Talk:Compact fluorescent lamp/Archive 1

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

The current picture is a little bland. Perhaps this one would be better? The only difference is that it doesn't show the bottom where the screw-in part is, but I don't think that's vital.PiccoloNamek 10:19, 26 November 2005 (UTC)

Actually, I think that there should be several photos, showing different styles of cf bulb. There are the straight kind, the spiral kind, and then there are various reflector and globe models. Nick 17:31, 26 November 2005 (UTC)

I think that's quite an attractive picure and if you haven't already made it the "lede" picture, I think you should do so.

Atlant 17:54, 30 November 2005 (UTC)

I also think that perhaps we should "gallerize" all of the different types of photos, instead of having them clutter of the side of the article. It would certainly look much nicer. I've gone ahead and done just that. If someone can find a better solution, please feel free to do so.PiccoloNamek 21:45, 30 November 2005 (UTC)

Also, someone kindly add why some of these bulbs only work in the base down position. Frustrating discovery that.

It's not generally true; I've operated several different brands of CFLs base-up. On the other hand, temperature control is important, both to regulate the

ballast

doesn't overheat, so it is entirely possible that certain CFLs in certain fixtures will not operate well. For example, many CFLs warn you against operating them in entirely-enclosed fixtures (as they will overheat).

Atlant 13:23, 16 January 2006 (UTC)

Someone please add why -some- of these will not operate with the base up. I have found bulbs that will only operate in a few fixtures and only base down. Some will operate in one base up location and not another. Is there a difference in the fixture itself that makes these bulbs incompatible? Or, is the the up-ness or down-ness of the base? Both variables applying here? Leave aside temperatrue issues for the moment as I am curious about why they won't even turn on when base up.

If they absolutely won't even "turn on" when base-up, then there's either a sensor within the lamp or a mechanical defect in the lamp. (I assume that the ambient conditions (temperature, humidity, line voltage, etc.) between your base-down and base-up tests are the same? You're not saying, for example, that "when the lamp is placed outside, base-up, in -20C weather, it won't turn on", right?

Atlant 14:56, 16 May 2006 (UTC)

I would like to see information about how mcuh extra energy is used in the manufacturing process of CFLs as compared to Incandescent bulbs. It seems to me that this could be a major factor in weighing up the environmental benefit of switching to CFLs and should be addressed on this page.

Does anyone have this sort of information?

Cheers Matt

I imagine it is not an issue. Both have glass bulbs. Both have screw bases. CFLs have a plastic cover for the ballast: that's low-energy to make. CFLs have a few electronic bits: they are too small to need much energy to make. Anthony717 06:54, 15 April 2006 (UTC)

Even so, this information is important, otherwise the calculations could be meaningless (particularly if the cost of manufacturing is subsidised and so not reflected in the cost of bulb purchase). —The preceding

unsigned comment was added by 86.129.160.253 (talk • contribs

) .

The general information that is currently circulating has suggested that some of the published figures are deficient in that they have not included all the energy costs associated with manufacture. The costs need to include all the energy used in manufacturing the bulbs, bearing in mind that there are three main parts made in separate factories (the glass bulb, the ballast unit and the plastic housing). The energy cost should, of course, include the indirect energy burden such as lighting and heating the factories. This is especially of significant as incandescent bulbs are made in a single factory. It would appear that the increase in total energy consumed during manufacture cannot be recovered during operation, especially given that CFLs rarely last anything even close to the claimed 8000 hours (actually achievable only under perfect conditions). But the environmentalists would rther that you didn't know this as they are more interested in pursuing a political agenda (i.e. forcing a change rather than actually achieving anything) 86.137.53.240 18:49, 27 February 2007 (UTC)

Hi 86.137.53.240. A couple of things about the way we work 'round here: First, try to add your comments to the end of ongoing discussions, rather than the middle. Second, things are only relevant to an encyclopedia if they follow from referenced, citable sources, so your comments above - for example about multiple factories, shortened practical lifespan, etc - may be true, and may not; may balance out the savings elsewhere, and may not. We can't tell unless you cite sources. Lastly, when someone wants to alter the tone of an article in the light of new evidence thay have found in reliable sources, then it's normal to discuss this here before going straight ahead as you did at 18:31. There's no fiendish environmentalists' conspiracy of silence going on here - we're just looking for the available published evidence and summarising it. --Nigelj 19:18, 27 February 2007 (UTC)

Nope, on indented discussion forums the practice is to add the discussion after the relevant point, further indented to separate it from the rest. Otherwise the context just gets lost.

Sadly, much of the information in the article is not from a cited source, or indeed citeable, because it's just plain false. Others have already highlighted the contentious points. 86.132.203.116 17:24, 2 March 2007 (UTC)

I, too, would like to see a thorough comparison of the energy input in manufacturing each. Also, don't fluorescent bulbs contain hazardous materials like mercury, and the phosphorescent coating isn't the most pleasant material to have around. And, what is in those ballasts, anyway? Even the disposal of fluorescent bulbs now seems to require a large energy input. 66.157.239.71 00:26, 25 July 2006 (UTC)

Old phosphors were hazardous but any phosphor made since the middle of the last century is pretty-much non-toxic. Old ballasts contained pitch and other tarry substances; high-power-factor ballasts often contained PCBs in their capacitors. Modern electronic ballasts are much more innocuous, containing nothing more hazardous than in any other electronic equipment including your cell phone; you can pretty well discount modern ballasts from your equations.

Mercury remains the most interesting toxic in fluorescent lamps, but as has been pointed out more than once, if we're comparing coal-generated electricity, the amount of mercury in a fluorescent lamp is substantially less than would be released by burning the coal to power incandescent lamps for equivalent light for an equivalent period of time.

Atlant 00:39, 25 July 2006 (UTC)

I agree that this is a non issue, let me explain why.

To begin, a CFL lasts 11 times as long as an Incandescent. So really, we have to compare the energy to manufacture 11 Incancescent light bulbs compared to 1 CFL.

Lets say you bought in bulk, and got 11 Incancescent light bulbs at $0.20 a piece for a grand total of $2.20. And then lets say you bought a CFL for $5.

Ok, so right there is a price difference of $2.80.

Now, lets now assume that the whole reason for that price difference right there, is entirely due to a difference in energy input cost. And If that energy came in the form of electricity, well then it required an extra

${\displaystyle \$2.80\times {\frac {1kW\ hour}{\$0.10}}=28\ kW\ hours\,}$

28 kW-hours to produce 1 CFL as compared to 11 Incandescent light bulbs.

Ok, but how much does using a CFL save in energy as compared to an incandescent? It saves:

${\displaystyle (60\ W-0.15\ W)\times (8000\ hours)\times {\frac {1\ kW}{1000\ W}}=360\ kW\ hours\,}$

The extra energy required to manufacture the CLF is at least an order of magnitude less than the energy it saves. JabberWok 01:37, 27 July 2006 (UTC)

A meaningful comparison would include all the direct and indirect monetary, environmental and energy burdens required to build the CFL and the incandescent bulb of same lumens, and to dispose of them. A Feit Electric 15 watt CFL, claimed to have the same brightness as a 60 watt incandescent and with a glass bulb around it weighs 74 grams compared to 27 grams for a GE soft white 60 watt incandescent. The extra material is made out of something and requires energy for its production. The CFL contains mercury, which requires special handling. Ifind the exposed spiral bulbs unpleasant to look at, so I have these more presentable bulbs in locations where the bulb is on much of the time and which are exposed (not inside a fixture). CFLs so far are not recommended for replacing incandescent bulbs inside fixtures(per the manufacturer's labelling). The packaging for a GE 75 watt equivalent CFL refers to www.lamprecycle.org, which says some states do not allow the CFLs to be disposed of with household refuse, and refers to hazardous waste collection drives. Usage patterns must also be considered when ooking at the lifetimes. A 1000 hour incandescent which is in the attic, a storeroom, the furnace room, or a closet might last for decades because of the low usage, so the economy claimed evaporates completely. The cost of new fixtures designed for CFLs evaporates the claimed savings for any but open air use. The poor performance or failure to start when it is cold (it hits minus 30 degrees F in parts of the US) render doubtful the use of CFLs outdoors and in outbuildings in cold climates. Only special CFLs can be dimmed at present. I have dimmers in bedrooms, TV room, living and dining rooms. New dimmers or special bulbs evaporate the savings there. CFLs are not recommended in applications where they are turned on and off frequently. These factors shoud be included in the article whereas at present it is very POV and get on energy saving bandwagon. Edison 02:01, 10 March 2007 (UTC)

So, someone says this is a "non issue", and then does some theoretical calculations... But, we want actual facts! Complete, real life-cycle analysis details about incandescents and fluorescents.-69.87.203.172 13:39, 2 April 2007 (UTC)

So what's stopping you? Get to it!

Atlant 14:00, 2 April 2007 (UTC)

not very in depth, but a good source:

http://www.greenhouse.gov.au/education/cfls.html#greenhouse

GREENHOUSE BENEFITS

QUESTION 1 - Does it take more energy to make CFLs than to make incandescent globes?

It takes a little more energy to make a CFL than to make an incandescent bulb, however CFLs use much less energy to run.

Over the life of the bulb, a CFL uses much less energy than an incandescent bulb.

David Woodward 05:16, 3 April 2007 (UTC)

Moved from Compact_fluorescent_lamp#Other_CFL_technologies

edit: CCFL's are not more efficient. Hot cathodes are more efficient... There is a reason why standard CFL's are hot cathode.

Well, less heat = typical CCFL is low powered - 3 to 5 watts. —Preceding unsigned comment added by 66.114.93.6 (talk • contribs) 16 May 2006

It's actually tough sometimes to determine which cathodes are really "cold" and which aren't. For example, "instant start" fluorescent lamps (or ordinary lamps operated as instant-start lamps with just one electrical connection per end) start as CCFLs but the impinging electron/ion beam rapidly heats a spot on the filaments to red-hot incandescence and they then operate as ordinary hot-cathode fluorescents. Even on older, large-scale CCFLs (the ones that look like "neon tubes"), there's often a hot spot that develops on the tubular electrodes.

I've not yet taken apart a modern LCD-backlightish "CCFL" to see what the electrode structure actually looks like, but I'd still be surprised to find they operate as truly cold cathodes; the ends of the lamps certainly get hot, suggesting there's a lot of heat developed by the "electrode losses".

Atlant 15:02, 16 May 2006 (UTC)

The term 'cold cathode' refers to the fact that the cathode cannot be heated by an applied current rather than anything to do with its temperature of operation. Whether the cathodes of hot cathode tubes are electrically heated depends on the method of ballasting the lamp. Generally most methods heat the cathodes during operation with the exception of 'switch start' systems where the cathode is heated only until a discharge is established. Thereafter it relies on the discharge current to keep it hot.

'Cold cathode' tubes have no means of preheating the cathodes, but nevertheless the cathodes run at similar temperatures to hot cathode tubes once the discharge is established. 20.133.0.14 15:16, 6 March 2007 (UTC)

I am not 100% sure, but the actual temperature of the cathode doesn't really determine hot/cold cathode. I think the key here is whether or not the cathodes are activly emitting and thus committing electrons to the arc stream or not. The idea is; "hot cathodes" actively contribute electrons to the arc stream, thus reducing voltage requirements from the ballast. So, typically cold cathode fluorescent lamps are very low current (and thus thin tubes) to keep the size and requirements of the ballast down. Given that, and if someone would produce a CCFL with the same light output and current rating as normal CFLs, the CCFL would probably (still) have lower efficiency, mainly because the ballast would be large and costly. So, the prospect of having large ballasts with multi-kilovolts at a few hundred milliamps keep the tube size (and current rating) down to ten milliamps or less. And as with all light bulbs, smaller = lower efficiency. —The preceding

Way back in 1908, a gentleman working at GE by the name of Steinmetz gave the correct formula to measure the power used by an AC load. It is still taught to all EEs today. http://www.invent.org/hall_of_fame/139.html Unbelievable pity that Wikipedia has such mistakes.

The formula for power used by a AC device is Power = V x I x Cos(theta); energy = power x time [kilowatt-hour] Cos(theta)of the V and I is the Power factor of your device. Restating - Cos(theta) is the cosine of the phase angle between the voltage and current flowing into the CFL. It has huge effect on the efficiency of any AC device.

CFLs operate on AC. But CFL ballasts needs DC to operate. So CFLs have bridge rectifiers [typically IN4001 diodes] at the ballast inputs. Bridge rectifiers cause a big phase difference between the voltage and current going into a CFL. This is the cause of the poor power factors in CFLs.

This is also known as supply line pollution. CFLs are major culprits in supply line pollution. As you may have noticed, large phase differences are bad for any AC device. They decrease the efficiency of your device. Hugely.

In a good CFL, the Cos(theta) factor is 0.6.In bad CFLs it is 0.5. This means that the efficiency calculation in this article is wrong. Pl get a EE to explain it to you. Then please correct the calculation. The power factor of a light bulb is 1. It is twice better than a CFL.

