ESP Logo
 Elliott Sound Products CFLs 

Rod Elliott (Elliott Sound Products)
CFLs - Not To Everyone's Liking, But Still A Good Option
Page Created and Copyright © 08 January 2013

Share |

IndexLamps & Energy Index
IndexMain Index


Compact fluorescent lamps (CFLs) have now been mainstream for several years, and despite some problems (including a few intractable issues) they are still a good option if used properly and sensibly. Unfortunately, 'proper' and 'sensible' can be very difficult - especially if you live in a rented property. Incandescent lamps were essentially banned in Australia as of 10 Feb 2009 (when GLS [general lighting service] incandescent lamp imports were stopped). MEPS (minimum energy performance standards) were applied at point of sale for all lamps in November 2009, and this effectively made it an offence to sell 'inefficient' lighting products.

All lights, with the exception of a few specialised types for ovens, be they gas, electric or microwave, must meet a mandatory minimum luminous efficacy standard, and standard light bulbs are unable to meet the MEPS requirements so cannot be imported or sold. There are some halogen lamps that just (just!) meet the standards as they are at present, but these, too, may shortly become non-compliant and will be banned. This includes the traditional 50W/12V downlight. A review of existing and currently compliant lighting products is to be held in October 2013 [2].

Meanwhile, CFLs have matured to an extent, and some are even dimmable - well, that's what it says on the packaging, but reality is cruel . Yes, they can be dimmed, but the results are often less than ideal. Despite claims that both leading and trailing-edge dimmers can be used equally well (see Dimmers for descriptions), in reality only trailing-edge or 'universal' dimmers should be used to avoid large current spikes that can damage both the lamp and the dimmer (see below for more on this topic).

While I refer to CFLs and LED lamps as separate entities, they actually share many characteristics. Both use electronic power supplies, and neither is suitable for use where the temperature is likely to exceed perhaps 60°C or so. This is not the temperature in the room, but the temperature inside the luminaire adjacent to the lamp itself. Indeed, at this temperature, the electronics will be a great deal hotter, and it's extremely important to keep the immediate ambient temperature as close to room temperature as possible.

These issues need to be addressed, but there is little useful information available anywhere. Certainly, the packaging rarely gives the user any usable guidelines, and even if they did, few people read the instructions for light bulbs - they never had to do so before, and see no good reason to start now. This is made worse because manufacturers, 'green' websites and resellers make no specific recommendations nor do they provide useful advice. They mostly just say things like "This 20W lamp gives the same light as a 100W incandescent, but uses 1/5th the power." To the buyer, they are equivalent products and can be used the same way. WRONG !

Claim & Counter-Claim

There are many sites on the Net that discuss CFLs and other 'alternative' lighting products. Some are balanced and explain the strengths and weaknesses of each, but there are others that are pure scare-mongers. According to these sites, using a CFL will almost certainly cause you to become ill, and if you are already ill then you are sure to die (or worse). Very few of these sites will give good reasons or (perish the thought) a proper scientifically reproducible measurement, but will regale you with tales of 'dirty electricity' and ultraviolet light (in vast quantities some claim).

You'll be told that your public liability insurance won't cover you for the damage these lamps do to your visitors (yes, I am serious - I've seen the claim made) and that they are dangerous and inherently unsafe products. There are also complaints about flicker, even though a CFL probably has less flicker than any other traditional light source. The power supply runs at over 30kHz - 600 times faster than a normal fluorescent tube operating from 50Hz mains. If a CFL shows visible flicker, it's faulty and should be replaced. It certainly doesn't mean that they all do it and are therefore dangerous.

Like all fluorescent lamps, CFLs use mercury vapour inside the tube to generate ultraviolet light. This is converted by the phosphor coating on the inside of the tube - when it's excited by UV light, the phosphors emit visible light. That's the way all such lamps work. Yes, there is some UV light that escapes the tube, and there is no doubt that some people are hyper-sensitive to UV. Lupus sufferers can be badly affected, but this applies to all light sources that emit UV - including quartz-halogen incandescent lamps! Other people can be affected too, and in general it's not sensible to be too close to any bright light, regardless of the technology.

