Bad electrical design

[SUNRAY]

Wow you fitted a lot in here

Since we are told we must fit two 30 mA at 40 mS RCD's in series with boats and caravans, can't really say it is wrong to fit two in series, even if a little pointless, you could always change garage one to 10 mA.

As to MCB where it is likely you can overload then having some discrimination may be prudent, my caravan has a 10A and 6A MCB which is fed from a 16A outlet which means if I put the kettle on with heater running in error I can reset without going outside. But before you go up in a rating you would need to inspect and test to see if you can without changing some thing else. I have seen many lighting circuits where the 6A has been swapped for at 10A however although the cable can take 10A, and your permitted up to 16A for a lighting circuit, the ceiling rose which acts as a junction box is only rated 6A (although sure it can handle a lot more) so technically you should not fit a 10A MCB in lighting circuits which have 6A ceiling roses.

So you would need to check on the ELI before you can change to a larger MCB.

As to volt drop and if the appendix is considered as a regulation well when one takes the C&G2382 exam they do ask questions about items in the appendix so it would seem C&G consider the appendix as part of the regulations. However in real terms looking at fluorescent lights. With the old wire wound ballast the fluorescent lamp was very sensitive to voltage, a 230 volt fitting can draw well over rated current if fed with 254 volt, and can fail to strike at 215 volt. I had a 240 volt 65W fitting in my kitchen, supply to house was around 245 volt, with a 58W tube as 65W discontinued it would strike most days, but then it stopped working all together. On testing voltage now house is at 230 volt, seems some solar panels have been fitted in the street and so they don't lock out on over voltage the voltage needs to drop.

This used to be quite a big issue, I did some exploratory work in a factory where the power ratings and power cost didn't tally, there were 2 similar buildings strewn with the same set of twin 8ft 125W fittings which ran 24/7, one was used as the factory with machinery and the other for storage. The metered power for both buildings was similar until the factory closed for a number of months for a refit, there was a bigger than expected reduction in power. We checked out the lighting and found the storage area was running on a much higher voltage, well over 240, and the factory was well under 230 [different sub's] and the increase in power was not linear with the increased voltage. We also concluded one phase of the factory was very near the voltage at which the lights extinguish at times and need more frequent tube changes.

This has been talked about a few times, we can test incoming supply and it may be 240 volt with an ELI of 0.20Ω but without telling you, the supplier can change that to 230 volt with an ELI of 0.35Ω as for ELI with all RCD protection not really a problem, but volt drop is another matter.

Been caught out with this where a circuit was designed for the measured ELI figures and it was marginal, DNO did some alterations (moved cables for a road development) and the ELI went up making the circuit non compliant.

I have been caught out a few times with volt drop, but fluorescent fittings and transceivers the latter receives OK with volt drop, but on transmit it can have mains hum which one is unaware of. OK licensed ham should know better and CB limited to 4W and pre switch mode power supply radios, or record players you can hear the main hum as voltage drops so well aware of problem.

Did a disco in a church hall a good mile of pole route from the village many many years ago and supply voltage as flaky as hell, the mains hum in the speakers was significant and I really thought my kit had developed a fault and tried providing my own earth spike and isolating from halls earth, the music was even modulating the lights in the church.

As said other than pre-HF ballast fluorescent fittings there is very little likely to fail due to volt drop. In the main the modern SMPS has removed the problem with volt drop. However one tends to use old equipment in garages, you buy a nice new radio alarm for bedroom, and old one it taken to garage.

I think fluo fittings are running much hotter than before as they are now designed to strike at the lower voltages but generally still run >240V, so much for improved efficiency .

 

[ericmark]

With old magnetic ballast I also found power used not as stated on tube. We got a load of 110 volt fluorescent fitting for lighting the below ground areas when building Sizewell 'B' power station, rated 58W so quick calculation 2 = 116W and 110V so around 32 on a 16A supply, didn't want to be on the edge so 25 x 58W fitting connected to a 16A supply. About 2 hours latter a message lights off, found MCB off, though some one messing around switched it on again. Same again tripped, then took lid off transformer and used clamp on ammeter, drawing 20A.

So got a fluorescent fitting on the bench and started to test it. The auto transformer was not centre tapped but marked 110 - 0 - 128 so moved tapping from 110 to 128 and current dropped from 0.8 amp to 0.6 amp, so opened up each fitting and swapped tapping, now last few would not strike, so last 5 swapped back to 110 volt and total of 25 lights now drawing 16 amp.

I did further tests, losing the capacitor could make the current jump to over 1 amp, so after that where I worked measuring the current used by fluorescent lamps was a regular thing, if current went up, then time to replace capacitors. However as the HF ballast hit the market place, it was found the 58W fluorescent actually used 58W did not matter if voltage 220 or 240 wattage remained static. Also tubes lasted longer, however a faulty tube could take out the ballast, so it was important to change tube the moment it started to fail, the old idea of wait until 25% had failed then change the lot no longer works, it costs too much in new HF ballasts.