Utility companies in many countries are now beginning to take action against users who have reactive loads. a reactive load is any load with non unity power factor. Soon people using CFLs will be charged an extra levy for polluting the power lines. The Indian capital of New Delhi is in an advanced stage of this amendment. I know that the Sydeny Energy counci is also proceeding on similar lines.

And yes, I am an Hons EE graduate working for the 'Big Blue' computer company. I am the one who yesterday wrote the para that was rm'd as 'unreadable rant'. Problem is I design illumination in million square feet offices for a living, and I know what I am talking about, but since my input is going against the un-informed attitude of 'see how wonderful CFLs are!', it is being labeled as 'unreadable rant'. Typical for Wikipedia.

I noticed a uttery hopeless discussion on cold cathodes in this thread. It was the blind leading the blind. So happnes, that I am also the orignal writer of the Cold Cathode article in this Wikipedia. http://en.wikipedia.org/wiki/Cold_cathode. I wrote it in 2003. 216.84.11.250 was my IP address at that time. —The preceding

It's not non linearity (well very marginally). It's because the circuit draws its current in pulses. The current flows only when the instantaneous voltage exceeds the voltage of the reservoir capacitor. The 'stray' currents in the neutral conductor are not stray currents at all. Any discharge lamp generates significant quantities of third harmonic currents (due to their non linearity). In a three phase system, the third harmonic currents are all in phase (think about it). There is nowere else for these currents to flow but in the neutral (and they can be as large as the phase currents). This current is known as the 'triplun' current. 20.133.0.14 13:11, 1 March 2007 (UTC)

If there is a power-factor problem associated with CFLs (and we have no cited source that there may be), it will work to mean that the

apparent power, not more. Lastly the simple calculation P = I x V applies perfectly well in AC, provided all three are RMS values. Have a read of the Power factor article too. --Nigelj

18:48, 23 November 2006 (UTC)

Just tried it. A GE 240 volt 21 watt CFL reads as having a power factor of 0.56. 20.133.0.14 13:11, 1 March 2007 (UTC)

I am haranguing on this because my 9 yr old was on this page, writing something for his class, I looked over, saw your calculation, and I felt truly tormented.

The calculation is just so misleading. You have simplified things so much that they misrepresent reality. I am troubled by the thousands of kids like mine who are reading it, and getting incorrect information.

This article, in its current form, can be justifiable labeled as 'misleading', and 'ignorant'. CFLs are not as good.

And yes, ordinary RMS meters only give correct readings for perfect cosine waves. For non cosine loads like a CFL, readings from RMS meters are completely useless. So watch out for RMS. [btw: this is the key problem with all the 'over unity' folks. They dont know that RMS meters read correct RMS only if the energy is following a cosine law, so they get all these over unity readings]

First, the wave is known as a sine wave. Nobody who even slightly understands these things would have made such an elementary error. Second, you obviously have even less knowledge of meters. While cheap multi-meters are of the type known as 'mean sensing RMS indicating' for which your observation is valid, better quality meters indicate the true RMS regardless of the applied waveform. 20.133.0.14 13:11, 1 March 2007 (UTC)

Well, "true RMS" to the degree that the waveform is within the

current probe

, and the 'scope's analysis/integrating software.

Atlant 13:28, 1 March 2007 (UTC)

It rather depends on the means that the instrument uses to calculate the true RMS (for that is what they do). Your $30 Radio Shack job will have a pretty limited frequency response (a few hundred Hertz tops). Many older true RMS meters were only accurate over a limited range of crest factors, but the modern ones have overcome that limitation. The one I use at home has a bandwidth of 100 Mhz and claims to read true RMS regardless of input waveform (It's certainly dead on for a square wave, which is usually outside the scope of the older instruments). Since it is a 'scope as well, the input waveform being measured can be seen as well. Cost well over a grand though. However, you don't need esoteric bandwidths to measure the true RMS current in a CFL. I have a $30 meter that measures RMS volts, RMS current, Watts, VA, Hertz, power factor and even kilowatt-hours. It reads within 0.2% (0.5% for power and VA) - it does as well. The thing is that it claims to read within 30 ppm for kwh, but I don't believe that. I think they meant the time element. 86.132.203.116 17:12, 2 March 2007 (UTC)

On reactive loads- any load that causes a phase diff between V and I is reactive. Diodes in AC circuits are reactive. In pf correctin work, we routinely build 'simulated inductors' using op-amps and hi voltage FETs. It not a reactive load according to you, but it is. Althought it has no reactive components. Device non linearity causes harmonics, not phase difference. Pl mind the difference.

You are showing your ignorance here. Although phase difference between the current and voltage gives rise to a power factor of less than unity, it is not the only phenomenon. Switch mode converter circuits (and electronic ballasted CFLs use a resonant switch mode converter) also have a power factor that is not unity - but the current and voltage are in phase. In fact, if you check the current waveform, you will discover that it is not a sine wave, but it in fact flows in pulses (whenever the instantaneous voltage exceeds the reservoir capacitor's charge). The RMS value of the voltage is calculated over the whole sine wave, but the power is only drawn over that part where the current flows. This gives rise to a VA (RMS Volts x RMS Amps) that is larger than the actual power. Typical power factor for a CFL is 0.5 to 0.6. 20.133.0.14 13:11, 1 March 2007 (UTC)

Now can you pl fix the wrong calculations ?

Things are only half as good as you are portraying them to be.

CFL pf = 0.5 to 0.6

ashvini1988.

Diodes aren't "reactive". And as the circuitry in CFLs becomes more and more sophisticated and utility companies become more demanding, don't be surprised to find power-factor correction included (although I don't think it is yet). Regardless, power factor probably isn't a factor here; I think you'll find that the stated power consumption of CFLs already is stated in terms of Watts (so real power), not VA or anything that needs further adjustment.

Atlant 13:18, 24 November 2006 (UTC)

I agree, Atlant. I don't think we can do anything with the ideas above that claim to be beyond all documentary evidence and outside of any scientific explanation. --Nigelj 20:45, 24 November 2006 (UTC)

Please try and comprehend that the CFL with '5W' on the box actually draws 10 watts, because its power factor is 0.5 (5/0.5 =10). This is not a matter of 'opinion', engineering never was polls based. You stick a meter in there and learn it. Please try understand that one will pay for 10watt-hour of energy per hour of a '5W' CFL. Not 5watt-hour, as the print on the cover would have you believe. The cover of a CFL doesn't tell you that it draws 5watts @ 0.5 power factor. Its the responsibility of people like us, with wiki's, to draw the attention of the world to this dirty secret.

What utter tosh. If the box says 5 watts, that that is what it takes. Of course if you measure the volts and the amps and multiply them together, you get around 10. But that's 10 VA not 10 watts. Once you multiply the ~10 VA by the power factor of around 0.5, you get the true watts of 5. And this is what you pay for, becaus your electric meter really does measure watt multiplied by time. 20.133.0.14 13:11, 1 March 2007 (UTC)

But I am sure you wont change your wrong methods because you are not educated enough to comprehend the true complexity of the CFL lie.

It is such a shame that the grand experiment of 'wisdom of the masses' has turned out into a 'squabble between people with various vested interests'.

It is now a repeating thread on most wiki articles. People with high school knowledge dominate the discussions, using the crutch of 'lack of citations' and 'not relevant' heads to keep articles in their wrong shape and misleading form.

What is needed is experts, not masses. I saw wiki as a liberator from the yoke of Encyclopedia Britanica. Not so. Wiki has evolved into a great squabbling ground between people with the 'know how' and 'people with too much time'. One must be weary of the Wiki.

Once again- the calculation of the relative comparison of CFL and Incandescent Lamps in this page is wrong. It is off by 200%. This article makes the CFLs look twice as good as they really are. That is because this article is ignoring the fact that the power factor of a CFL is a poor 0.5 [which is only half as good as an incandescent lamp with a pf=1.0]. This means that your actual 'on the books' energy savings will be 200% lesser than what this article makes you believe. You will only save half as much money in your monthly utility bill as this article makes you believe. Beware of the misleading 'efficiency' of a CFL. Take power factor into account. A CFL is not as efficient as it says on its cover. Thanks . ashvini1988.

Actually, it is exactly what it says on the box. You are showing your ignorance of power calculations in AC circuits. 20.133.0.14 13:11, 1 March 2007 (UTC)

I added the following paragraph to this section: "However, virtually all of the energy from light bulbs of any type is converted into heat. During cold months when a building is being heated, the heat produced by a light bulb is helping to heat the building, and this needs to be taken into account when calculating the energy cost of different types of bulbs. In a typical home using electric heat, for example, light bulbs of any type do not produce any electricity cost during cold months, since the heat produced by the bulbs simply offsets heat that would have been produced by the home's heating system".

It is inaccurate to look at the cost of different types of bulbs without taking into consideration the above. An inefficient incandescent bulb may turn out to be 100% efficient if the "waste heat" from the bulb is used to heat the building the bulb is in. --Xyzzyplugh 13:15, 24 June 2006 (UTC)

What you are proposing is a theory (is there references for this?), which is why I removed the paragraph for the moment. But at least it's a testable theory. Let me put that paragraph into different words: "Using CFLs instead of Incandescent bulbs results in a utility bill that is no smaller, because any savings in electricity have to be made up for in heating costs." That's what another way of saying it, correct? I'm stating it this way because now it's more obvious how it can be tested.

So now the question is: taking into account both heating and electricity costs, does using CFLs save money overall? I haven't seen any references personally on this, so I can't comment. Anyone else? JabberWok 16:54, 24 June 2006 (UTC)

I highly doubt that a source will be found on the final sentence. Everything I've ever read on the comparison between fluorescent and incandescent bulbs totally ignores the fact that the waste heat generated by bulbs might actually be useful. If you want to remove the final sentence from the paragraph entirely due to it being unsourced, that would be understandable, although I find it to be a simple logical conclusion. The first part of the paragraph is simply self-evident. Light bulbs produce heat, homes need to be heated, therefore the bulbs are helping to heat the home. Sources could easily be found demonstrating these things, but I don't think that would be necessary. --Xyzzyplugh 17:09, 24 June 2006 (UTC)

Actually, it just occurred to me why the paragraph you're proposing is incorrect. And it comes down to one thing:

• Energy from electricity costs more than energy from Natural gas. (And the reason for that is because you have to burn things - like natural gas - to make electricity.)

So because of that, heating your home with a space heater (which is what an incandescent light bulb essentially is) costs more than heating your home with natural gas.

That's why, for any home heated by natural gas, switching over to CFLs results in a overall lower cost.JabberWok 17:12, 24 June 2006 (UTC)

This doesn't contradict anything that I said. If you notice, I specified "In a typical home using electric heat" in the last sentence of my paragraph above. What you're pointing out is true, of course. The usual method which is suggested to compare CFL's and incandescents is a gross simplification. It doesn't take into consideration whether the home needs to be heated, or the method of heat, and it also doesn't take into consideration whether the home is being air conditioned. If you're air conditioning a home, this doubles the inefficiency of incandescent bulbs versus fluorescents since you are paying for the extra electricity used to produce all that waste heat that incandescents produce, then you're paying to remove that waste heat with your air conditioning system. --Xyzzyplugh 17:32, 24 June 2006 (UTC)

Just so I can get a rough order of magnitude...

Lets say, during a winter month, a home uses something like 50 Therms of natural gas. In kW-hours that's:

${\displaystyle (50\ therms)\times {\frac {105,000,000\ J}{1\ therm}}\times {\frac {1\ kW*hour}{3,600,000J}}=1465\ kW*hours}$

Ok, so it takes about 1,500 kW-hours to heat a home home for a month.

To compare with energy cost on the main page, using a 60 Watt light bulb for 8000 hours (which would take several years) would create about 480 kw*hours of heat.

So the heat released from one bulb is a small effect, at least.

But now, is the heat from a light bulb useful heat? People turn up or turn down their thermostats depending on air temperature. So does the heat from a lightbulb mostly end up in the air? Or does it end up in a nearby wall or ceiling where it doesn't contribute to a person's general feeling of warmth?JabberWok 18:22, 24 June 2006 (UTC)

Your statement that light bulbs produce almost no heat is an odd one, if you think about it. If bulbs aren't producing a significant amount of heat, then they must not be using a significant amount of energy, and so why not just use incandescent bulbs?

I have electric baseboard heat in my home, all of my heat comes from electric heating units along the walls near the floor (and from light bulbs and any other electric appliances which might be running). As the heat from the baseboards manages to circulate its way through the entire home, including warming hallways and bathrooms which don't have any baseboard heaters at all, clearly electrically produced heat doesn't just sit in one spot. The bathrooms are not noticeably cooler than the rooms which have baseboard heaters in them, either. The baseboard heaters don't have fans or any such thing, they just get hot and the heat makes its way out of them. It may well be that ceiling light fixtures are less efficient as more of the heat may be absorbed into the ceiling. However, many light bulbs will be placed in lamps or chandeliers where they are not directly next to the ceiling.

The simplest demonstration that the heat from a light bulb is not all being absorbed into the ceiling is to just feel the ceiling nearby. An incandescent light bulb gets so hot it burns you if you touch it, while the ceiling a few inches away doesn't feel hot at all. Furthermore, if the heat were all sitting in one spot and not circulating into the room, the ceiling would quickly catch fire.

