Light bulbs blowing after only days

[Ron Lowe]

Hi,all.

We have a staff house in town, which is the accomodation on a couple of floors above the shops on the high street. This was all fully refurbished for us, including the electrical system, by (presumably ) reputable proffessionals.

The lightbulb failure rate is astronomical.

We have tried normal lamps, and the energy-saver flourescent jobs. They are lasting around a week!

All lampshades have been removed, to eliminate any possible overheating issue. We've just got bare bulbs dangling from pendants.

The electricians havn't found anything wrong with the installation, and I've eyeballed the electrical installation too. There's nothing obviously wrong. There are several 6A lighting circuits, for various parts of the building, just as you'd expect.

I can't really imagine what kind of fault within the building's wiring could cause this.

I can only think of high-voltage spikes or other noise on the mains, perhaps induced by some impropperly - suppressed large machinery controlled by triac switching somewhere close by? But it's a light commercial area, not an industrial area. The RMS voltage reads 240 to 250 depending on time of day, with a calibrated fluke meter, so the basic voltage is OK.

I've suggested hiring a Power Quality Analyser from someplace to log the supply voltages and spikes, harmonics, etc, becuase without some hard evidence of what's going on, we're just shooting wild guesses.

Any other suggestions ?

[Jonathan Schneider]

Before you go hiring expensive equipment or professional exorcists how about using the min-max feature of many DVMs.

Do things fail at a particular time of week ? If not it ought to be possible to measure something interesting very quickly.

Why not go chat to some of the nearby shopkeepers and residents in case they are also suffering.

Are the problems confined to the lighting circuits or one lighting circuit ? Would a desk lamp on the socket circuit suffer in the same way ?

Are you by any chance getting all your noname brand light bulbs from the same source ? If so perhaps they're dodgy.

Jon

[Ron Lowe]

I'm not sure the max/min feature has a fast enough response time for the kind of glitch that may be happening. But yes, we could do that. Good idea.

Not that we have been able to determine. We rather get the info second-hand from the people who live there. And they are a transient bunch of students who come to our training school for a few weeks at a time.

Have done, and they are not aware of anything.

Now, that's an interesting question. I don't have hard data, but I think the answer is that the bedside lamps are not failing. Which points to something in the lighting circuit, I think.

It also seems that the worst offenders are lights in hallways and corridors.

Switches? These hallway circuits typically have several lamps per switch. So possibly a bit more arcy-sparky in the switch? I'd bet the developers who refurbished the place almost certainly used the cheapest no-name bulk-pak switches too. Perhaps as a cheap test, I could replace some switches... Hmm again...

That was our first thought. We have had batches from our normal bulk vendors, and from local tesco and asda. And we've had so-called long life bulbs, and the low-power flourescent thingys.

[Christian McArdle]

If you contact your electricity company, they may be willing to supply some logging gear that will measure the supply voltage and spikes over a period of days. The supplier can then work out if there is a supply problem and fix it.

Christian.

[Jonathan Schneider]

It's conceivable a nasty switch could upset the lights it controls but not others on the same circuit.

Foreign long-haired students ? Nothing against these people as such.

Here's a possibility. Rather than have proper mains adaptors somebody is plugging something into a light fitting which is now easier to get to what with the shades removed. When the landing light is switched off by a flatmate your nice low energy lamps get an indictive load dump.

Now I'd imagine a heating hairdryer wouldn't do this because the element would be in circuit but a cold hairdryer with low energy lamps just maybe.

Jon

[Ron Lowe]

Foreign, yes. Long-haired, no.

Na, the ceilings in this old place are so high, we had to buy a special super-extra-long stepladder to change the bulbs! This lives in a maintenance cupboard which is locked.

They'd have to ballance 3 chairs on top of a table, and then stand on someone's shoulders to do this!

The issue has outlasted several changes of students, too. ( and has been present since we moved into the place. )

We also provide a stock of universal plug adapters which we loan out to anyone who needs them.

It amazes me that people visit a foreign country, and are surprised that their plugs don't fit! Our American guests are the worst offenders. " Gee, you guys have different plugs here! " If I'm in a less than charitable mood with them, I just hand them an adapter and say nothing, and wait to see if their laptop power brick is multi-voltage or not, because they never think about that either.

[David Hansen]

On Wed, 21 Jun 2006 14:39:54 +0100 someone who may be "Ron Lowe" wrote this:-

Sadly, most have been multi-voltage, with automatic sensing, for a considerable time. However, it is still worth a try.

How do they get on with trying their US dollars in the shops?

I can only imagine that there is some sort of external spike. Are the shops open in the evening?

I wouldn't put too much trust in what they, or the students, say. The only thing to do is probably to hire a gadget to keep a record of what is happening.

[Dave Liquorice]

OK but 250 is rather high and presumably in the evening when the industry has gone home but the lights are on... Upper limit is 230+10% = 253v.

The voltage here used to hover around 250v and light bulbs didn't last well. Complained to the board, they came out measured, and adjusted the tapping on our pole transformer. We now have voltage around 240 and bulbs have the expected life, like chnaging one is a rare event compared to one a week...

As to short transients I wouldn't expect tungsten lamps to be that bothered due to the thermal inertia of the filament. Surges due to big loads suddenly switching off is another matter.

[Frank Erskine]

On Wed, 21 Jun 2006 13:14:19 +0100, "Ron Lowe" had this to say:

Are the bulbs you're buying rated at 230V or 240V? A 230V bulb operating on 240 will have a much shorter life than it oughta.