Some of the scare campaigns you find on the Net are unbelievable, and while they often quote real research by qualified professionals, it's often taken out of context and/or a minor point is exaggerated to the point where it looks very scary indeed. There are risks, and I've covered most of them in great detail, but for most people in normal houses the claimed risks are fairly small. The real risks are usually overlooked because the authors of scare campaigns usually don't understand the technology used anyway.

TV 'news' items (usually only partially) citing research and naturally talking to concerned parents help promote FUD (fear, uncertainty and doubt), but rarely provide the full story. Don't expect to see a comparative UV light test performed on various different light sources - in fact, don't expect to see any test figures at all. Never ruin a good story with facts seems to be the motto.

What Are The Risks?

Personally, I far prefer LED lighting to CFLs, but there are some places where the expense of LEDs is not justified, or no suitable LED lamp exists at an affordable price. As explained in the first article I wrote on the topic (Incandescent Lamps), CFLs should be considered an interim solution, but in the meantime, it's only reasonable that local councils provide proper recycling facilities to prevent mercury contamination. Thus far, I don't know of any local government area that has done so, and that's highly irresponsible. That to me is a major risk, and there is little or nothing being done about it.

What's even worse is that very few light fittings (aka luminaires) are suitable for use with CFL or LED lights, because they lack any real flow-through ventilation. Unlike incandescent lamps which get hot because it's what they do, the electronic circuits used in CFLs and LED lights cannot tolerate heat. If the internal circuits are subjected to more than around 60°C their life is shortened dramatically, and at 100°C or more failure will occur in a matter of days or even hours!

CFL failure modes vary, but it's not at all uncommon for the electronics to fail with a flourish, which can include burnt plastic and glue, or even explosion. These are real, and are well documented (including in my articles on the topic), and many such failures are the direct result of the CFL being installed in an inappropriate fitting that doesn't have proper ventilation.

Despite complaints worldwide about the lack of suitable luminaires for CFL and LED lights dating back several years, there appears to have been almost zero progress. Fully sealed oyster fittings are still common, as are suspended ball fittings in various sizes. When any electronic lamp is used without adequate ventilation, it will run far hotter than ever intended, and the life can be shortened to as little as a few days. Some lamps will survive while others fail - statistical analysis could be applied if there were any useful figures available but this isn't the case.

Figure 1
Figure 1 - Suitable And Unsuitable Fittings For Electronic Lighting

It's about time that lighting designers started to take notice of electronic lighting, because it's here to stay whether anyone likes it or not. It doesn't matter if the lamp is a CFL or LED - neither can tolerate heat, and both will have dramatically reduced life if they are run in unventilated fittings that cause the electronics to overheat. This point cannot be stressed too much - it is essential that any electronic lamp operates at the lowest possible temperature.

If a CFL (in particular) is installed in a sealed lamp housing, there is every reason to expect not just failure, but spectacular failure. This can include the CFL exploding, and possibly damaging the housing in the process. See 'Normal' Failures for some examples including photos of failed CFLs. This is not intended to alarm users, it's just factual information that you need to be aware of.

Utter Nonsense At Large

One of the most troubling aspects of CFL 'information' is just how poor and misleading it can be. As an example, I offer the following Q&A, taken from South Australian TAFE - Energy Efficient Lighting. This is typical of the misleading claims seen, but to find it in an educational website is an indication of how little regard most people have for the fact that CFLs are electronic lighting products, and simply cannot be used as an equivalent to GLS (general lighting service) light bulbs.

When one reads the blurb to see where this drivel came from - it's from the South Australian Government (Department for Transport, Energy and Infrastructure)! How on earth are people ever going to be able to make informed decisions when their government provides completely inappropriate 'advice' like this ...

Q     Can I put a compact fluorescent lamp into my existing light fitting?

AYes, provided that the fitting is not controlled by an analogue dimmer switch*. Compact fluorescent lamps come with bayonet or Edison screw bases, so they can be used instead of incandescent light globes in existing light fittings. Note that the higher wattage compact fluorescent light bulbs are often larger than equivalent incandescent globes, so this may also be a factor. You may find that some lamps have a restricted light globe wattage, meaning a Compact fluorescent lamp can give you a brighter light safely.