We don't need to be doing original research to edit this article though, not even simple research like feeling our ceilings. But the fact that heat from a light bulb heats the air in a room is self-evident. The very existence of electric baseboard heating, which contains no fans, ought to be evidence of this. --Xyzzyplugh 04:55, 25 June 2006 (UTC)

It surely heats the room. However the effiency of incandescent bulbs in heating a room (in comparison to an eletric heater which of course depends on the type of eletric heater anyway) is debatable and uncertain. Remember Without any reliable research, we should only mention that it may reduce heating costs since the effect could easily be negligible. (BTW, you might want to remember the 3 ways of heat transfer; conduction, convection, radiation and the relevance. Also remember that hot air travels up. It's easily possible incadescent bulbs as 'heaters' may increase the heat gradient in the room i.e. there will be a great difference between the temperature close to the ceiling and the floor and so you may find they don't actually have a great effect in reducing heating costs). Nil Einne 22:18, 4 August 2006 (UTC)

Here is an article talking about the same thing this guy is. Apparently there exists "High-Power Factor CFLs" to eleminate the problem... http://64.233.167.104/search?q=cache:bnqqTcWLnPYJ:hem.dis.anl.gov/eehem/93/931113.html+power+factor+CFL&hl=en&ct=clnk&cd=1&gl=us 75.132.24.225 02:01, 27 February 2007 (UTC)

Some comments from a typical non-EE user of CFLs. One, we measured the wattage of a standard 60W incandescent bulb with a "killer watt" phantom load device and it drew 63 watts. Then we plugged in a 11W CFL and it drew 10 watts. So the argument about power factor is just plain wrong - CFLs draw what they say they draw. Or at least our name brand bulb did. And second, here in Texas we do not want hot bulbs in our houses - we start running the AC in May if we're lucky, earlier if not, and keep it on until November or so. Hot incandescent bulbs are terrible. 12.110.209.151 20:43, 29 March 2007 (UTC)Britt, 3/29/07

Xyzzyplugh added and JabberWok reverted-out the following text:

However, virtually all of the energy from light bulbs of any type is converted into heat. During cold months when a building is being heated, the heat produced by a light bulb is helping to heat the building, and this needs to be taken into account when calculating the energy cost of different types of bulbs. In a typical home using electric heat, for example, light bulbs of any type do not produce any electricity cost during cold months, since the heat produced by the bulbs simply offsets heat that would have been produced by the home's heating system.

There's some truth to this but it's still more complicated than that. One, electrically heated houses are pretty rare, and compared to any other fuel source, even an electric heat pump, the heat produced by any lamp is costly pure-electric heat. Two: It's not entirely clear that all of the heat produced by all household lamps is captured as useful heat into the envelope of the house. In particular, certain ceiling-mounted lights lose heat into the ceiling and roof spaces that would ordinarily not be heated.

Still, in northern climates where the need for heating predominates over the need for cooling there's quite a lot of sense to this argument that you need to at least consider the "electric heat" provided by lighting; it's certainly not the 100% loss that a simple calculation might suggest.

Atlant 22:57, 24 June 2006 (UTC)

Note my answer above. Also, one thing which many people seem to be missing is the issue of luminous efficacy is complicant. The difference is about 2-3x it appears depending on type of CCFL and incandescent bulb. Some of this is obviously going to heat however some of it to infrared. While the infrared is obviously going to have some effect on heating in the room I would assume (it's been a while since I did physics), most of it will effect on the objects in the room, the floor, ceiling etc. However this will affect the heating in the room I don't know. Or in conclusion it's clearly really complicated as you say and we really have no idea how much of the heat is usefully captured into the envelope of the house... Nil Einne 22:43, 4 August 2006 (UTC)

In some cold countries (like Sweden) direct electric heating, even without any sort of forced air, is quite common. There's a general underlying theme in the above argument that electricity is a "higher form" of energy than, say, oil or gas, because it is easier to convert and it takes a lot more heat energy from oil, gas, or coal to make a given amount (joules) of electric energy. This is also a bit misleading in some countries, where much of the electricity comes from hydroelectric or nuclear sources (Sweden, Norway, Canada)---electricity is relatively less scarce in those countries and using it for direct heating isn't as stupid as it sounds from an American perspective.

Another effect that really, really should be considered is that it is dark at night, when it is cold (duh!!!!); it is dark in the winter, when it is cold (double-duh!!!). Even in a relatively warm climate, it is quite possible that incandescent lighting adds very little to the a/c bill for this reason alone. And that the contribution to heating is bigger than a simple averaging calculation would yield. Also please remember that the atmospherically vented gas furnaces common in the United States are not 100% efficient, either. I'm not sure how efficient they are, but I'd be surprised if it were much over 80%. To get higher efficiencies, "condensing" furnaces are necessary---furnaces that allow the water vapor generated (together with CO2) by burning the fuel to condense and thereby release its enthalphy of vaporization rather than carrying it up the chimney. MikaNystrom 02:26, 4 March 2007 (UTC)

I'm putting the following in place: "The above calculations are a simplification, as virtually all of the energy from light bulbs of any type is converted into heat. During cold months when a building is being heated, the heat produced by a light bulb is helping to heat the building, and this needs to be taken into account when calculating the energy cost of different types of bulbs. In addition, during hot times of the year when a building is being air conditioned, CFL bulbs produce extra savings over incandescent bulbs because extra energy is being used by the air conditioning system to remove the waste heat which the bulbs are producing".

I removed the previous final sentence with which there was disagreement. Anyone still have problems with this new paragraph? It may not be written as well as it could be. I'd like to see some mention of the fact that the usual calculations given are a simplification, but going into too much detail would probably be tedious and unnecessary.--Xyzzyplugh 19:13, 30 June 2006 (UTC)

A little problem with "The energy that is not used to create light is instead converted into heat energy." While technically true, it's a bit deceiving because that light will also eventually turn into heat energy. Visible light produced from a light source (cfd, incandescent, etc.) bounces around the room until it is ABSORBED by a material (thereby increasing the material's internal energy). In other words, ALL of the energy emitted from a light source will (eventually) heat the room. (except, of course, for the fraction of radiation that escapes through a window). e.g. Even a perfect, 10-Watt light bulb that emitts only visible light would still be a 10-Watt heat gain on the room. Chrike 00:25, 21 February 2007 (UTC)

  yup.  Does one have to be a theoretical physicist to understand this?   Wuggsy 18:56, 3 February 2007 (UTC)

Except the part that goes out through the window! MikaNystrom 02:16, 4 March 2007 (UTC)

Turns out some companies now offer CFLs that can be dimmed: http://www.nolico.com/saveenergy/dimmable_lamps.htm Kar98 17:33, 9 July 2006 (UTC)

I trimmed down the last paragraph about heat effects to a more manageable 60 or so words.

Although it is too anecdotal to put in the article, I read that Japanese residents (especially in small apartments that don't have central heating) often use incandescent bulbs in the winter and CFLs in the summer.

Radiant heat sources are comparatively wasteful because they lose more heat through floors, walls and ceilings. This is a seriously important problem. Fan-forced induction heaters (including electric central heating systems) are substaintially more efficient than space heaters with exposed coils or exposed incandescent bulbs for increasing indoor air temperature. Anthony717 00:07, 10 July 2006 (UTC)

Does the spectrum of fluorescents limit their use in grow lights for houseplants or greenhouses (or whatever)? Photosynthesis only works with certain wavelengths, right? — Omegatron 00:33, 18 July 2006 (UTC)

If we're talking small-scale, there shouldn't be any problem. Plants grow well-enough under a wide variety of light colors and certainly grow under good-old "cool white" phosphors.

If we're talking large scale (say greenhouse scale), then you'd avoid CFLs because other light sources are notably more efficient; T8 and T5 fluorescents seem like a good step up from CFLs but

metal halide lamps

seem to be popular among certain sub-cultures. ;-)

Atlant 12:50, 18 July 2006 (UTC)

That's not why I was asking, but I'll keep it in mind in case I ever decide to take up a new hobby.  :-)

I was leaving a lamp on to compensate for low light levels and a sickly plant, and a friend said it was a waste of power because they don't grow in light from fluorescents. Just curious if they were practical for that. — Omegatron 13:47, 18 July 2006 (UTC)

It appears the 3.5 million car figure comes straight from the cited "Make the Switch" campaign. This is at least somewhat in the neighborhood of the one bulb per household being equivilent to one million cars off the road claim made on the Energy Star page. —The preceding

Well, I have been playing with a number of different CFLs from GE I got at Walmart. All of them emit some type of buzz. I even recorded the buzz as WAV file. (If somebody wants me to upload them and link to hear I will). It is incredibly hard not to notice it, especially when your job has everything to do with perceiving various types of noises. Anatoly larkin 21:49, 26 January 2007 (UTC)

The chart, as currently displayed, substantially under-reports the typical light output (in lumens) of incandescent bulbs of various wattages. For instance, a 100w conventional bulb is said to generate 1200 lumens of light output. However, many 100w conventional bulbs put out something on the order of 1500-1600 lumens of light -- prehaps 33% higher than the original figure. Similarly, 60w bulbs are represented as generating 600-700 lumens, while they are often actually rated at 750-850 lumens -- again, possibly 33% higher than the original figure. This shading of the data as to disfavor conventional bulbs is evidence of either dishonesty or negligence.

I refer the reader to the Sylvania lighting products company webpage for further information on the light output of incandescent bulbs, as an example: http://www.sylvania.com/ConsumerProducts/LightingForHome/Products/GeneralPurpose/. —The preceding

To fit enough arc tube in the volume that was previously occupied by an "A-lamp" incandescent light bulb with as few sharp bends as possible.

Note that lots of the traditional CFLs (shown in the one image in the article) just use a series of conjoined straight tubes.

Atlant 00:40, 1 February 2007 (UTC)

Firstly, the discharge in a medium pressure fluorescent lamp (which is what a CFL is) is not an arc, it is a discharge. You need to go to the high pressure lamp for an arc. It doesn't in fact matter whether the discharge tube is a spiral or a series of joined straight tubes, the discharge will happily follow whatever path is laid out for it. The spiral design appears to be an attempt at making the lamp a similar length to its equivalent incandescent bulb. Early CFLs consisted of 4 tubes joined in a square formation and (in the the 20 watt version) was nearly twice as long as a its equivalent 100 watt incandescent buulb.

Later the design evolved into 3 looped tubes side by side making the whole lamp about 25% shorter. The spiral design has shortened the lamp further still. It is a constant sourse of irritation that when a CFL fails, that you cannot seem to buy one of the same design.

What I don't follow is the claim that the phosphor is thicker on the lower side of the twist. The ability to evenly coat glass tubes has existed since the dawn of fluorescent lamps. 86.132.203.116 16:54, 2 March 2007 (UTC)

It is a discharge

Well, you get to chose between two types:

arc discharge

. It's an arc discharge.

And one reason you want an arc discharge tube with as few sharp bends as possible is that that makes it easier to coat the tube uniformly with the phosphor. The older PL shape was easier still, but as you observe, rather too long to replace "A" lamps.

Atlant 00:25, 14 March 2007 (UTC)

As with many challenges in engineering, there are competing factors. When designing a CFL, you are aiming for the best compromise across them all. There's a fixed power wasted at each end (electrode) proportional to the tube current. In a short tube, this becomes a significant part of the power used, so you want the tube as long as possible. For something which is designed to replace a standard filament lamp, this means the tube will have to be rather bent/folded/sprialled/etc. If you consider a simple case of a straight tube folded in the middle to create two parallel tubes, some of the light emitted from each side passes into the other side. Whilst some of this will pass through, much is lost by the time it's gone through another couple of layers of glass and phosphor, and possibly been reflected a number of times at the related optical surfaces. Thus folding the tube in half has reduced its efficiency. However, merely folding a tube in half still doesn't get you a long enough tube which will fit within the maximum acceptable dimensions of the lamp. So various techniques are tried to squeeze as much length of tube into the space as possible, whilst minimising the extent that any part of the tube obscures any other part of the tube. One of the best tube layouts for this turns out to be a loose spiral (significantly more open than the one shown in the picture), and I think one of the manufacturers had a patent on this tube layout.

The "phosphor is thicker on the lower side of the twist" probably means the phosphor is thicker on the inside of the spiral. This may or may not be true, and may or may not have much affect on efficiency. The phosphor coating can be applied either before or after bending the glass. It's cheaper to apply it beforehand to a staight piece of tubing, but heating the glass to bend it does have a detrimental affect on the phosphor which becomes slightly less efficient. If it was applied before bending, it will be thicker on the inside, whereas if it was applied afterwards, it probably won't be. Phosphor coatings are not transparent, so too thick a coating doesn't let much light out, whereas too thin a coating won't catch all the UV and convert it to visible light—getting that balance just right is a chalenge. (This is another one of those competing factors.) However, a thicker layer of phosphor on the inside of the spiral would cause more of the light it generates to be directed back into the tube and out the other side, rather than into the centre of the lamp where it's likely to be obscured by other parts of the tube. Yet again, another factor which might give some gains. Look up details of Aperture fluorescent tubes—this is partly how they work to generate directed light beams. Creating an aperture spiral tube with the aperture along the outside edge would likely create a very efficient lamp, but it would probably be prohibitively expensive (or someone would have done it already).