When comparing a fluorescent with LED with magnetic ballast the LED looks good, but with HF ballast the fluorescent lasts longer than LED and costs a lot less to replace, and 95 lumen per watt to 100 lumen per watt just not worth changing, fluorescent more cost effective.

Had the LED tube lasted the 10,000 hours claimed then maybe, but one I tried died after 1000 hours so fluorescent wins.

 

[SUNRAY]

With old magnetic ballast I also found power used not as stated on tube. We got a load of 110 volt fluorescent fitting for lighting the below ground areas when building Sizewell 'B' power station, rated 58W so quick calculation 2 = 116W and 110V so around 32 on a 16A supply, didn't want to be on the edge so 25 x 58W fitting connected to a 16A supply. About 2 hours latter a message lights off, found MCB off, though some one messing around switched it on again. Same again tripped, then took lid off transformer and used clamp on ammeter, drawing 20A.

So got a fluorescent fitting on the bench and started to test it. The auto transformer was not centre tapped but marked 110 - 0 - 128 so moved tapping from 110 to 128 and current dropped from 0.8 amp to 0.6 amp, so opened up each fitting and swapped tapping, now last few would not strike, so last 5 swapped back to 110 volt and total of 25 lights now drawing 16 amp.

I did further tests, losing the capacitor could make the current jump to over 1 amp, so after that where I worked measuring the current used by fluorescent lamps was a regular thing, if current went up, then time to replace capacitors. However as the HF ballast hit the market place, it was found the 58W fluorescent actually used 58W did not matter if voltage 220 or 240 wattage remained static. Also tubes lasted longer, however a faulty tube could take out the ballast, so it was important to change tube the moment it started to fail, the old idea of wait until 25% had failed then change the lot no longer works, it costs too much in new HF ballasts.

When comparing a fluorescent with LED with magnetic ballast the LED looks good, but with HF ballast the fluorescent lasts longer than LED and costs a lot less to replace, and 95 lumen per watt to 100 lumen per watt just not worth changing, fluorescent more cost effective.

Had the LED tube lasted the 10,000 hours claimed then maybe, but one I tried died after 1000 hours so fluorescent wins.

LED tubes in place of fluo just don't make sense to me, the light O/P is not as good IN MY OPINION (I know others have a different view and I won't bother getting into discussion on it) and they cost no less to run, they cost more to replace...

We have gone through the exercise of changing fluo to LED and back again several times now in some fairly big installations.

 

[bernardgreen]

For what it is worth.

A council commisioned a project to detect failed or vandalised flourescent lamps in pedestrian under passes without the need for council employees to have to go into the underpass.

Measure the amount of current used by lighting in an underpass and from rate of change of current it should be possible to detect that a lamp had failed. Measuring and and logging was simple current transformers fitted on the supply leads to the individual underpass

A tube could be smashed without making any significant change to the current taken by a group of 20 or so lamps. Nothing was consistant enough ( in terms of change of current ) to detect failed / vandalised lamps.

 

[JohnW2]

LED tubes in place of fluo just don't make sense to me, the light O/P is not as good IN MY OPINION .... and they cost no less to run, they cost more to replace...

I personally think that makes good sense IF/WHEN the fluorescents are properly specified. However, as I've said before, for what it's worth, my personal experience has been that fluorescents are often over-specified (including by myself) - i.e. unnecessarily over-powered. I suspect that's probably because one tends to choose them by length, so as to illuminate, say, a workbench, room or garage etc. of a certain length and, having chosen the length, one doesn't really have any choice about power (only whether to have a single or double tube fitting).

As a result, whenever I have changed fluorescents to LEDs (roughly half the power and half the light output), I have found the amount of light has been more than adequate. In other words, I have roughly halved the running costs, not because LEDs are more efficient, but because I was able to halve the amount of electricity used (and light produced) in way that I generally could not do by staying with fluorescents.

In cases where I've had (usually inherited) double fluorescent tube fittings, I've achieved the same by sticking with fluorescents and removing (or, usually, disconnecting, which is aesthetically better!) one of the tubes.

Kind Regards, John

 

[ericmark]

You make a good point, yes in a corridor you don't really want a lot of light, but an even light over its length, so with fluorescent 58W per 6 meters as any bigger gap leaves dark area, but with LED 20W per 6 meters is ample. However with a ware house some one has gone to a lot of trouble working out the light level, to then reduce the light by half is not really an option, with this example the 5 foot unit uses 50W and has a 4500 lumen output, it is claimed a 58W fluorescent tube with a HF ballast actually used 51W so this LED unit is within a watt of the fluorescent fitting, with the initial lumen of 5200 lumen allowing for the lumen to drop it still does not compare. However the advert clearly states

With multiple wiring options and a smooth and even light distribution it provides up to 50% energy saving against T8 fluorescent.

Note used same supplier so can do a good reference.

What it seems is the LED batten is being compared with a magnetic ballast fluorescent which have been discontinued, also compared with halophosphor tubes rather than new triphosphor tubes so it is not really valid.

