CFL short life

[Julian]

Polonium 210?

[Dave Liquorice]

I know that but 32' or there abouts is not very practical to hang on the hall wall. 32' is roughly the ridge line on a two storey house with a pitched roof...

[Derek Geldard]

It was indeed myself.

And I gave you the reason at that time which was that I hadn't got a stock of part life expired GLS lamps to make measurements from. I have a *life* instead. I had had to buy a single 60 watt (pearl) GLS lamp from Tesco for 16p just to make my measurements. I explained (from memory) that all the lamps in my house are CFLs, and I had made a start about 20 years ago with Phillips Jamjar lamps. I do want the CFL concept to work.

I have no recollection over my lifetime that the light output of GLS lamps deteriorated significantly over their (admittedly short, but cheap) working life. I wouldn't expect them to, over 1,000 hours or less. Whereas the several processes which cause the deterioration of light output in CFLs are well understood.

30% less than they promised out of the box, and a further deterioration of 48% in the first year, is the bottom line.

However I do believe that products should deliver the performance they promise, and at the moment CFL's do not, and I take issue when the government talks of taking powers to outlaw / ban the sale of GLS lamps.

At the current state of the technology this can only have happened if this government of artists and lawyers that we have have been suckered / sweet talked / bribed into believing that which is not true. No surprise there then.

The problem being that if the general public have the alternative GLS option outlawed, then spending on improving CFL technology will yield lower returns and probably will never happen, and we will be stuck with a technology which yields a light output that is 30% down on what was promised when it was sold, which then goes on to diminish by 48% over the first 12 months, and which has a median lifespan of about 18 months, not "8 Years" as implied on the packaging of the Phillips E-ON lamp I have here.

Do I detect here an attempt at a "proof by selected instances" ??

Then they weren't "Compact" and they didn't have electronic ballasts, as they have today, and which would have given them a higher efficiency. They would have had wound inductive ballasts, which by and large don't fail. The ballasts would probably not have failed but the phosphor would have been worn out years ago, and the tube ends blackened further reducing their measly, miserable light output, and filaments / heaters / cathodes would have failed at random intervals.

20 years my arse !

But, you see, it's no good asserting that. Everybody knows they don't.

When you mention "Well over 20 years" and "A very long time" you are dissembling. The CFLs you buy today have an electronic ballast in the tube base, this will include a power rectifier and several high voltage miniature electroytic capacitors amongst other high power components. It is beyond credibility to expect these to last about 20 years in routine daily use. They don't, at the high temperatures experienced in light fittings (about 80 C) they last about 12 - 18 months.

This is not news, it's not even history, it's everybody's common experience of mains powered electronic devices made over the last 20 years.

Oh and the technology of semiconductor diodes and of electrolytic capacitors has not improved AFAIK over the last 20 years, and certainly not to the extent that a Chinese sweat shop can knock them out cheap enough that CFLs can be sold here in the UK against a tarriff of 10% (??) at 2 for a pound. Whereas reliable GLS lamps made in Europe which deliver the performance they promise can be sold at

16p. So, you replace 2 or 3 per year, big deal. They can probably be obtained cheaper then that anyway if you are into skinnin' turds for ha'pennies.

HTH. DG

[Geoff Berrow]

Message-ID: from Andy Burns contained the following:

Umm, not really practical. The idea is to create a vacuum at the top of the column of liquid. However, water boils at room temperature in a vacuum.

[Andy Burns]

I think you may have missed my humorous quotation marks ...

[Dave Liquorice]

No, you can make a barometer using water. The space at the top is not a vacuum, water (or Hg) does vaporise into the space.

Remember it is not the vacuum "sucking up" the column of fluid but the air pressure pushing it up from the bottom, until force from the weight of the fluid column equals that of the force from the air pressure. This is why the column height varies with the air pressure.