Andrew Gabriel, 81.187.162.107 09:19, 31 March 2007 (UTC)

wondering if anyone knows about the embodied energy, and the material cost (cost of production), and also toxicity of flouros verses incandescents. I know that they use mercury, which is pretty toxic, and I would assume that the twists of glass tubing would have a higher embodied energy than an incandescent, and I would also think that there is generally more material used (heavier, thicker glass, etc.). it would be good to get some of these facts in there if anyone knows. --naught101 02:46, 2 February 2007 (UTC)

When you do the mercury trade-off, though, you have to reconcile the fact that coal-fired power plants emit a lot of mercury. So over the life of the CFL, it's probably a net win on mercury emissions, especially of the lamp is properly recycled. The other points are very interesting questions, though.

Atlant 13:46, 2 February 2007 (UTC)

Brr, it is cold outside. Today is February 3rd, 2007. Smack dab in the middle of winter.

Congradulations with your energy efficient lightbulb purchase, you have now saved 100 watts.

Hold on a second, your home heater has something to say.

"Dang, I must now put out 100 more watts," home heater. "You fool," says your home heater, "You have saved nothing, I must still maintain that 68.5 degree temperature your thermostat is set at, and when you replaced that inneficient lightbulb you removed 100 watts from this enclosed system which is YOUR HOUSE!"

"Go back to school," says your home heater, "and learn this concept called the Conservation Of Energy."

Wuggsy 18:52, 3 February 2007 (UTC)

"But wait," says mister gas-meter; "I supply you - home heater - with gas, not electricity. Electricity is a lot more expensive per unit of energy than gas, so he's now heating at a much lower cost."

"Indeed," adds the air conditioning unit, "and think about in summer when you are using me - I'll have a reduced need to get rid of that excess heat."

"Besides", comments the lightbulb, speaking to himself as much as anyone else, "if he were really worried about saving energy, then he could turn down the thermostat and put on a sweater."

Neo 01:56, 4 February 2007 (UTC)

Plus incandescent light bulbs are (in the UK at least) often in ceiling fittings, this they keep the ceiling (and the upstairs neighbours if you live in an apartment) warm, not the room. —The preceding

talk • contribs

) 21:56, 4 March 2007 (UTC).

From the point of view of the environment, does it matter if you are heating your own place or your neighbor's? MikaNystrom 09:37, 5 March 2007 (UTC)

"But wait," Sayas the Home Heater, "I am a heat pump and run on electricity. So, I am using more energy. But that's provided by nuclear power, so I expend less carbon burning and cheaper energy. But that has a half life of thousends of years and makes people sick. Oh goody, I'm now switched to hydro-power. Oh no, those poor fish are slaughtered. Ah, now I am running on ethanol. Too bad the humans can afford a $20 bowl of corn flakes. What? Out of corn? Well, you may be starving the world, but at least you're warm." The Light Bulb looks down and adds, "And they can see in the dark too."

Basically this argument is stupid and childish and can go on forever. So drop it. 70.108.121.242 22:16, 7 March 2007 (UTC)

In the section "Efforts to Encourage Adoption" the first sentence of the second paragraph reads: "Some governments have attempted to encourage CFL usage by distributing them for free and by appealing to people's moral beliefs." However, presently there is no citation for this information. This article ([1]) mentions Hugo Chavez' government initiative to distribute free "incandescent bulbs" to neighborhoods across Venezuela. If these "incandescent bulbs" are the same as CFLs then this source could be cited for some of the information in the sentence ("Some governments have attempted to encourage CFL usage by distributing them for free..."). I'm just not positive if the incandescent bulbs mentioned are the same as the CFLs this article is about. The same goes for Castro's initiative, also mentioned in the news piece, where he "launched a ...program two years ago, sending youth brigades into homes and switching out regular bulbs for energy-saving ones to help battle electrical blackouts around the island." Does anyone know if these initiatives used CFLs or some other kind of energy saving light bulb? 2headedboy 21:39, 21 February 2007 (UTC)2headedboy

The appropriate use of CFLs can undoubtedly reduce the end-user’s electricity bills. However I understand that a mix of conventional and CFL bulbs is recommended by responsible authorities.

In my experience there are numerous problems with energy saving bulbs:

1. Their claimed light output is misleading. It is compared with "soft colour" (low-efficiency) bulbs not normally used for illumination. This could be why CFLs seem dimmer than the bulbs they are bought to replace.
2. Using a photographic light meter, I have compared the light output of a CFL and various pearl bulbs.
3. The output of a one-year-old 20 Watt CFL (claimed equivalent 100Watt) rated at 230V-240V was JUST LESS than that of a normal 60W 2000hr 240V pearl bulb, and around 40% of a 100W 240V bulb.
4. CFLs start off dim and the light meter shows that they take more than five minutes to reach full output, so where they are only switched on briefly, they are less suitable than conventional bulbs.
5. Their brightness is temperature sensitive, so they may be unsuitable for outdoor use in cold weather.
6. They sometimes do not last anything like the claimed hours.
7. They are more fragile, and are more often broken on arrival.
8. They are complex and are more often dead on arrival, or suffer infant mortality.

It has also been said that

1. Their brightness reduces over their life.
2. They are not compatible with most existing dimmers, though compatible dimmers are available.
3. They consume more energy to manufacture, polluting the country of manufacture if clean energy is not used.
4. Their manufacture would be uneconomic if not subsidised.
5. Their use is promoted by the Government backed Energy Saving Trust.
6. The technology is not yet mature.
7. It could be that we should be using another type of low energy light source such as LEDs.

I hear the clatter of deckchairs being frantically rearranged on the ecology bandwagon. The Australians will discover this to their cost, if conventional bulbs are phased out in the short term as planned. GilesW 18:41, 25 February 2007 (UTC)

Probably the worst disadvantage with CFLs is that they don't like being on for short periods of time. I think GE recommends theirs are left on at least 15 minutes per go. I have lots of lights that are on for less than that when they are on---I use incandescents in those, and will continue doing so, regulators be damned. Switching to CFLs would mean I'd have to leave those on, or suffer less lighting and shorter bulb life. What uses more power, a CFL on half the day or an incandescent on for two minutes? MikaNystrom 09:42, 5 March 2007 (UTC) CONCLUSION: Assuming the "100W" CFL bulb's power consumption is 20W as claimed by the maker, and that its light output part way through the CFL bulb's life is equivalent to a 60W 2000h pearl bulb (at 230V), NOT a 100W 1000h bulb as labelled, then the energy ratio of (CFL / equivalent pearl) is 1/3, not 1/5 as indicated by the maker (a 66% over claim), and not 1/4 as indicated in the Wikipedia article page. The differences are significant. I request that someone repeats the simple experiment, using a variac etc to supply 230V & 240V, and compare the light outputs of CFL & pearl bulbs at each voltage.

Independent data about actual bulb life is also needed. The 12000 hour & similar claims for CFLs do not seem to me to be realistic, though others support the claim.

If my findings are substantiated, the calculations in the Wikipedia article need to be revised relative to both CFL efficiency and actual bulb life. I have raised these Misleading Claims with the ASA. The data need to be understood so that costs and benefits can be properly evaluated. Otherwise Government policy will again be based on misleading claims by interested parties. GilesW 22:37, 25 February 2007 (UTC)

If the real problem is that the output of a 20W 230V-240V CFL is equivalent to that of a 240V 100W pearl bulb as claimed at 240V, but when supplied with 230V is only equivalent to a 60W 240V pearl bulb, then locations with 230V supplies would need CFLs rated at 230V and not 230V - 240V.GilesW 17:33, 26 February 2007 (UTC)

I reformatted your post to make your points stand out a bit more - I hope you don't mind. A couple of days ago, someone edited the article to include many of the disadvantages of CFLs. About half were established facts. The remainder, though having an obvious basis in fact were not necessarily directly citeable, at least quanitavely. It was interesting to note, that some obvious environmentalist completely removed any and all references to any possible disadvantage, restoring much that was inaccurate (see below for discussion on those points!). Unfortunately such people will not allow anyone to critisise their holy grail. 20.133.0.14 12:43, 1 March 2007 (UTC)

Please look for reliable sources which satisfy the nrew

WP:ATT to examine the light output, safety, compatibility, and environental variables related to CFL vs incandescent. I note that many present incandescent bulbs are in enclosed fixtures which CFL manufacturers warn against, or are on dimmer circuits. Do the CFLs catch fire, fail prematurely, or just not light in enclosed fixtures or on dimmers? Or are people supposed to throw away the glass enclosure and have ugly exposed spiral CFLs in place of former enclosed fixtures with incandescents? Or must they have an electrician replace the light fixtures or do do-it yourself rewiring? I have some antique incandescents over 100 years old which still work when I try them, so I suggest stocking up on incandescent bulbs for applications with dimmers or where the bulbs are in enclosed fixtures or where they are used only occasionally or when they are outdoors. That would be about 3/4 of the light bulbs at present. Can anyone find published sources which address these concerns to include in the article? Most of the sources so far seem to be the manufacturers of CFLs or POV environmentalist groups or those with a like agenda. There need to be links to and discussion of the enclosed fixtures which can be used with CFLs and the special dimmers and CFLs to be used with them. Edison

15:44, 12 March 2007 (UTC)

I removed (not blanked) a lot from the "CFL energy consumption compared to incandescent bulbs". The only information it was trying to convey there was that CFLs use a certain amount less. You don't need flashy pictures or dollar values to say this. Just say it uses 25%, and move on. You've made the point. Dollars don't mean anything to those outside the US, while percent electricity cost saved, does. Wikipedia is not an advertisement, and it doesn't belabor simple arithmetic. MrVoluntarist 20:40, 27 February 2007 (UTC)

{{Multiple issues|disputed=March 2008|POV=March 2008}}

The article has been edited to only include the advantages of Compact Flourescent Lamps (CFL). The environmental advantages are nowhere near as great as is claimed once the disadvantages are taken into account. By removing the disadvantages, the article has a clear bias. As an encyclopeadic article, it should cover both the advantages and the disadvantages. Currently it appears to be a vehicle to push an environmental agenda, and appears to give environmental advantages that are not as great as claimed (if indeed there are actually any at all).

There are many inaccuracies, such as the claim that they have a guaranteed life of 8000 hours. They most certainly don't, and I have never seen a CFL that carries such a guarantee. They have a claimed life of 8000 hours, but this is achievable only under perfect conditions. The real usage life is comparable with an incadescent bulb and is often shorter. There is also a claim that they contain 1/5th of the mercury of a watch battery. Watch batteries have been mercury free, for at least the current century, or even longer. They don't contain 'traces' of mercury. If you can get hold of one without any flourescent coating (they are made as ultraviolet lamps), the mercury is clearly visible sticking to the glass (that is not a trace quantity, but a significant quantity). Having recently attempted to dispose of a few CFL lamps and flourescent tubes, I can confirm that there is no recycling facility (in Europe) and that they all go to landfill. The European

20.133.0.14 14:00, 28 February 2007 (UTC)

Watch batteries have been mercury free, for at least the current century, or even longer. They don't contain 'traces' of mercury.

I think you're considering just lithium "coin cell" watch batteries. Many of the more-traditional "button cell" form factors (A76/LR44, 392, etc.) still do contain mercury. See this PDF for a lengthy exposition: [2]

Atlant 15:32, 28 February 2007 (UTC)

Regardless of your claims, all watch batteries sold in Europe are all guaranteed to contain 0% mercury, and it says so on the packaging (and, as it happens, 0% cadmium). Indeed virtually all batteries sold are so marked. There is considerable consumer resistance to batteries not so marked. The cite is specific to the US, who are widely regarded as one of the most environmentally unfriendly countries on the planet. 86.137.53.240 18:47, 28 February 2007 (UTC)

Actually, the citation I provided does speak of the situation in Europe.

Atlant 19:01, 28 February 2007 (UTC)

The citation is actually hopelessly out of date. 86.137.53.240 is also a bit behind the times. Mercury in batteries was outlawed in Europe last year by the ROHS directive which virtually all of Europe has enacted into law. This places an upper limit on the amount of mercury (and other materials) that is permitted to be contained in batteries (IIRC 0.05% Hg by weight - not a lot given its high density). American supplied batteries have to comply if they wish to sell in Europe (and many US companies do, though the batteries are usually made in the Far East).

The manufacture of Mercury Oxide batteries is still permitted, but only for specialist uses that demand that particular technology (Mercury batteries make almost perfect voltage reference sources). Disposal methods are naturally heavily legislated. I B Wright 19:18, 28 February 2007 (UTC)

Some of 20.133.0.14's bile seems to have leaked out of this article onto my talk page. I quote in full, as I think he really wants to make his points about this article and not just about me personally.

== Compact Flourescent Lamps ==

Are you perchance an environmentalist [spits]? Why have you removed all the disadvantages from the above reference article?. You have succeeded in restoring much that is inaccurate and dowright lies. This is what environmentalists do to push what is basically a political agenda. They don't like people finding out that what they push is rarely the whole picture.