The above is simply wrong!

*   I am unsure why the author chose the term "analogue dimmer switch", as it is not common terminology in Australia (the term 'dimmer switch' is very common in the US though). A dimmer is not a 'switch' - it's most commonly a rotary control (although home automation systems may use up/down buttons). It doesn't matter a damn if the controller is analogue or digital - dimmers are a no-no for any non-dimmable CFL - period! "Analogue dimmer switch" indeed .

However, of far greater concern is the claim that you can use a CFL in the existing light fitting, but without a single cautionary word about ventilation. This is exactly the kind of disinformation that has spread throughout the Net (as well as the popular press, TV, 'green' websites, etc.). It might be possible to use a CFL in the "existing light fitting", but only if it meets the ventilation criteria shown above. The implication is that if the fitting is rated for a 60W GLS bulb, you can use a 20W CFL (supposedly equivalent to a 100W incandescent lamp) without a care in the world. Apparently, who cares if the CFL gets so hot that the electronics fail? Not the government, that's for certain.

Unfortunately, it is just this kind of nonsense that you'll find on government websites everywhere, and it's almost impossible to get to anyone who is willing or able to make changes to the disinformation. I have tried on several occasions, but to absolutely no avail - no-one wants to know about reality if it conflicts with 'government policy'.

Regardless of government (or other) drivel to the contrary, a CFL is not 'just a light bulb'. It is an entirely new product that operates in a very different manner from the bulb it allegedly replaces. There are certainly instances where there is no likelihood of problems, but they are actually in the minority. More than 50% of existing (and new) light fittings are unsuitable for use with CFLs or even LED lights, because they are either inadequately ventilated or have no ventilation at all.

Don't expect your government or news media to tell you, but even ignoring the electronics, if CFLs get too hot their luminous efficacy falls dramatically. Getting exact figures is not easy because it's not something that seems to be widely publicised. Depending on the type of lamp, the housing (luminaire) and ventilation strategy (if any), you could expect a light output reduction of as much as 25% [4].

In general, try to keep the lamp's ambient temperature (i.e. within the housing containing the CFL) to no more than about 45°C. The electronics will be happier with lower temperatures, and if you can keep the lamp's ambient temperature to 25°C long term survival is probable (other than normal component failure that can happen at any time).

If you care to do some research of your own on the temperature effects vs. light output (as well as base up vs. base down or horizontal mounting), you'll discover that it is a minefield of conflicting requirements. The fluorescent tubes like to be hotter than the electronics for maximum light output, but only for one group of lamps. Another group may show a dramatic drop in output if used base-down, unless they use a mercury amalgam. The conflicts are wide and varied, and are only determined by rigorous testing on a wide range of lamps and fittings.

For domestic CFLs, it looks like no-one has done much research for you, and I doubt that anyone will. You might find some info, but it won't be for the CFL you can buy at the supermarket unless you are very fortunate indeed. In general, you need to maintain an ambient temperature (for the lamp, not yourself!) of between 10°C and 45°C. Lower temperatures will cause low light levels and/or difficult starting, and anything higher will both shorten the life of the electronics and reduce light output [5].


As described in the article Incandescent Lamps [1], all CFLs have a significant amount of electronics inside, and the power supply doesn't like the very fast risetime created by a leading-edge (aka TRIAC) dimmer. This limitation is not limited to CFLs though - LED lamps are no happier when used with leading-edge dimmers, because they also have electronic power supplies. The simple fact is that despite manufacturers' claims to the contrary no lamp with an internal electronic power supply should be used with a leading-edge dimmer, with the possible exception of 12V lamps operated via an electronic transformer. Even this cannot be guaranteed!

Dimming with CFLs is generally not very successful, and as noted above the makers always claim that both leading and trailing edge dimmers are suitable. IMO, you should never use a leading-edge dimmer because they cause excessive current spikes. Trailing-edge or universal dimmers are far more friendly to the electronics, and are the only ones that should be used. Look at Figure 2, which shows a leading-edge dimmer (left) and a trailing-edge dimmer (right). I tested a leading brand dimmable 20W CFL for these measurements, with the brightness set for approximately half. Both dimmers were set for the same light level.