[meow2222]

the reason is irrelevant, the point is you failed to compare the 2. You didnt present the relevant data.

seems to work ok here

The mechanisms of light output reduction in filament lamps have been known for over a century.

I recall you failing to measure light output from various angles, and as is known output varies in differing directions with CFLs. IIRC this was brought to your attention before.

If youve had 48% drop in a year, I suggest a) you clean it before remeasuring b) you clean the surrounding items that contribute reflected light Or alternatively, dont buy junk lamps

The packet equivalence figures are compared to soft lamps, not GLS. It is misleading, not incorrect.

I'm with you on that one

all wrong

varies widely according to cfl, and lifetimes are measured in hours not years, since they depend entirely on hours of use.

So you bought junk.

Packs claiming in years normally say how many hours use per day the claim is based on. I've not seen any evidence of such claims being incorrect.

Refuting your claim. Honestly

In fact the glow starters are failure prone.

it wasnt

not significantly

no evidence or reason then

nobody here claimed they would. You can buy ones that do, but those arent usually used in homes.

Many of my fittings dont get anywhere near 80C. A few barely hit 30.

depends on use obviously

You must think me a rare exception then. I dont.

Shottky diodes, supercaps, higher manufacturing yields, and closer manufacturing tolerances leading to improved reliability. But other than that.

Is that what youve been paying! No wonder you ended up with junk.

Let us know when you understand the subject.

NT

[Derek Geldard]

It is perfectly valid observation to measure the initial shortfall in the output of CFL's, and then meaure their rate of deterioration over time.

That's Hansen's Strawman, nowt to do wi me, I never said I did.

No, you are flat-earthing for CFLs, there is a difference, I want them to be actually *workable*.

BzzzzT, No commercial Tungsten lamps 'til 1911. ;-)

Anyhow this isn't a pissing contest over the oldest reports of lamps deteriorating. The simple fact is reports of CFLs failing have been flat earthed, we have now established they do fail and how, and why, That's at least *some* peogress.

I used a regular lampshade as a diffuser, taped the sensor from the lightmeter onto it so it didn't move and the experimental setup was not disturbed between measurements of new and old, and the whple set of measurements was then repeated. The 48 % figure is surprisingly consistent amongst a wide range of supermarket / IKEA (*) quality CFLs which is precisely what the hoi polloi will be buying.

• and also some bought in bulk from a wholesale internet supplier.

None of these issues influenced the results.

One does one's best. I also did a very expensive experiment with a pair of GE Genura lamps one of which lasted just over a year despite being sold under the pretext of lasting 18 years. At £19 a pop too much of an expensive risk to be considered.

Andrew Gabriel on here did say that they suffer from overheating if they are used in a shade in a cap up configuration. I think he's probably right. Since I presume their intended function is to provide useful light from a fitting mounted on the ceiling (they fail miserably in the Objet D'art stakes) having to have them pointed upwards is a bit of a right bastard.

In fact it is meaningless. Nowadays there are many and varied soft coloured lamps some quite deeply coloured almost like the lamps on fairground sideshows and roundabouts.

All perfectly repeatable.

Not according to the boxes I saw on display in Sainsburys today.

Not the case, lifetime is influenced by ambient temperaturemains voltage variations, and frequency and pattern of switching on / off

It would be more helpful if you were to tell us of the lamps you do approve of, which you believe we can rely on to honestly live up to their claims. It seems to me it would be a much shorter list than us putting up commonly available lamps and you shooting them down.

What evidence can you produce that verifies their claims. Which, in law are probably no more than "Salesmen's puffs"

I don't doubt you've had one, I've probably had one myself but gave up using it because of long warm up time, poor light output, humunguous weight, and it's generally obstroculous shape. But to claim that that would be the typical experience of people buying ordinary CFLs available today from an ordinary supermarket would be dishonest.

Indeed they are.

What was that you were saying about 20 years, and lifetimes " depend entirely on hours of use", (and therefore not switching cycles).