For example, no CFL has a guaranteed life of 8000 hours. No CFL in normal use comes anywhere close. In real usage, most rarely last longer than a normal bulb.

Also: they do not contain 1/5 of the mercury of a watch battery. Watch batteries have been mercury free since at least the beginning of the century. CFLs are most definitely not mercury free.

If you must correct articles, then at least don't include total bollocks.

20.133.0.14 14:09, 28 February 2007 (UTC)

I, however, have better things to do in the evening than to be spoken to in such a manner, in such a public forum as this. I simply reverted a major POV attack - which included some extraordinary claims and no cited references - that I found on the article the other day, suggesting that the anonymous author discuss such revisions here and cite some sources.

I would not have chosen the word 'guaranteed' referring to any light bulb's life - it is clearly the wrong word for that context - but I did not write it. I have an ordinary General Electric CFL in front of me whose packaging claims a life of 15,000 hours, saying that that is equivalent to 15 ordinary bulbs. It makes these claims under IEC60969, which specifies that "life to 50% of failures shall be not less than value declared by the manufacturer".

Indeed 'guaranteed' is the wrong word. The problem with the cited specification is that although it specifies limits of temperature and voltage etc. that the lamp is to be operated under, it specifies little of any value and relevance. It would be perfectly OK for a manufacturer to power up 100 of his CFLs and time how long it takes 50 to go out. Chances are that he would get a life of 8000 hours or 15,000 in the case of some lamps. But people just don't use lamps that way (OK a few might). Every time a lamp is started, some destruction of the cathodes occurs. Real world usage (perhaps turning on and off when entering and leaving a room) drastically shortens the life. In a real environment, a lamp will typically fail to outlast its incandescent equivalent. I have given up using CFL lamps in my house (except for one which stays on all night). All of the remainder have been outlasted by 100 watt incandescents (some of which are a good few years old - none of the CFLs have lasted longer than a year or two). 86.132.203.116 20:35, 1 March 2007 (UTC)

Curious. My experience has been quite different. Of the three CFLs that I've had fail (ever), two were run continuously right to the end of their life (quite possibly several years) and one was an infancy failure, probably of the electronics. While there's no doubt that starting fluorescent lamps is bad for them, there's not much evidence that, with CFLs, it's that bad for them.

Atlant 20:40, 1 March 2007 (UTC)

Running the lamps continuously is pretty well the ideal conditions that I have been talking about. The infant mortality is probably just bad luck. However all types of discharge and arc lamps suffer degradation of the cathodes on starting (it is a phenomenon that has been known about since discharge lamps were invented). Different types of lamp experience the effect to different degrees. Medium pressure mercury discharge lamps suffer from it more than low pressure mercury discharge. Standard flourescent tubes are examples of the latter and suffer from it less (but are by no means immune). Compact Flourescent Lamps are medium pressure lamps and display the phenomenon to a greater extent. That they have unacceptably short lives when frequently switched is well known and renders them unecconomical to use both cost wise and environmentally. But the environmentalists don't want you to know this. 20.133.0.14 10:37, 2 March 2007 (UTC)

You may have missed my point. Of all of the CFLs I have, only three have ever failed. The rest are still happily cycling, on and off, on and off. The two that ran to end-of-life, well, that's what happens eventually. The only failure that I can call abnormal was the one infancy failure.

Atlant 23:42, 7 March 2007 (UTC)

If my individual experience is of use: In my hall light, I was going through incandescent bulbs every 3 or 4 months. I stopped turning it on and off, seeing if the electrical surge was blowing them, then I changed the fixture - no difference. Over two years ago, I first bought Sylvania brand CFLs at the Dollar Store for, you guessed it - $1 ea.. Some either didn't work, or lasted only a few months, some are still burning. 19 months ago, I bought several packs of IKEA CFLs and replaced the rest of my incandescents including one outdoor light. I don't want to praise IKEA because their furniture is you get what you pay for, but the hall CFL is still burning over 13,700 hours later - and none of the others have yet failed. The hall light is on continuously, and the other house lights are turned on and off at least daily. Iagree with the critique of the mercury (Quebec generates electricity via dams - not coal) and the extra packaging (extra cardboard)in my purchases.

(Quebec, March 12, 2007)

—The preceding

Atlant 12:41, 12 March 2007 (UTC)

Without this level of personal abuse I'd be happy to try to contribute to these articles, but right now, my level of interest is low. --Nigelj 20:15, 1 March 2007 (UTC)

Abuse is something we all get to endure occasionally on here. I take it you are not much of a usenet user, where you just learn to take it on the chin.

I suggest that you may like to have a look at some of Wikipedia's

policies and guidelines - you'll see that this is neither a usenet forum nor a place where contributors just have to learn to take personal abuse on the chin. If you, and those you sympathise with, are planning on teaching us to do so, that may be a problem. --Nigelj

20:14, 2 March 2007 (UTC)

Hey, don't have a go at me. I couldn't agree with you more. I was merely pointing out the realities of life. The rules may prohibit personal abuse, but if you complain in the designated manner, Wikimedia will do nothing about it - believe me, I've tried. I don't condone it, but it happens and it seems that unfortunately that there is nothing that you nor I can do about it. I B Wright 12:53, 3 March 2007 (UTC)

I can sympathise with 20.133.0.14 to a certain extent (though maybe not to the extent of condoning abuse). Your reversion seems to indicate that you objected to the information included, and gives the impression that you supported the original article (though I note that 20.133.0.14 didn't contribute it). The problem is that much of it was factual, the remainder though having a basis in fact was open to arguement as to the magnitude of the effect (though in my view, was a probably a fair summary of the position). By reverting the article, you give the impression that you did not like the negative aspects aired, and I can understand the allegations made (if not the manner in which they were made). The problem is that many are becoming exasperated by environmentalists who continue to spout their one sided arguements on the principle that if they say it often enough, people will believe it. The uneducated do, but the more intelligent realise that nothing has actually been proven.

I hope that you do not allow events from deterring you from contributing, but do allow others the opportunity to balance up the article. The 'Totally Disputed' flag is certainly justly earned at present. I B Wright 20:51, 1 March 2007 (UTC)

In the summary that you attached to your edit, you claimed to have edited out the content that you clearly decided that you didn't want to be there. You did not claim to heve reverted the article. You removed much relevant information (as others have noted) and much that had at least had a degree of relevance (again as others had noted, but the details could well trigger discussion - had they been allowed to). You removed that which attempted to redress the balance of the article by putting the negative aspects across (and there are many). You claimed that it was uncited POV, but most of the current article is that anyway, and certainly contains much that is patently false (the one you replaced had attempted to redress that).

Since you clearly unilaterally decided that the negative aspects could not be permitted to taint the otherwise unbalanced and false impression that the article attempts to create, the allegation that you are an environmentalist that won't tolerate critisism of your environmental sacred cows - stands.

In case there is confusion, I should make it clear that I did not provide the version of the article that is causing the problems. However, it was clear that it did provide a degree of balance that the article craved so much. And it certainly had its basis in fact. 20.133.0.14 10:27, 2 March 2007 (UTC)

"rv" is very frequently used as an abbreviation for "revert" in WP edit summaries, along with "rvv" for "revert vandalism". I suggest you go and find out what POV stands for. Then read

WP:A. --Nigelj

20:14, 2 March 2007 (UTC)

I'd like to see a Total Environmental Impact section in the article that would go beyond the rather simplistic electricity savings benefits. For example, someone else on this page has pointed out that a CFL weighs about 3 times as much as a comparable incandescent bulb. If you include the weight of the heavy-duty plastic box most CFLs I have seen come in, the weight ratio is probably 4:1. Now what about all that extra energy that is needed to haul this extra weight around the world in gas-guzzling and global warming -causing trucks and airplanes? --72.244.191.252 05:13, 12 March 2007 (UTC)

For light weight objects, shipping costs depend more on volume then weight. Mostly you are paying to move the truck. But even with your 4:1 number, given that the CF's last more than 4 times longer, shipping effects would be a net plus for the CFs. And how about the environmental effects of extracting and shipping the fuel needed to make the electricity the is saved? --11:33, 12 March 2007 (UTC)

Good point. But what about all the energy that goes into producing all this extra material. In the picture I see a printed circuit board, a large capacitor, and other electronic components. Then of course there is the big plastic box in which most CFLs are packaged. How much energy is used and pollution generated to make all this stuff. And in the end, all this (including the mercury) is going to end up in thousands of landfills all over the world. We could have a long discussion about this. I don't have all the answers and neither do you. My concern is that 50 years from now we (our kids and grandkids) might be looking at CFLs the way we now look at leaded gasoline, lead paint, DDT, etc. They all seemed like great ideas at one point. They were also heavily promoted by the U.S. government and/or the companies that made them. Perhaps we should try to learn from the mistakes of the past.

The bottom line is unless the article can include a balanced view of pros and cons of CFLs including a reference to a trustworthy total environmental impact (full life-cycle) analysis, it should be limited to just a description of CFLs and an explanation of their function. --72.244.191.252 14:41, 15 March 2007 (UTC)

Whatever amount of energy is consumed in making and packaging CFLs must cost less than the wholesale price of the bulbs, before subsidy. The companies that make them aren't losing money. CFLs typically save consumers several times their cost in energy savings over the life of the bulb, even at retail unsubsidized prices. So there is certainly a net energy savings. There are plenty of published sources to back up the information in this article. Published sources and not original research determine what belongs in Wikipedia.--agr 19:02, 16 March 2007 (UTC)

Hmm - that depends on how you measure environmental cost. Without wishing to be controversial, there are questions one can raise about the effect of replacing standard incandescent lamps in landfill with CFLs, as CFLs contain mercury and other metals which are not environmentally benign. In theory, this is obviated by recycling, but this too has a cost. The embodied energy of a CFL is far higher than an incandescent as well, so looking at the total lifetime cost of manufacture, use, and disposal is not a simple as looking at the wholesale costs of lamps and their respective electricity use. Other things add to the complexity of the analysis: CFLs have a power factor of less than unity, which means that mass replacement of incandescents with CFLs will force expensive re-engineering of the electricity supply network. It may well be that, taking these and other considerations into account, use of CFLs is still of net benefit, but it is by no means obvious. I agree that verifiability of facts in Wikipedia is necessary, so I'm not defending the addition of unsourced material - simply pointing out that the situation is not as clear cut as a lot of sources make out. Regards, WLD^talk|edits 19:34, 16 March 2007 (UTC)

The energy cost of manufacturing CFL has to be less that what they are sold for. There is no reason to suppose that carting one CFL and its packaging to a land fill costs appreciably more more than carting 4-6 incandescents and their packaging. As for power factor, even CFLs with a poor power factor draw fewer VAs than the incandescents they replace and most homes have inductive loads most of the time, such as refrigerators and HVAC pumps and fans. Most CFLs are capacitative loads. BC Hydro says "using CFLs in a home will not affect power quality appreciably." [3] For large facilities one can purchase high power factor (>0.9) CFLs or add external power conditioning. Such considerations are a normal part of large facility engineering. Power companies are the people who are most concerned with power factor management and they are begging (and often paying) people to switch to CFLs. --agr 21:32, 16 March 2007 (UTC)

It's cheap to dump mercury in landfills, just as it was cheap to use

tetra-ethyl lead in gasoline - until the overall environmental impact was determined. Similarly, asbestos was a cheap thermal insulator, but companies are being bankrupted paying out for mesothelioma compensation. I didn't confine myself to solely energy costs. The simple energy cost calculations for CFLs externalise many environmental factors. Thanks for the BC Hydro ref. Reading it, I come to a (slightly) different conclusion to you - the sentence is ambiguous. I'd say that use of CFLs does not affect power quality appreciably in the home - but overall, the power companies would have to engineer for changes in power factor - hence large facilities are encouraged to buy high power factor CFLs or purchase power conditioning (an extra cost). WLD^talk|edits

23:09, 16 March 2007 (UTC)

I would argue total expenditure is a VERY good proxy of total environmental impact. So if Product + Energy Costs of CFLs is cheaper than Product + Energy costs of traditional lights, then CFLs probably do have lower overall environmental impact. If you think about it the cost of something is basically directly related to the energy necessary to create it (whether through mining, processing, etc). 21:47, 16 March 2007 (UTC)

The examples I've given above show that cost was a very poor proxy for environmental impact in those cases. CFLs contain mercury. Their PCBs (if outside the EU) may well use lead-containing solder. Now, CFLs are unlikely to have a similar impact to either tetra-ethyl lead or asbestos, but the point is that there are costs to the environment not captured simply by energy cost. The problem is that these external costs are not easy to quantify. I have no doubt the Dutch or Scandinavian governments either are working on it, or have worked on it, as they tend to be clued up on such matters, but I've not seen any reports. There is this [4], which is related, but ongoing. In the EU, RoHS Directive, (2002 95 EC) which became law July 1 2006 and restricts CFLs to containing 5 mg of mercury or less (see Restriction of Hazardous Substances Directive and Waste Electrical and Electronic Equipment Directive). There's a telling quotation here [5]

General Electric has been making compact fluorescents for 20 years. Now the company admits that the little bit of mercury in each bulbs could become a real problem if sales balloon as expected.

