Weired Energy efficiency bulb problem

heliosuk

Hi there,

Am a Professional automotive engineer by trade who is in the doldrums after being laid off like many thousands of others at the moment.

So you know the scenario, left a list of jobs to do by wifey, one of them read, "fix the second light in the bedroom, bulb on window sill". So go and look, put low energy bulb in, turn on and all OK! Think, ok, job done, she obviously couldn't reach the fitting.

When I'm working, I generally go to bed later than her and am at work before she gets up, so think she has taken bulb out so when I turn light on it doesn't wake her up.

However, last night, go to bed as normal, in the dark and I see a flash. Look again a few seconds later, and it flashes again.

This morning I ask her if this was the original problem to be told yes.

So think that's easy, it's the switch. It looks mickey mouse so must be that. So put new crabtree switch ( it's a one gang fitting) on. Then look and it does it again.

So take switch out and it still does it. Turn off rcd and it will stop. Put in normal bulb and it will stop. So start to study when circuit live again and it flashes every 10 seconds without fail.

To me it indicates the low energy bulbs have a CR circuit in them and there must be some sort of back feed powering the start up circuit of the bulb. Changing the bulb for another won't make it go away either.

Has anyone seen this before or know what's going on and how to cure it?

Thanks as ever.

Regards.

RF Lighting · Posted: Tue Jun 09, 2009 9:13 pm Post Subject:

It's more common than you might expect

http://www.diynot.com/forums/viewtopic.php?t=127623

__________________
I

downlights

mattylad · Posted: Tue Jun 09, 2009 9:18 pm Post Subject:

They do that in my house too.

I have no earth on the lighting circuit for the rooms this happens in, something to do with capacitance & transference from the live wire through to the switched live happening because there is no CPC to prevent it.

Search the site for "flash" & you may find this discussed & explained better.

securespark · Posted: Tue Jun 09, 2009 9:19 pm Post Subject:

I guess it is two way??

__________________
As George Michael once said, "..if you're gonna do it, do it right, right?"

heliosuk · Posted: Tue Jun 09, 2009 9:38 pm Post Subject:

No it's a one way which is why I was amazed that taking out the switch and making the circuit live again it still did it. Started to think of back feeds etc but ruled out because ordinary bulb behaved normally.

heliosuk · Posted: Tue Jun 09, 2009 10:02 pm Post Subject:

Ah ha!

Seem to be on the right track.

Have read the previous posts on this but they all seem to refer to 2 way switch/2 gangs and nothing on one.

Given that some of these posts go back to 2005, surely the bulb manufactuers know of this so why haven't they done anything about it?

Bulbs are made by Phillips.

I will go and buy a new bulb from a different manufactuer tommorrow and quality switch. I'll change one by one and if it is the bulb then an exocett will be launched.

I cannot believe they don't know about this and to my knowledge, and from some of the posts, the fix is easy.

Thanks for the help everyone.

Kind regards.

Steve · Posted: Tue Jun 09, 2009 10:53 pm Post Subject:

Theres nothing wrong with it, stop fussing. It has no ill effect on lamp life.

Then again if the missus dont like it, it has to be fixed!!!

TicklyT · Posted: Wed Jun 10, 2009 9:08 am Post Subject:

The problem is the lamps is energy efficient - it's trying it's best to convert the little bit of energy from the induced voltage from capacitive coupling in the cabling into light.

One solution would be to fit less energy efficient lamps that can't utilise the energy derived from capacitive coupling.

As noted by Mattylad, capacitive coupling is often more pronounced in wiring without a CPC present, so maybe the lamp is commenting on the (lack of) continuity of the CPC in the switch drop wiring.

You've already proved it isn't the switch if the effect continues without a switch connected.

A high value resistor between the switched live and neutral would probably be sufficient to dissipate the induced voltage, but I imagine lamp manufacturers would be loathe to do this as it would not be of any benefit to the majority of users, increase the component count, thus their manufacturing costs and make their product less energy efficient. They probably can't afford that in a market that's giving low energy lamps away with cornflake packets.

