied in one of two ways - EITHER by limiting length to 3m and 'installing in such a manner as to minimiise rsik of fire or danger to persons' OR by satisfying the requirements of 434 for protection against fault current.

So, provided the fault current protection requirement is satisfied (which it surely should be?) it seems that the first option, imposing the 3m limit, doesn't apply - so you could have as long a spur as you like.

Of course, the OSG recommends ('as a rule of thumb') that unfused spur length should not exceed one eighth the cable length from to spur to the furthest point of the ring - but that's a different matter.

That's how I see it, anyway!

Kind Regards, John.

 

[SimonH2]

Thats my problem , the inverter is floating

You link it's "neutral" to earth and it's no longer floating.

As to your other thread, you got it locked a bit quick. The inverter would need it's own reliable earth*, and for the RCD to offer any protection AND for safety since the loads will all be single pole switched, you'd want to link it's neutral to earth.

* The incoming MET would be an obvious place - but bear in mind that if the supply gets disconnected, then so does your earth. The difference is that with the normal mains supply, if the supply is off then you have no power so lack of earth is a moot point.

The obvious course of action is to consult a qualified electrician - but I know quite a few who would respond with

 

[SUNRAY]

Thats my problem , the inverter is floating

You link it's "neutral" to earth and it's no longer floating.

As to your other thread, you got it locked a bit quick. The inverter would need it's own reliable earth*, and for the RCD to offer any protection AND for safety since the loads will all be single pole switched, you'd want to link it's neutral to earth.

* The incoming MET would be an obvious place - but bear in mind that if the supply gets disconnected, then so does your earth. The difference is that with the normal mains supply, if the supply is off then you have no power so lack of earth is a moot point.

The obvious course of action is to consult a qualified electrician - but I know quite a few who would respond with

Sadly there are far too many 'electricians' who go in the Controls environment. Or rather they don't ask but do the job and someone has to rip out the mess they make and do it properly.I guess this is the case with PV and wind too.

 

[ban-all-sheds]

Under 16th it was permissable to use a ... 16A mcb off a ring.

No it wasn't.

 

[ban-all-sheds]

I think 20A DP switches are built to a BS number which is not on the list of permitted BS numbers in the reg for ring final circuits.

BS EN 60669.

So put them on a radial.