"Given what we anticipate to be the significant increase in the use of these products, we are now beginning to look at, and shortly we'll be discussing with legislators, possibly a national solution here," says Earl Jones, a senior counsel for General Electric.

In my view, CFLs are not an unmitigated good. WLD^talk|edits 23:09, 16 March 2007 (UTC)

The U.S. EPA has a published analysis that shows that even if all the mercury in CFLs is released into the environment, there is a net reduction in mercury compare to producing the same amount of power by coal. And in most parts of the word, coal is the variable source for electric power, even if most of the power comes from hydro or nuclear. So if energy is saved, likely less coal is burned. When discussing options for improving the environment, there are few "unmitigated goods." CFLS come pretty close. --agr 01:15, 18 March 2007 (UTC)

You haven't cited which analysis you are talking about. Its probably the one used as backing for articles/leaflets like these: http://www.energystar.gov/ia/partners/promotions/change_light/downloads/Fact_Sheet_Mercury.pdf and http://www.nema.org/lamprecycle/epafactsheet-cfl.pdf. These factsheets have some carefully chosen statistics in them. As the first says, talking about coal-fired power plants, "A power plant will emit 10mg of mercury to produce the electricity to run an incandescent bulb compared to only 2.4mg of mercury to run a CFL for the same time." - but the first states "EPA is implementing policies to reduce airborne mercury emissions. Under regulations issued in 2005, coal-fired power plants will need to reduce their emissions by 70 percent by 2018." - this is not taken into account, as a 70% reduction in power-plant emissions of mercury will mean that CFLs end up releasing more into the environment than coal, because even if the power generated to run the CFL has a zero mercury emission, the CFL itself contains mercury. Places like France, which generates over 75% of its electrical power by nuclear [6]; and Norway, which generates over 99% of its electrical power by hydroelectric [7]would actually see an increase in mercury emissions if they move to CFLs. (Note that Norway's consumption figures are over 95% is hyrdoelectric - the difference is made up by electrical power imports). From the NEI figures for nuclear, Vermont generates over 70% of it's power by nuclear, meaning that it too will experience an increase in mercury emission by moving to CFLs. So the key is recycling. There's a good Canadian study here: http://www.mercurypolicy.org/new/documents/FluorescentLampsReportOctober2005.pdf, which although very positive about future benefits, makes it clear that current recycling rates are nowhere near high enough. Overall, the pros- and cons- of CFLs and their mercury content are not clear cut, and in my view it is disingenuous to say they are. WLD^talk|edits 08:29, 18 March 2007 (UTC)

By the way, I'm not sure what you meant by "And in most parts of the word, coal is the variable source for electric power" - could you clarify? WLD^talk|edits 08:29, 18 March 2007 (UTC)

Coal plant output is most easily adjusted to meet varing loads. It supplies the marginal demand. France and Norway are connected to the EU grid. If people there reduce their use of electricity, the excess power can be sold to other countries who can then burn less coal.--agr

Not it is not. See Base load. The most easily adjusted is hydro-electric, followed by gas- and oil-fired power stations. Coal (and nuclear) take the longest to vary their output, and are used to service the 'base load' - coal does not service the marginal demand. Regards, WLD^talk|edits 07:41, 19 March 2007 (UTC)

I don't see how replacing tons of mercury released from a few hundred smoke stacks with (maybe fewer, but still way too many) tons of mercury dumped into thousands of landfills is an "unmitigated good" for the environment. Why is nobody suggesting that we try to reduce/eliminate the mercury emissions from the power plants? Is it really technically impossible, or just not as profitable as replacing every incandescent bulb with a CFL? --72.244.191.252 21:04, 18 March 2007 (UTC)

Err - 'they' are suggesting just that: "EPA is implementing policies to reduce airborne mercury emissions. Under regulations issued in 2005, coal-fired power plants will need to reduce their emissions by 70 percent by 2018.". In theory, CFLs should be recycled (rather than just dumped in landfill), so the mercury contained in them should remain out of the general environment. However, recycling rates need to improve. WLD^talk|edits 21:23, 18 March 2007 (UTC)

As to the statement "The energy cost of manufacturing CFL has to be less than what they are sold for", that might be true. However, nobody really knows what the real manufacturing cost of a CFL is except for the manufacturers, and they keep it secret. The only thing we know is the current(!) selling price. But what if this price is significantly lower than the real manufacturing cost? What if the selling price is just an "introductory offer" meant to create a truly mass demand for a product with much higher profit margins than the product being replaced? Would not be the first time. What if the selling price of CFLs doubles after the incandescents have been outlawed to reflect the real manufacturing cost? What if the price triples when the manufacturers have to include the cost of safe disposal in the price (as they should be required to do)? Or maybe the price will remain the same, but our grand-children will end up paying the cost of the cleanup of thousands of toxic waste sites (the current landfills).--72.244.191.252 22:23, 18 March 2007 (UTC)

It's highly unlikely that several independent companies have been selling CFLs below cost for twenty years in the hope of making a killing when incandescents are banned, a notion that has only surfaced recently. In any case, the cost to manufacture can be estimated by looking at the photo in the article of one that has been taken apart. All the components are standard electronics catalog items. The glass envelope and base are not likely to be much more than a factor of two or three times more expensive to make than an incandescent. What else is there?--agr 04:00, 19 March 2007 (UTC)

I carried out a trial for another forum. I though the results were applicable here. For the benefit of non-European readers, all appliances sold in Europe are marked with an energy efficiency rating ranging from A (most efficient) to G (least efficient).

While out shopping I had a look at the low voltage halogens to get the current claimed energy efficiency. Here I got my first surprise. While every other light bulb type (including CFLs) had the energy efficiency rating and the lumen output on the packaging, both were curiously missing from every low voltage halogen that I looked at (in 2 shops selling Philips and Osram). The packaging had an area of white space where the energy efficiency diagram would have gone. Hmm.

Lumens is an odd unit to use, because without any further information it doesn't convey a lot of information. Lumens is not simply emitted light intensity.

I set up to test 3 types of lamp. A 20 watt Philips Compact Fluorescent Lamp (CFL); a 20 watt 12 volt halogen capsule lamp (LVH) and (because the 20 watt CFL claimed to replace a 100 watt GLS lamp) a 100 watt GLS lamp. The photo detector was a cadmium sulphide device which, on checking the spectral response, fairly closely matches that of the eye, so differences in spectrum of the different lamps should be mostly eliminated. This type of detector has a completely liner response so comaring differences in brightness is easy. Each lamp was operated at the centre of a lamp shade that had white sloping walls that should direct most of the light in one direction.

First up was the 100 watt GLS lamp. We will call the light intensity 100% as a reference.

Next was the 20 watt LVH. The intensity was 43%.

Finally the 20 watt CFL. The intensity started at 24% and rose to final steady intensity of 39% after 8 minutes.

That doesn't sound like much of an equivalence. A search around the internet turned up much discussion about how much of the supposed energy saving of CFL lamps came from using CFL lamps of lower intensity than the lamp it supposedly replaced. Thus giving a 5 times energy saving. The claimed lumen output of a 20 watt CFL is 1200 Lumens. It took quite a bit of searching and I failed to find a specified lumen output for a 100 watt GLS. What I did find is that the 20 watt lamp should really replace a 60 watt GLS. I also turned up what I assume is the environmentalist response, "Most people are unlikely to notice the reduction in brightness". All I can say is that if that is true, then most people have a serious sight defect. You can't help but notice the difference, and it is fiddling the numbers. So I repeated what I had done with a 60 watt reference.

60 watt GLS: 100% (reference)

20 watt LVH: 75%

20 watt CFL: 45% warming up to 69% after 8 minutes.

So the 20 watt CFL isn't even an equivalent of a 60 watt bulb. I didn't have a 40 watt bulb in my spare cupboard, but on the numbers the 20 watt CFL would be equivalent to a 41.5 watt bulb. Not a *5 times* energy saving but only a *2 times*. We have been comprehensively had.

But look at that 20 watt LVH again. It is brighter than the equivalent wattage CFL. That is *it is more efficient* - nearly 9% more efficient. Further, the CFL gets dimmer as it ages, the LVH does not.

It is clear that the CFL is the wrong solution to the environmental problem, and the advantages claimed are considerably exagerated. Clearly the low voltage halogen lamps should be marked 'energy efficiency A', and the CFLs should really be marked 'energy efficiency B'. Also, LVH lamps don't have any of the disadvantages of the CFL. And they certainly don't contain mercury.

But why don't the lamp manufacturers put the energy efficiency on the LVH lamps? One reason is probably that they simple and ridiculously easy to make. They don't have anything like the scale of profit of the vastly more complicated CFL. But the environmentalists have already made up their minds that everyone has to switch to CFL lamps regardless of the facts.

20.133.0.13 09:33, 16 March 2007 (UTC)

Arbitrary subheading for ease of editing (1)

Something is wrong with your numbers. There is no possibility that the LVH lamps were more efficient than the CFL numbers. Even manufacturers' data sheets don't support this contention.

Atlant 11:40, 16 March 2007 (UTC)

The Philips numbers are available from their website.

For the UK (240V AC supply):

• 60W Frosted A-shape is 700 lumens (11.7 lumens/watt)
• 100W Frosted A-shape is 1330 lumens (13.3 lumens/watt)
• 20W Capsuleline 4000 Hour LV Halogen is about 300-310 lumens depending on model (15.0 - 15.5 lumens/watt)
• 75W Capsuleline 4000 Hour LV Halogen is 1575 lumens (21.0 lumens/watt)
• 100W Capsuleline 4000 Hour LV Halogen is 2200 lumens (22.0 lumens/watt)
• 20W MASTER-PL Electronic CFL is 1200 lumens (60 lumens/watt)
• 23W MASTER-PL Electronic CFL is 1500 lumens (65.2 lumens/watt)

Incandescent lamps produce pretty much the same output throughout their lives (There is some darkening of the lamp due to deposition of tungsten from the evaporated filament on the inside surface of the envelope, but this doesn't apply to halogen lamps). Fluorescent lamps drop in output over their lifetime. If you want to replace the 100W non-halogen incandescent above, you'll need either the 75W LV Halogen, or the 23W CFL, as the 20W CFL does not have sufficient output. If the 75W halogen is on a dimmer, the output can be regulated to produce the same luminous intensity as the 100W non-halogen - which will save some power, and extend its life, so long as it is still hot enough for the halogen cycle to work. Dimmable CFLs are appearing, but I don't know what effect on power usage and lifetime dimming has. So it looks like the CFL is far more efficient. However, one problem is that the colour rendering index (CRI), which for the incandescents is generally rated between 95 and 100 (often assumed, incorrectly, to be 100); is for the CFLs above, 82 (from the manufacturers figures), so even though the total light output is equivalent, the quality of the light produced is radically different - rather than being a smooth spectrum, it has significant peaks. These peaks probably don't interact well with the CdS sensor (which to match better the spectral sensitivity profile of the human eye should be a CdS/CdSe sensor, the CdSe being Cadmium Selenide).

So, your CFL may possibly be old , and therefore not producing as much output as when new, and the spectral sensitivity of your photocell/sensor may not be what you expect. Regards, WLD^talk|edits 13:07, 16 March 2007 (UTC)

You omitted the official lumen rating of the 100 watt standard GLS lamp. It is also officially rated at 1200 Lumens. But if you compare a brand new 20 watt CFL with a 100 watt GLS bulb, the CFL is very noticeably dimmer. I would say that under half brightness is probably correct because the eye is non linear. The problem is that measurement 'the Lumen'. Lumen is not a measurement of light intensity (the unit of that is the Candela). Lumen is a much more hazy measurement of light flux (broadly the light per unit area landing on a surface - though it isn't quite this). The Lumens can vary depending on how you measure them and where in the spherical surround of the lamp you measure it. It wold be easy to pick a sot with highest light flux for a CFL (easy given that the light varies enormously depending on the direction - a function of the lamp design). It would also be easy to pick a direction on the 100 watt GLS that has a realtively low flux. Voila, two lamps of equal brightness - at least on paper, but not when used to light a practical room. I dug around some historical information, and I found varying values for the lumens the largest of which was 2500 Lumens. Presumably the variaton is down to the method of measurement.

I do know that because of the relatively thick filament, the low voltage halogen lamp has a much lower surface area to volume ratio. This allows it to operate at much higher temperatures than convenional bulbs and consequently is much more efficient. As a rule of thumb it is considered that low voltage halogens give a little less than same amount of light as standard bulbs of three times the wattage. We are told that CFLs are five times as efficient, but if they are really dimmer as seems to be the case, then that five times figure is called firmly into question. It must be less than that - three times perhaps? It might be that low voltage halogens might at least be nearly as efficient as CFLs. I know the official Lumen rating is lower, but I have plenty of low voltage halogens, and 40 watts of them is definitely brighter than a 100 watt GLS. But the quoted numbers don't support this. But then numbers can be manipulated can't they.