[meow2222]

you can put those back now

none

thats the one thing that strikes me about your whole approach here. Until you gather the facts reliably, youre blowing in the wind.

Forgive me for stating the obvious, but do you think they might be swapping their dead room bulbs with working corridor ones to avouid buying replacements? Its not exactly difficult to give someone a leg up to reach a high ceiling.

Lets establish a few facts.

It wont be spikes, as electronic equipment would be dying faster than lightbulbs. A filament bulb can take a hell of a big spike without dying.

If you've used 2 or more brands of bulb it isnt the bulbs either.

Switches and bad wiring wont reduce the life of filament lamps. If anything it would increase it.

Harmonics have nothing to do with it, as filaments couldnt care less about anything but rms voltage.

The fact that your neighbours have no such problem ends any discussion about bad mains supply.

It all leaves only one conclusion. If it were me I'd fit secure fittings, ie a special tool is needed to reach the bulb, and only special bulbs will fit, such as unballasted low energy or even 2' linear fl.

NT

[Bob Eager]

Or a UPS with a PC monitoring it, set to tight margins.

[The3rd Earl Of Derby]

So are you.

He did say the bulbs have been blowing since the place had been rewired.

[Ron Lowe]

Yes, I'm only just getting started on this myself. I havn't been down there for quite a while, I really need to go back and take a closer look. I only did a cursory check with a fluke last time, shrugged my shoulders, and we just asked the electrician to go check it out. That was months ago, and I've just had the handyman whinging at me again, so that's why I need to re-investigate. It's not really my day-job, just one of the various hats I have to wear!

I'm fairly certain they are not swapping bulbs, it really is not leg-up stuff. You'd need to be an Eastern European circus troup to get up to that height onother people's shoulders. Then again....

Inclined to agree.

Possibly agree, I wondered if nasty switching transients might be a problem, if they were repeated often enough.

Well, yes. I was just indicating the kind of parameters these machines monitor. I'm sure that particular parameter is not significant. I'd be interested to see what the supply voltage fluctuations were over a week or so, at the actual ceiling rose.

Hmm, possibly. I don't know how reliable their evidence is. Also, they may be on different phases, which is something that set me thinking again...

Possibly, but like I say, I don't believe they are messing with them.

I've just been driving home, and musing this in my head. I have been able to dream up an obscure scenario......

The incoming cable to this building is 3-phase +N. This splits off to the various premises in the building, including us.

If the 3 phases are well ballanced, the N current is zero. But if the load was not well-ballanced, then there may be significant N current. And that in turn may cause voltage drops along the N line. Which means the N voltage at our distribution board may be slightly elevated, not a true N. For a single phase supply, this would be in-phase with the L current, and would subtract from the voltage available at the load. But in a 3-phase system, this N voltage would actually be an additive value on 1 or 2 of the phases, depending on the exact phase of the N current, which in turn depends on the relative L1, L2 and L3 currents.

Now I think about it, it is possible the lights are on a different phase from the power rings. And that may well be significant. I need to go down there and make some better measurments, and with a closer eye on what phases things are on. I might just also clamp a current meter round the incoming tails...

[unopened]

I thought it was against Regs to feed different circuits (e.g. power & lights and/or upstairs power & downstairs power) in the same residence off different phases - otherwise people can get 415V unexpectedly. I'm not an electrician, but ISTR that asonishment was expressed when someone found out that German regs allowed it. Andrew Gabriel may be along to correct me presently....

Sid

Sid

[Sparks]

That can't be right, or that would mean a property can't have more then one phase!

I would think you can have lighting on one phase, sockets on another, but you are not allowed to have more than one phase's wiring in a single phase item (for example, a normal 13A socket can't have wiring for another phase in the same box)

But I stand to be corrected too!

[Ron Lowe]

Well, our office is supplied from a Merlin 3-phase distribution board, with

3-phase loads obviously coming off all 3, but individual rings and lighting ccts come off different phases.

Indeed, if someone were to put their finger in the L of live socket, whilst sticking their tounge in the L of another ( they'd need a long tounge... ) then yes, they'd get a 415v tingle instead of a 240 one. Or if they drilled into the ceiling with a 240 drill powered off one phase, and drilled through a lighting cable on another, then within the drill, there would be

415 v available...

Perhaps there's a difference in the regs as regards domestic and commercial premises. I also don't know if our hostel premises would count as domestic or commercial. ( I'm in Scotland, so the regs might be different again from English ones, possibly. )

I can't even remember if the feed to our staff house board is 1 or 3 phase! And that's a rather fundamental question. That's why I need to go back and take a closer look.

[<me9]

The neutral was corroding away underground on the 3 phase supply at work causing similar problems (paper insulated aluminium cable - nasty stuff!) and the supplier metered it (conveniently) between the phases and declared it ok. Slightly later it declared itself by the whole cable going open circuit. Then they got round to fixing it. It may be something similar.

[Owain]

There are, or were, rules about separation of phases, and anywhere that had multiphase present (eg consumer units) had to have warning notices.

Not uncommon to find the off-peak circuit to a fan-assisted storage heater on one phase, but the fan circuit on another phase.

Owain

[Harry Bloomfield]

Ron Lowe was thinking very hard :

240 to 250v would seem a little high, combine that with perhaps cheap no name lamps and they may not last long. I once bought a batch of 230v European lamps which lasted no more than a week or two each. Our supply generally measures 240v or occasionally a volt or two over.

As someone else suggested - as a first step use the min/max logging facility included in some meters.

[Dave Plowman (News)]

Hmm. Students selling the good ones and replacing them with faulty?