Maybe it hadn't been used. Because of long warm up time, poor light output, humunguous weight, and it's generally obstroculous shape

Ditto.

But yhey do.

And no attempt to gainsay it either.

Just take a walk around, say, IKEA. I did this week and saw that many of their CFLs on display were heading towards end of life status, and IIRC the buildiing was re-furbished just 1 year ago.

Are too ...

What are they then ? We've already dismissed the GE Genura.

Can we quit with the dissembling. Bugger the coolest and bugger the average, we are trying to get a grip on the problem cases.

How hot do the hottest get? They'll be the ones which have a compromised service life.

Ordinary use. In my case I'd admit to heavy ordinary use, because of the way the house is orientated, and has been designed and modified.

Certainly not, there's "Mad as a box of frogs" Hansen.

Taught to me at uni in 1965. We called them Schottky diodes BTW, or even Hot Carrier Diodes. I've only had to source 1 of them in my 28 year career in electronics. FWIR it was needed in a precision video camera used in digital cardiology, for it's precise characteristics.

Pray what is their relevance to fluorescent tube ballasts ?

If they can be used to improve the reliability of a 20 watt fluorescent tube ballast crammed into the end cap of a tube I'd be obliged if you could tell me how, then I could tell them at work - we have been building dimmable electronic ballasts since 1989 though we would never cram one into the tube end connector. Ours are the size of a pack of biscuits and we still get failures caused by overheating ...

continue

Not electrolytic capacitors and been around since 1957, no relevance at all to fluorescent lamp ballasts...

continue

A financial benefit for Chinese manufacturers. Relevance to the user of electronic fluorescent tube ballasts? ...

continue

The benefits gradually acheived thereof set against the reduction in reliability due to miniaturisation, cooling issues, price pressure, chinese sweat shop manufacturing etc. If the technology of making reliable subminiature electrolytics economically had been cracked several of the big name motherboard manufacturers would have availed themselves of it and saved a lot of damage to their reputations, and they are not even running at 339 volts DC. (= 240ac x 1.414)

Next !

What alternative can you propose that would ensure better performance and reliability whilst fitting into luminaires already in service. (Not the Genura) ?

2 for a quid is the current price, that being the case, what price do you expect the public will accept in future? They expect newer technology to bring prices down. IME By offering poxy products dirt cheap they've pissed in the soup. The customers will bitch about poor quality and reliability, on the one hand, and regard price rises as rip-off attempts on the other, and people flat-earthing about the current standard of performance and reliability just make it worse.

I't might be I'd never be happy to make such a claim. I aim to keep on bettering my understanding.

But I've 38 years experience in the manufacture and design of electronic ballasts (dimmable since 1989) and my understanding of the subject has been good enough to make me quite a lot of money as I've gainied the experience, and it keeps on coming.

;-)

DG

[David Hansen]

On Sun, 05 Aug 2007 21:18:05 +0100 someone who may be Derek Geldard wrote:-

A lot of words, but these fail to disguise the fact that he has either not performed the appropriate comparison, or has performed it but does not wish to state the results.

[Steve Firth]

I'm sure if you ask a responsible adult they willbe able to summarise it down to a level that you can understand.

"Oh look, CFL."

"See, see, the CFL."

"See the CFL glow."

"See the dim light."

"See the light fail."

"Oh watch, oh watch the light fail."

"Dim, dim, forever dim."

"See the dim light."

[Huge]

*applause*

I'm not buying any more of the damn things until they work 'as advertised'.

The 20W one in the porch failed last week, after less than a year in service. I've had to increase the wattage of the ones in the lobby and hall after complaints from "the management" about how dim they were. I don't have any complaints about the colour of the light, just how much of it there is and how long they last. They cannot possibly be "green" in any meaningful way given that they obviously cost more to make than a filament bulb, contain mercury and the lifetime claims are blatant lies.