When I read the lead part of this, I dropped into my local DIY store on the way home. The energy rating and the lumens were indeed present on all the lamps except the low voltage halogen lamps (they were own store branded). 81.157.133.142 17:19, 16 March 2007 (UTC)

Arbitrary subheading for ease of editing (2)

I don't know where you got the official lumen rating of 1200 for 100W General Lighting Service (GLS) Lamps. A quick look at manufacturer catalogues gives higher numbers. The Wikipedia lumen article states "A standard 100 watt incandescent light bulb emits approximately 1700 lumens in North America and around 1300 lumens in 220V areas of the world." I agree than numbers can be manipulated, but I don't buy a grand conspiracy to manipulate lumen ratings - I suspect there's a problem with your method somewhere. As I understand it, the lumen rating is meant to be the total usable light output of the lamp, summed over the entire sphere around the lamp, so shouldn't be direction dependent. If I'm wrong, please say so. WLD^talk|edits 18:09, 16 March 2007 (UTC)

If we're comparing "standard life" ordinary incandescent lamps (say, 750-1000 hours) with "standard life" halogen lamps (say "2000 hours"), halogen lamps are nowhere near "three times" as bright as ordinary incandescents of the same wattage. Also, it has very little to do with the "thickness" of the filament; instead, it has everything to do with the halogen cycle allowing the filament to operate just a bit hotter without rapidly evaporating away.

Here's Sylvania claiming "20% greater energy efficiency" for halogens over odirnary incandescents: [8]

Atlant 23:25, 16 March 2007 (UTC) (revised 00:18, 17 March 2007 (UTC))

You are quoting figures for mains voltage halogens. Low voltage halogens are approximately two and a half times as bright as a similar wattage GLS lamp. They also have a life of 4000 hours (it says so on the packaging). And it has everything to do with the filament thickness. Why do you think American 100 watt lamps are quoted as 1700 lumens compared with 1300 lumens for the European version. It is because it uses almost twice the current and therefore the filament has twice the cross sectional area. Since the surface area to volume ratio is lower, the filament can be operated at a higher temperature and is thus more efficient. The inverse relationship of filament temperature to cross sectional area has been known about almost since Edison stole the invention. The halogen cycle is but one factor in the equation. 81.157.133.142 17:22, 17 March 2007 (UTC)

Arbitrary subheading for ease of editing (3)

You are correct, I mis-spelt 1300. When I read the above, I became somewhat intrigued. It is possible to rig the Lumen rating because of what the Lumen represents. It is light flux rather than light output. The correct figure to quote for a light bulb should be 'total Lumens' measured using an integrating sphere (basically a large metal sphere painted white on the inside - a photometer peeps in through a hole). However , just plain unqualified Lumen can mean almost anything. It could represent the maximum light flux in some specific direction (some lamps have to be measured this way - such as projector lamps or spot light bulbs).

I think the figure quoted for lamps is total lumens measured using an integrating sphere. See here for an accessible definition of the lumen http://www.wisegeek.com/what-is-a-lumen.htm and here http://www.theledlight.com/lumens.html. This reference may help also http://www.fordav.com/publications/cb17.pdf. WLD^talk|edits 21:15, 17 March 2007 (UTC)

Since the lamp in my spare room has a large paper almost spherical lamp shade, I considered it might prove ideal for an experiment. I managed to find several Cadmium Sulphide cells in one of my scrap boxes - one was adequate. This is the ideal type of photo detector to use because its spectral response closely matches that of the human eye (and essential requirement to measuring lumens) - and (as noted elsewhere) its characteristic is perfectly linear. Placing the photo cell in small vise on a table underneath the lamp, I noted the resistance. I replaced the 100 watt bulb with a Philips 20 watt CFL (a brand new one - it is one of the type that has 3 U-shaped tubes). I allowed it to warm up for 10 minutes and then took the reading. Just by observation it was definitely nowhere near as bright. And the photocell agreed. Assuming the 100 watt bulb to be 1300 Lumens meant that the 20 watt CFL was just 489 Lumens. Hmm, that is quite close to the results that started this thread where he made it 507 Lumens (39% of 1300).

Not quite sure what you mean by "its characteristic is perfectly linear". The spectral response is certainly not flat. See Figure 6 here http://www.aicl.com.tw/cds/p6-12.htm and Figure 1 here http://www.ecs.soton.ac.uk/publications/rj/1994/transduc/ross/ross.html - and to get a spectral response that peaks at the same point as average human vision requires a mixture of Cadmium Sulfide and Cadmium Selenide. Note also that fluorescent lamps have a 'peaky' emission spectrum, unlike the fairly smooth spectrum from an incandescent lamp, which probably also affects the measurement.WLD^talk|edits 21:15, 17 March 2007 (UTC)

Indeed the spectral response of the CdS is not flat. But then neither is the human eye. The CdS cell response is fairly close to the human eye's response (that is why they were extensively used in camera exposure systems). The measurement of Lumens requires a detector thay matches the eyes response. If the photcell meets these requirements then peaks in the spectral output of the lamp will be taken into account by the cell, just as the eye does. 81.157.133.142 15:27, 18 March 2007 (UTC)

You still haven't said what you mean by "its characteristic is perfectly linear" - please clarify. In addition, take a look hear for a quick overview of the spectral response of the human eye http://photo.net/photo/edscott/vis00010.htm - it varies according to the level of illumination (depending on whether you are using rods or cones) and in colour vision has at least three peaks. I've linked elsewhere to similar plots for CdS/CdSe sensors, and you can see by inspection that they are not identical. They may well be similar, and good enough for photography, but that does not make them identical. If the spectral peak in the emission spectrum of a fluorescent lamp corresponds to a low point in the spectral sensitivity of the human eye, the lamp will look dimmer to the human than the sensor. This is why specifying phosphors is difficult, because people's sensitivities vary. The same applies in reverse as well, the eye can be more sensitive to particular wavelengths than the sensor, so what the sensor sees as dim, the eye sees as bright. WLD^talk|edits 17:02, 18 March 2007 (UTC)

I think you are trying to be far too pedantic. The characteristic curve that I have here (published by Philips) for the ORP12 (which is what I ended up using), plots light flux against resistance. The line is dead straight. It is necessary to take the reciprocal because, of course, the resistance is inversely proportional to light flux. The spectral response curve is close enough to the eye for the purposes of the trial. You could critisise much else about the 'desert island' measurement techniques, but it doesn't make the results invalid. You are going to have to come up with a lot else to explain a 60% error. 81.157.133.142 18:46, 18 March 2007 (UTC)

Well, it looks like the comment (if true) by 212.183.134.130 below explains it, which is a very interesting point. As for being pedantic, believe it or not, but I was trying (ineffectually) to help - poor experiment design is the downfall of many experimenters who think they have extraordinary results. WLD^talk|edits 19:19, 18 March 2007 (UTC)

Oh, and there's a particularly good example of the difference between photosensor sensitivity and human eye in this paper :http://irc.nrc-cnrc.gc.ca/pubs/cp/lig3_e.html - look at Figure 4. 17% difference at a critical point in the spectrum right there. Please note that I'm not trying to prove you are wrong, I'm just trying to identify and (hopefully) minimise the sources of error in your method. If CFLs are better than incandescents, then they should be when all the facts are known, and the key is to get at all the facts. From a purely personal point of view, I strongly dislike the quality of light produced by CFLs - but despite that, approximately half the GLS lamps in my home are CFL, because I like the cost savings. Those that are left are either short duty cycle, or on dimmers. Regards, WLD^talk|edits 21:54, 18 March 2007 (UTC)

Points taken. I never claimed, the method was a scientific bullet proof trial. However, I feel it is adequate to the task (see the addendum below). The photocell used isn't perfect, but adequate to the task. Yes there are errors, but they certainly don't account for the huge differences between the Lumens measured and the Lumens claimed. 81.157.133.142 19:32, 20 March 2007 (UTC)

So here is the dilema. The quoted Lumens for the 20 watt CFL is 1200 Lumens (It's actually marked on the lamp). But many people have observed that they are not as bright as the 100 watt bulb they replace. I doubt that anyone would notice the difference between 1300 and 1200 Lumens (a mere 7 1/2% reduction). It occured to me that the structure of the lamp makes it far from an isotropic radiator and I tried to find a direction where the photocell indicated a flux of 1200 Lumens. But I couldn't find one - it certainly varied though. So the question is: where did that figure of 1200 Lumens come from? The obvious lower brightness of the CFL lamp sugests an element of skulduggery, but I just couldn't figure out how.

To measure the lumen rating of a lamp, you need an integrating sphere. One vendor is here http://www.sphereoptics.com/, another here http://www.avantes.com/Accessories/integsphere.htm - there are others. I don't think a single photocell and approximately spherical lampshade does it, I'm afraid. WLD^talk|edits 21:15, 17 March 2007 (UTC)

No you don't. To measure total Lumens you do. But the term 'Lumens' on its own is vague, because it can refer to light flux in one particular direction (i.e. for a non isotropic radiator). 81.157.133.142 15:27, 18 March 2007 (UTC)

It's not vague in the slightest. The definition is given in the Wikipedia article on lumen (unit) "If a light source emits one candela of luminous intensity into a solid angle of one steradian, the total luminous flux emitted into that solid angle is one lumen. Alternatively, an isotropic one-candela light source emits a total luminous flux of exactly 4π lumens. The lumen can be thought of casually as a measure of the total "amount" of visible light emitted.". I suspect you are not completely understanding the definition. The need to use the integrating sphere is to compensate for anisotropy. Once can, of course, measure the luminous flux into a solid angle less than a sphere (which is done when specifying projector lamps), but you then need to specify the solid angle in some way. A lumen measurement on its own is total lumens, measured using an integrating sphere. WLD^talk|edits 17:02, 18 March 2007 (UTC)

But if the 20 watt CFL really is only 40% as bright as the 100 watt GLS, then the claim that they use 80% less energy for the same amount of light is greatly exagerated (and really comes as no surprise). It seems to be closer to a 50% reduction (i.e. twice as efficient). I am on the case, but whether I find the answer is another matter. I suspect that the environmentalists don't want it found.

The problem is that the low voltage (12 volt) halogen is officially two and half times as efficient as its GLS counterpart (A little less than double in the US). Unfortunately, I didn't have a suitable low voltage halogen for a comparison. It's almost worth buying one - purely in the interests of science. 81.157.133.142 17:22, 17 March 2007 (UTC)

While your enthusiasm for doing your own research is to be commended, please be aware that Wikipedia is not the place for publishing personal research. The policy is Wikipedia:Attribution, which means that no matter how interesting your results, they cannot be included in Wikipedia unless you can give verifiable reputable published sources. The aim is not truth, but verifiability. Don't let that stop you from carrying out your practical research, but please don't try and include the results in the article. WLD^talk|edits 21:15, 17 March 2007 (UTC)

I think you will find that that applies in the articles themselves. This is a discussion page where much material has no attribution, not is it ever likely to.81.157.133.142 15:27, 18 March 2007 (UTC)

Which is why I said "please don't try and include the results in the article". You are free to add anything you like to the discussion page for the article, so long as it is relevant to improving the article itself. WLD^talk|edits 17:02, 18 March 2007 (UTC)

I'm sorry? Where exactly did I include any results in the article?81.157.133.142 18:32, 18 March 2007 (UTC)

You didn't. The point is was making was that you shouldn't. I suspect you may have got that particular point. <grin> WLD^talk|edits 19:19, 18 March 2007 (UTC)

I got around to buying a 12 volt 20 watt capsule halogen lamp. Sure enough there is no lumen information on the packaging or energy efficiency rating. Recalibrating my photocell against the 100 watt lamp ordinaire, I made the lumen output of the low voltage halogen 452 Lumens, a reduction of around 6% on the CFL. I note that 20.133.0.13 made the brightness of the low voltage halogen higher than the CFL, but I wonder if he took account of the fact that 12 volt low voltage halogens aren't actually rated to operate at 12 volts. They are rated at 11.5 volts (at which I ran mine). The extra half volt would give more light (and shorten the life). 81.157.133.142 19:32, 20 March 2007 (UTC)

I've already referenced a lumen rating for a 12V capsule lumen rating above. Philips publish them on their website. Go to http://www.philips.co.uk, choose 'Lighting'/'Product Catalogue' from the top edge drop down menu; then 'Lamps & Gear' when the next menu (eventually) appear, and navigate through the various submenus to LV Halogens. Capsuleline 400 Hr 20W 12V gives 300 lumen. The 2000 hour variant is 250 lumen. Incidentally, I've not found any support for the statements given elsewhere on this talk page for the phosphor efficiency drop when using 'incandescent white' phosphors. WLD^talk|edits 20:01, 20 March 2007 (UTC)

I assume you meant to type '4000 Hr' and not '400 Hr'. If so, something is wrong here. The 2000 Hr variant of the lamp should have a higher Lumen output not lower, because the shorter life is the price paid for running the filament hotter.