Figure 2
Figure 2 - Leading-Edge and Trailing-Edge Dimmer Performance

The leading-edge dimmer causes ±2.4A spikes, with the RMS current measured at 165mA, while the trailing-edge dimmer had current spikes of only 400mA (1/6 the peak current with a leading-edge dimmer), with an RMS current of just under 92mA. This is a significant difference, and shows quite clearly that the trailing-edge dimmer is much kinder to the internal electronics. It wouldn't take many of these lamps to exceed the peak current rating of a typical cheap leading-edge dimmer, most of which use a TRIAC (a type of electronic switching device) that's only rated for perhaps 10A repetitive current peaks.

Although not shown, I also tested the same CFL without a dimmer, and the maximum RMS current measured 126mA with 450mA peaks. When this is compared against the dimmed current, it's quite obvious that the trailing-edge dimmer is a great deal less hostile and will minimise component stress.

So, although most manufacturers claim that leading-edge dimmers are suitable, in reality this is dubious at best. With any dimmable electronic lamp, assume that you should only use a trailing-edge or universal dimmer - regardless of any claims to the contrary by the maker or anyone else. The current waveform measurements that I show are usually not shown or even performed by anyone else! Without this important measurement you are in the dark (possibly quite literally).

Please Note: You should never operate a non-dimmable CFL or LED lamp with a dimmer in circuit, even if it is set to maximum! This is very dangerous, and causes very high peak current that can easily cause the electronics to overheat and either fail quietly, or fail with a flourish - including but not limited to fire!

I tested a standard 11W CFL, and the normal current drawn was 70mA RMS, with a measured peak current of 250mA. When used with a trailing-edge dimmer the light output was low, and the lamp only drew power during negative half-cycles of the mains. This is bad, but when I substituted a leading-edge dimmer, the peak current rose to 15A with very sharp peaks lasting only 25µs (25 millionths of a second). That's 60 times higher peak current than normal - I don't know about anyone else, but to me that indicates a very serious problem indeed.

I was so surprised at the exceptionally high current that I re-measured the CFL several times to make sure I hadn't made a mistake. I knew that the current would be many times greater than normal, but when I saw 15A peaks I was taken aback. Several measurements later, and the measured value was confirmed. I'm still surprised! Take heed of the warning above - this is a very real problem.

Dirty Electricity

This has become one of the most over-used and least understood terms imaginable. If you ask those who proclaim (loudly) that 'dirty electricity' is dangerous and will cause all manner of ailments, it's highly unlikely that any of them can actually explain what specifically is 'dirty' and how it can cause harm. Many use completely mysterious boxes that measure the 'dirt', but none that I've seen displays this measurement in any meaningful way. Even one of the available meters displays its results in 'GS units', because no officially recognised standard terminology is available. That alone should tell you something . Many of the sites that purport to tell you all about 'dirty electricity' are riddled with factual errors, but in some cases the comments left by others can be amusing, so all is not lost.

If you look at the waveforms shown in Figure 2 again (especially the one on the left), that is an example of so-called 'dirty electricity' - at least as I understand it. The spike waveform indicates that there are many harmonics of the mains frequency, so additional frequencies are created at odd multiples of the 50Hz (or 60Hz) mains. While the additional frequencies are certainly higher than the mains, there is no evidence that any of these frequencies are in any way harmful. The voltage and current in the speaker leads to your hi-fi speakers can carry signals that are not only much greater than those shown, but extend to higher frequencies and contain more frequencies - not just those that are multiples of the mains frequency.

No-one I've heard of has ever suggested that the audio signals in your speaker cables or speakers are harmful. Provided you don't make direct contact with the wires, there is no claim of risk that I've ever found. Even if you do make direct contact, even that is usually harmless unless you have a very powerful amplifier.

All electronic power supplies create harmonics to a greater or lesser degree, including those used to charge mobile phones, run your portable or desktop computer or power any number of low voltage devices. I don't see the anti-CFL brigade waving their silly meters at those, and I know why - most will register much higher electrical fields than a CFL or LED lamp, mainly because they are far more powerful. However, the 'dirty electricity' brigade have a field day, even if they do get many of their 'facts' woefully wrong.