[Steve Firth]

[CFL]

I put 4xCFLs into the ceiling downlighters in the living room they are R63 "equivalent" to replace 4x25W incandescent bulbs. Each downlighter is rated 11W and is supposed to be "Equivalent to 60W" this is pure unadulaterated hogwash.

They take about 5 minutes to get bright enough to see by but the state of the living room is now permanent twilight.

[The Natural Philosopher]

Indeed. And the energy they save is peanuts compared to what it takes to heat the average house.

[meow2222]

The mechanisms were known well before tungsten filament lamps were invented. They are precisely the same mechanisms that happen to all filament lamps (with the one exception of platinum lamps).

To be frank I dont have the time or inclination, you generate far too much misinformation.

NT

[meow2222]

Its weird that so many people have such a hard time putting in a suitable bulb (18w). They never have this difficulty with tungsten.

NT

[Steve Firth]

Go on, list the maker of an 18W R63 bulb.

[The Natural Philosopher]

Well more fool us for believeing the blurb.

Just replaced another CFL that failed..outside in the cool that one is.

I find that if you basiclly have their claims - i.,e. ==100W is really =60W and so on, and ==60W is really =25W, you get fairly close.

All the CFLS I have sen that failed are showing signs of overheating and are all sealed. O noted that te newer ones that haven' filed yet have breather holes in them. I suspect the stupid things just got too hot when turned on all night (waiting for the dog to find his way home)

I use em because they last a bit longer than the standard bulbs, but they are nothing like as good as they claim, and the energy saving is

(a) half of what they say for the same output (b) half of what's left, if the room needed heating anyway, with powerstations around half the efficiency of oil boilers..

So let's take that 11W/60W equiv lamp.

Actually you need about 18W to get 60W equivalent light..

So you are saving 42W. However given that about 75% of your light is needed in the winter when you also need heat, and the efficiency of your boiler is say 60%, you need to add back 53W of oil power to keep the room warm.

So in fact *your* total energy consumption has GONE UP. from 60W to 71W.

The only saving is the fact that the *oil-to-lamp* consumption is down..since to generate that 60W of electrical power would have taken around about 120W of fuel....more or less.

Now this is a net benefit to the environment, sure, but its nothing like the 82% claimed: In fact it's probably around 40% only.

On domestic lighting.

Compare and contrast that with te sort of savings you might make by NOT going to te supermarket once a day or twice a week, or by staying at home once a week..

[unopened]

Hmmm. Tungsten boils off the filament, deposits on the glass of the bulb.

Tungsten halogen invented to combat this by running a hotter glass envelope and using the halogen to react with the boiled-off tungsten and redeposit on the filament. Works well. The filament fails eventually as the tungsten redeposits on the cooler parts of the filament, which get thicker, while boiling off the thinner, hotter part of the filament faster.

CFLs dim for different reasons (obviously), but they do dim, and quite dramatically. See:

"CFL Lumen Maintenance" slide. A definition used in (I think) the US Energy Star programme is "At 40% of rated life, must be 80% of initial (100-hour) rating" I beleive it is better now, but not many years ago, nearly all CFLs failed that criterion.

Cheers,

Sid

[Matt]

Ditto, it's an old thread but I'll add my experiences

One CFL, an Osram Dulux 23W recently removed from service lasted just over 10 years, with 9-10 hours a day usage in winter reducing to 4 hours a day usage in summer. The date of installation was marked on the lamp housing.

I have also had in the past some cheap CFL's that have catastrophically failed (with a big bang) within seconds or minutes of being energised for the first time.

I also know of a 60W incandescent, horizontally mounted and subject to regular but intermittent operation in a bathroom. It's been in service for well over 10 years, maybe as many as 15. Until it actually fails and I check the date on the label inside I won't know for sure. When it fails I will also need to buy some new extra high stepladders, the ones I used previously to reach the light fitting went to the dump at least 10 years ago, hence why I know the lamp has been there so long.