So fluorescent tube 5 foot £2.60 and HF fitting £14.50 the only LED I could find is IP65 so price is a little steep at £32 however one would also need to bring all the runs closer and fit extra units to get same lumen output.

Before fitting one I would have said less maintenance with LED, however bitter experience is they don't last as long as fluorescent, and since with the one selected you can't replace the tube it is a sealed fitting with 3 year warranty (19080 hours) with fluorescent tube 20000 hours with magnet ballast, HF ballast will increase both life and lumen output so 32,000 hours and 5980 lumen for fluorescent and 19080 hours and 4500 lumen for LED.

How can anyone say LED is better than fluorescent? Using data from TLC it would seem fluorescent wins on all counts.

 

[JohnW2]

You make a good point, yes in a corridor you don't really want a lot of light, but an even light over its length, so with fluorescent 58W per 6 meters as any bigger gap leaves dark area, but with LED 20W per 6 meters is ample.

Exactly my point but, as I said, I would go a lot further than just corridors. Since one tends to think of/specify in terms of tube length, even for illuminating a work surface, garage, or whatever, using fluorescents can force one to use about twice the power (and about twice the lumens) that are actually needed. I presume that there is a technical reason why they never offered fluorescent tubes with a range of wattages for a given length.

However with a ware house some one has gone to a lot of trouble working out the light level, to then reduce the light by half is not really an option ...

That may be true, but this is a DIY forum and I doubt that many of those here are concerned with the lighting of warehouses.

How can anyone say LED is better than fluorescent? Using data from TLC it would seem fluorescent wins on all counts.

Maybe in the 'warehouse' scenario - but, as I've said, in a domestic situation use of LEDs may result in an energy saving which could not be achieved with fluorescents, good for both the pocket and the environment.

To illustrate my 'over-specification' theory. my (very large) cellar came with multiple 2 x 58W 5ft fluorescent fittings (primarily over workbenches etc. I replaced each of them with 2 x 24W LED, with individual switching of the two tubes. I have found that, in most cases, and for most purposes, having just one of the LED tubes on results in more than adequate illumination - hence 24W per fitting rather than 116W. Over the five fittings, that's the difference between 120W and 580W (aka around £1,500 difference in running cost over 20,000 hours).

Kind Regards, John

 

[mattylad]

If the Op can fit an MCB before the existing RCD to supply the outbuilding, then why not fit an RCCBO in its place that will then give the entire supply feed RCD protection as well as overcurrent without affecting the inside circuits?

 

[JohnW2]

If the Op can fit an MCB before the existing RCD to supply the outbuilding, then why not fit an RCCBO in its place that will then give the entire supply feed RCD protection as well as overcurrent without affecting the inside circuits?

He could, but since he already has an RCD in the garage, it would probably be simpler to do as he suggested and simply move the garage supply to the non-RCD side of his house CU (the SWA does not need RCD protection). Furthermore, if there is to be only one RCD (or RCBO) protecting the garage supply, it would (once in a blue moon!) be 'more convenient' for it to be in the garage.

Kind Regards, John

 

[ban-all-sheds]

I may have missed it, but did you replace the entire lights, or just swap the tubes out for LED ones?

 

[JohnW2]

If that question is directed at me, the answer is somewhere between those two, but essentially/effectively the former. I gutted the original fittings (other than for the tube connectors etc.!) and substituted LED tubes for the fluorescents.

 

[ban-all-sheds]

But you didn't go out and buy new luminaires. I wonder what the payback period is if one does, particularly in a location where the lights aren't used very much

 

[JohnW2]

I didn't. However, since there's virtually nothing in an LED luminaire other than the tubes, 'budget' ones do not cost all that much more than the tubes alone. Although one might have to wait a long time to win financially, one might have at least saved a few trees (or branches thereof), and a little CO2, in the meantime.

 

[plugwash]

Since we are told we must fit two 30 mA at 40 mS RCD's in series with boats and caravans

From the site side you have unarmoued flexes being used outdoors by untrained personell in an environment where vehicles are moving around, I would sure as heck want those to be RCD protected.

From the caravan side the caravan may be taken to all sorts of different locations in the UK and abroad, so you can't rely on the site operators working to any particular standards.

 

[ericmark]

From the site side you have unarmoued flexes being used outdoors by untrained personell in an environment where vehicles are moving around, I would sure as heck want those to be RCD protected.

From the caravan side the caravan may be taken to all sorts of different locations in the UK and abroad, so you can't rely on the site operators working to any particular standards.

In the early days, yes there were some UK sites where RCD protection was not installed, not worried about sites abroad they don't follow BS7671, however two points, one is very rare not to have RCD protection on the sites today, and second is we should not need to protect from some one braking the rules.

Same rules apply to boats, however with boats I can see a reason, to stop electrolyte erosion of the hull, it is common to fit isolation transformers, so the shore RCD will not protect the person in the boat. Also think with boats RCD is a silly name, ELCB-c or any other name which can't be confused with recreational craft directive or RCD for short.

I have not tested, so unsure what would be the result in fitting diodes in the earth connection?

So yes good reason with boats, but caravans not really required for safety, but required by regulations.