Yes, I did mean to type '4000 Hr'. I agree that it looks odd, but I'm simply reproducing Philip's numbers. I've given the source, so you can check. WLD^talk|edits 18:05, 23 March 2007 (UTC)

I've not been able to find anything either. However, I did recall something being mentioned when I studied fluoresent lighting at college. I managed to find my old college notes in the loft. Unfortunately they didn't cast any light (!) beyond a mention that of the three principal colours of fluoresent tubes available at that time, 'daylight' tubes were brighter (i.e. more efficient) than 'white', and in turn 'white' was brighter than 'warm white'. There was no information as to the degree of brightness difference. The 3 colours are in descending order of colour temperature, so there may indeed be something in it. This was in the 1970's so there is no information on 'incandescent white'. Current manufacturer's data gives the same Lumen output regardless of phosphor colour, but common sense dictates that different phosphors are highly unlikely to have the same efficiency. In fact, the Wikipedia article on 'fluorescence' mentions different yields of different phosphors, but doesn't elaborate much. Of course, the inability to find readily available information may be by design.

Well, it makes sense that 'warmer' (redder) phosphors are less efficient, as the transition that produces the redder colour is further away from the UV driving frequencies emitted by the excited mercury vapour, and while there is a lot of work being done on finding phosphors that emit two photons of visible light for every one UV photon emitted, they are not in production as yet. It's possible that phosphor formulations and their efficiencies are trade secrets, which could explain why there is a dearth of public data. WLD^talk|edits 18:05, 23 March 2007 (UTC)

The one thing I have been able to find out is that CFLs loose around 45% of their light output in the first one eighth of their life, and then around 45% of what's left in the next eighth and so on (a greater fall off than the low pressure fluorescent tube). Halogen lamps, of course, don't loose any brightness at all during their lives. 86.134.124.16 17:14, 23 March 2007 (UTC)

That figure looks a bit higher than those generally bandied about. As usual, any reliable/credible/verifiable source for it? It may well be true for 'first generation' CFLs, but I think more recent ones don't suffer output fall-off so badly. I think manufacturers do actually quote figures for this. A quick Google finds this [9] and this [10] and this [11]. I think the term is 'lumen depreciation' or 'lumen maintenance'. Ah - this [12] is a good set of slides - see "CFL Lumen Maintenance" slide. A phrase I'm seeing in various places is "At 40% of rated life, must be 80% of initial (100-hour) rating" WLD^talk|edits 18:05, 23 March 2007 (UTC)

Some remarkably interesting stuff. Interestingly: one of those sources (4) quoted actual test results for output fall off after just 100 hours that ranged from 90 to 95%. using the Lumen fall off quoted above of 47% after 1000 hours (1/8 of life) and extrapolating that to the end of the life, doing a least squares fit to find an equation for light fall off (it turns out, not unsurprisingly, to be an exponential equation and it gives a good close fit - as it should).

L=0.9993637^h

where L=percentage of starting light output
      h=hours

Solving for 100 hours gives 93.8% of new output - nicely in the range referenced above. 86.137.167.111 14:54, 31 March 2007 (UTC)

Arbitrary subheading for ease of editing (4)

A gross error has been made. When dealing with fluorescent lamps, the quoted Lumens is the nominal total Lumens for that lamp design. But not all versions of any particular design have the same actual total Lumens. The actual phosphor used varies the output very significantly. As any lighting designer will know: to get the actual total Lumens for a specific lamp you have to multiply the nominal total Lumens for the lamp design by the conversion factor for the actual phosphor used. The manufacturers apparently like it this way because then they can use the same packaging (marked with the nominal lumens) for all versions of the lamp.

The nominal total Lumens for low and medium pressure mercury fluorescent lamps is measured using a phosphor that equates broadly to a colour temperature of 6500K (what is described by many European manufacturers as the 'daylight' phosphor). But 6500K is very cold unwelcoming light, and something warmer is usually wanted. The problem is that the phosphors used in the fomulation to provide the redder (i.e. warmer) colours are less efficient (generally, the greater the shift in wavelength from the ultra-violet, the more inneficient the phosphor). The phosphor used in Compact Fluorescent Lamps (CFLs) made for domestic use is aimed at closely mimicking the colour of a normal general lighting service (GLS) filament lamp (~2500K). As such it is one of the warmest and least efficient phosphors.

The 'daylight' phosphor is assumed to have a conversion factor of unity, so a 20 watt (a popular size) CFL with the daylight phosphor would indeed have a total lumens of 1200. But the actual phosphor used (known as 'incandescent white') by many European manufacturers (The American company GE have decided to call it 'warm white' - why is a mystery because there is already a 'warm white. (about 3200K).

The conversion factor of the 'incandescent white 'phosphor is close enough to 0.4 that that is the figure usually used. When you multipy the nominal design total lumens (1200) by the conversion factor of the phosphor you get the real total lumens for that lamp.

1200*0.4=480 total Lumens.

Individual lamps, of course, will vary. Those contributors who measured their 20 watt CFL lamps against the 100 watt GLS lamp are actually in the right ball park. This works out to 24 lumens/watt. Pretty close to the 22 lumens/watt of 12 volt capsule low voltage halogen (LVH) lamps. The gradual decline in brightness of the CFL lamps does mean that for most of its life the LVH lamp will be brighter (and thus more efficient). But we've known that all along. As to whether the failure to factor in the conversion factor of the phosphor is deliberate or through ignorance: I wouldn't like to comment. 212.183.134.130 18:26, 18 March 2007 (UTC)

Thank-you very much, 212.183.134.130. In my case, it was out of sheer ignorance, and I am very grateful for being put on the correct path. This is such an important point, I think it should be in the article - can you cite credible, verifiable sources for the above - it's not that I don't believe you, it's just that without verifiability, it's likely to be summarily removed from the article. It seems to run a coach-and-horses through the proposition that CFLs are as environmentally friendly as they are claimed, which, in the context of the recent decisions in Cuba, Australia, and the EU is newsworthy and important. WLD^talk|edits 19:19, 18 March 2007 (UTC)

If true, this certainly answers my perception that these CFL lamps are noticeably dimmer than their supposed equivalent filament lamps. I do recall that the lighting gallery (sadly now dismantled) in the London Science Museum had an exhibit where the relative differences in brightness of fluorescent tube phosphors was very ably demonstrated (can't remember any numbers).

In my own company, we had a firm of designers in to redecorate the lavatories and at the same time to change the lighting to something more energy efficient. Although I wasn't privy (!) to the discussions and arguments, I do know that the designers, to many peoples' surprise insisted on low voltage halogen rather than any form of fluorescent lighting. Maybe they really did know something that the rest of us didn't. I have to say that for the relatively small number of 50 watt lights, it is well lit. I B Wright 09:15, 19 March 2007 (UTC)

What the lighting designers "know" is that if they put in halogen lighting, everyone except the comptroller will be happy, but if they put in CFLs, they'll get complaints from a small but very vocal proportion of the users of the facilities and they won't get any more business from that company or many of its employees. So they go with what won't get complaints and will get them more income in the future.

Atlant 12:36, 19 March 2007 (UTC)

I'm not sure what point you are making here. Why would the user of the facilities care what lights are used as long as they can see what they are doing? I B Wright 15:59, 19 March 2007 (UTC)

I had to visit another building this morning for a meeting. I just happened to notice that the security uplighters (provided to illuminate the buildings at night so that the security patrol can see what's going on) are fitted with what is obviously two different types of lamp. One is what most people expect CFL lamps to be, a nice warm almost incandescent glow (presumably the incandescent white refered to above), but the other a much colder white light (could be the 'daylight' refered to above, but I wouldn't actually know what daylight looks like). The important point is that the whiter light looks much brighter than the warmer light - and I mean that it looks very much brighter. I B Wright 15:57, 19 March 2007 (UTC)

Has anyone experimented with "dimmable" CFLs to see how they work? That is, in what technical ways are they different from "non-dimmable" CFLS?

My first guess is that they are not much different, and are only modified:

1. To lower the value of the bulk-storage capacitor after the rectifier so they will actually respond to the phase-control dimmer's waveform.
2. To allow them to keep the arc struck even at very low duty cycles.

On the other hand, I could imagine that they could be as complicated as using a microcontroller to read the duty cycle of the incoming mains power and adjust a high-frequency pulse width modulator to vary the light output in a way that accurately replicates the brightness "commanded" by the incoming duty cycle.

It's also been widely alleged that operating a non-dimmable CFL on a dimmer will cause fires. Aside from operating in the regime where the lamp is continuously extinguishing and re-striking, I can't see the mechanism where a dimmed but still normally-operating lamp would be an increased fire hazard. Does anyone have any clues for me? Increased

So, does anyone know? Or do I have to buy one for dissection?

Atlant 13:26, 16 March 2007 (UTC)

Dimmable fluorescent lamps (both compact and ordinary) can be dimmed using conventional quadrac dimmer units. The only difference is that the filaments must be permanently powered at full voltage, and the ballast must be designed to take a sinusoidal current at unity power factor. The dimmer unit does need to be modified to provide a full voltage feed for the filament circuit and has to have to be equiped with a load resistor to keep the qudrac firing stable when operating at low brightness.

Operating a non dimmable fluorescent on a dimmer causes the quadrac to fire erratically. Although I doubt very much that anything will actually catch fire, it could easily damage the semiconductor components in the dimmer (and probably won't do the lamp much good either). 81.157.133.142 17:19, 16 March 2007 (UTC)

Here's one answer to my question:

• http://www.irf.com/technical-info/refdesigns/cfl-3.pdf

And yes, the basic problem is that capacitor-input filters simply don't dim much until they suddenly reach extinction. Not only does the app note say this, but I tried it and as I "dimmed" them, my lamps dimmed slightly, reached a point where they briefly "half-cycled", then went out completely. And yes, the app note reminds us that there are also problems with the EMI suppression inductor "ringing" with the capacitance of the CFL's input stage.

Atlant 20:49, 16 March 2007 (UTC)

I removed the following:

"However, it is unclear what kind of pollution environmentalists are specifying when they make such statements. The issue of

global warming has prompted environmentalists to allege that carbon dioxide is the main culprit for the apparent rise in global temperatures over the past decade. This debate has not been settled, and there is substantial evidence that carbon dioxide is not the main cause of the greenhouse effect

. If true, then power stations using carbon-based fuels cannot be regarded as polluting the environment."

This is unsourced and POV. There is a separate article on global warming and arguments about it belong there. There are multiple sources of pollution associated with electricity production. There is nothing unclear about it.--agr 14:48, 19 March 2007 (UTC)

I don't think you can blithely talk about pollution without explaining what exactly you mean. You seem to have a POV which neglects wider issues. Please read the article on air polluton or at least cross-reference so that readers can check your broad and I think unjustified statement. Politicians and environmentalists are currently trying to make an issue here, and Wiki artcles should mention both sides of the argument to remain neutral. Wki s no place for a biased POV. Peterlewis 14:54, 19 March 2007 (UTC)

I've added a quote from Environment Canada to explain exactly what I mean. I've also linked to the pollution article, a good suggestion. If you have a reliable source that says electricity generation does not contribute to pollution, feel free to add it.--agr 15:34, 19 March 2007 (UTC)

<fx>Puts on pedantic hat.</fx> Errm.

• Wind power - little or no pollution
• Tidal barrages - little or no pollution, but has other environmental effects
• Hydroelectricity - little or no pollution while generating. Methane from flooded valleys may be significant, as are other environmental effects from disrupting the watershed.
• Nuclear - little or no pollution while generating, and little to some after, as strenuous efforts are made to keep nuclear waste out of the biosphere. Opinions on this are polarised and strong.
• Natural Gas - lots of CO2, relatively little other pollutants
• Oil - lots of CO2, some sulphur dioxide, depending on scrubbers
• Coal - lots of CO2 (considerably more than natural gas or oil), sulphur dioxide (depending on scrubbers), arsenic, selenium, heavy metals (including mercury and radioactive elements)

So generating electrical power per se is not necessarily polluting - it depends on how it is done. Obvious and extremely pedantic, you might say, but the sweeping statement that "generating electricity is polluting" is plain wrong, and needs qualifying. Certain methods of generating electrical power are polluting (for certain definitions of 'polluting'). Unqualified sweeping statements do not belong in an encyclopædia.

<fx>Doffs pedantic hat.</fx> WLD^talk|edits 15:53, 19 March 2007 (UTC)

Unless they are true. Zaslav 03:36, 20 March 2007 (UTC)

That wouldn't be a qualification of my sweeping statement would it? :-) WLD^talk|edits 08:14, 20 March 2007 (UTC)

There's been a huge battle over a wind farm (Cape Wind) off the coast of Massachusetts, with many environmental groups opposed (and some, to their credit supportive). At the least there is pollution associated with their manufacture and installation (see same argument wielded against CFLs above). There is also ongoing maintenance, introduction of traffic and waste into pristine areas, disposal of the equipment when worn out, effects on birds and micro-climates, etc. In any case, most electrical supply systems are gridded, so reduced demand allows the most polluting source to be kept off line longer. --agr 16:21, 19 March 2007 (UTC)

Wind farms are visually polluting of pristine mountain environments, and there is much argument in the UK about their introduction. The costs of installation and transmission are high, again largely because they need to be situated in remote areas. The wind is not reliable, so thermal power stations have to be installed in case the wind fails. This means that in the UK, they are subsidised by the consumer. Peterlewis 06:51, 20 March 2007 (UTC)