__________________
If it isn't broken, keep fixing it until it is

toasty · Posted: Wed Jun 10, 2009 10:57 am Post Subject:

I had this with my utility room light. You can stop the flicker by connecting a mains rated capacitor over the light terminals (i.e connect the capacitor in parallel with the bulb.

I have a bulkhead fitting in the utility room , so just put the capacitor in there connected to the lamp holder connections (with a bit of heatshrink over the bare wire leads)

That stopped the problem for a total cost of some 40p!
A suitable capacitor is this one from Maplin:

http://www.maplin.co.uk/Module.aspx?ModuleNo=98169&C=30869

Another 'solution' is to replace the low energy bulb with a filament bulb

ColJack · Posted: Wed Jun 10, 2009 8:49 pm Post Subject:

if it's a one way switch, there is a way to cure it but it involves using a 2 way switch backwards ( so the live goes to L1 rather than com ) and connecting an earth to L2 ( and so earthing the switch wire when switched off ).. this is not for amateurs

__________________
If you make everything idiot proof, evolution will just make a better idiot.

ColJack · Posted: Wed Jun 10, 2009 8:55 pm Post Subject:

TicklyT wrote:

As noted by Mattylad, capacitive coupling is often more pronounced in wiring without a CPC present, so maybe the lamp is commenting on the (lack of) continuity of the CPC in the switch drop wiring.

can we just clarify..

it's more pronounced in wiring without a CPC in between the switched live and live wires.
the presence of a CPC does not automatically negate the effect if it is not between the live and switched live, such as in conduit wiring or with MICC cables where the sheath is the earth.

__________________
If you make everything idiot proof, evolution will just make a better idiot.

heliosuk · Posted: Thu Jun 11, 2009 12:06 am Post Subject:

Hi Guys,

From this forum it's obvious it's a known problem. Whilst I'm capable of putting in the fixes it's not right that anyone should have to go through all of this to cure it. If I (or anyone else for that matter ) took the time into account to fix it, we'd be better off just whacking in ordinary bulbs for a few months.

The point is, now I now know it's a known issue why don't the manaufactuers do something about it? Afterall it's in their interests to do something about it.

If this issue was on a car everyone would be up in arms.

We are told to replace ordinary bulbs with these low energy ones but wouldn't an ordinary consumer expect it to be a like for like replacement?

It's not right in this day and age especially more so that all the blurb say's we have to go this way and the fact of the matter is they are fundamentally underdeveloped.

To the uninittiated it's called FMEA, Failure, mode, what the affect is and analysis.
If this issue is known about then it can't possibly comply with CE reg.

Get my drift???

bernardgreen · Posted: Thu Jun 11, 2009 6:35 am Post Subject:

Overlooking the effects of stray capacity in domestic electrical wiring is just one more example of the continuing reduction of quality in engineering design processes.

The design of the item may be perfect and meet exactly the requirements of the specification handed to the designer, that part of the process is seldom the problem.

The problem is often the blinkered view of the writers of specifications who do not look for the worse case environments the item will be used in.

Some blame must be attached to those designers who do not question the specification but accept what they are given to work to.

Space cat · Posted: Thu Jun 11, 2009 9:59 am Post Subject:

ColJack wrote:

using a 2 way switch backwards ( so the live goes to L1 rather than com ) and connecting an earth to L2

Better to connect a neutral to L2 if you can get one to the switch.

bernardgreen wrote:

Some blame must be attached to those designers who do not question the specification but accept what they are given to work to.

Having been an electronics design engineer myself, I tend to agree. It always pays to look beyond the written spec and ask a few pertinent questions. If you don't do this, everybody else will blame you anyway when the damn thing doesn't work properly.

The downside is that, having saved the company wads of money by NOT making something that won't work, they still blame you for holding up the job!

__________________
Their base is so secret, even the space cats don't know where they are.