I do not deny that there may be some individuals who are more sensitive to electronic fields than the rest of us, but unwarranted and unsubstantiated scare-mongering does no-one any favours. Anyone who is sensitive to CFLs or LED lights will be unable to use a computer because the electric fields will be far stronger, but this seems to be an area where there is intense disagreement. One quote that sums up the topic says it all - "In 30 years, 25,000 studies have failed to find a definitive link between adverse health effects and 'dirty electricity'." [3]

I remain open minded on the issue. My natural inclination is to dismiss the claims out-of-hand, but for every site that refutes any notion of harm, there are several that provide evidence to the contrary. What I don't know is whether they are providing real evidence, or 'evidence' that's only backed up by their own research (another word that has a very different meaning when put in quotes). No matter what you happen to believe, there is a website somewhere that agrees, but that doesn't make it true. It's also worth noting that there seem to be more sites promoting the concept of 'dirty electricity' than there are debunking the idea - again, the number of websites does not indicate the truth or otherwise of any claims made.

I know from experience that there are many people who will cheerfully supply 'products' to detect and/or fix any known (and even many unknown) problems, regardless of type, cause or effect, and regardless of the efficacy of the products offered. There are countless examples, and so far I'm not at all convinced that any meter of filter for 'dirty electricity' has any basis in science or reality. One site I looked at even recommends that users change CFLs to 'candescent' (sic) or LED lamps - not realising that every LED lamp also includes an electronic switchmode (high frequency) power supply. Exit all credibility from that site, despite the 'doctor' who is referenced throughout. Just because someone has a PhD doesn't necessarily mean they are smart!

CFL Benefits And Disadvantages

Compact fluorescent lamps (like any technology) have pros and cons, and it's important to recognise the points that are important against those that are not. There is no doubt whatsoever that they are dramatically more efficient than incandescent lamps, providing as much or more light at a fraction of the power consumption. As power prices increase this is obviously a benefit, but lighting in general isn't a massive part of the household electricity consumption. Estimates vary, but assume around 10-15% of your electricity bill is due to lighting - assuming the use of incandescent lamps. If you switch to CFLs, this percentage will drop, but it won't be substantial in most houses. Don't expect to save $hundreds every bill unless you run thousands of lights.

Overall, the savings to the nation are enormous, but the electricity network doesn't benefit as much as many websites would have you believe. The current waveforms shown above and the measurements taken are proof of this. A 20W lamp should draw 87mA from 230V mains, but the CFL I tested measured 126mA - substantially more. Even though the kWh meter in your meter box only registers 20W, the lamp actually draws almost 30 volt-amps (VA). You aren't charged for the extra current, but the utility has to provide it, and also make sure the infrastructure can handle the nasty current waveform. Better power supplies would help, but that's not going to happen with a cheap consumable product.

Compared to a roughly equivalent 100W incandescent lamp (which would draw 438mA), both you and the network are better off because there is much less current and power, so there is a net saving that's certainly worthwhile. Over an entire city (or country), the savings become enormous when high efficiency lights are used. Each individual lamp is a tiny drop in the ocean, but put together the benefits cannot be denied.

On the downside, CFLs almost never last as long as claimed. When they were first introduced in quantity in Australia, it wasn't uncommon to see claims of 20,000 hours life, but this has been revised down several times since. The 20W dimmable CFLs I have claim 6,000 hours! Another pack I have in the workshop claims "average life 8,000 hours", but then shows a 'typical application' in a fitting with no flow-through ventilation! Many people have found that CFLs don't last very long at all, but that may well be because the housing isn't ventilated and the electronics cook themselves to death. Like any electronic product, failure can happen at any time. To expect the very long life originally claimed is unrealistic though, but they will normally last longer than a typical incandescent lamp. One would hope so, because they are substantially more expensive to make, ship and buy than the very simple bulb that they replace.

Then we have the issue of disposal, which has not been addressed properly in the majority of countries where CFLs are sold in large numbers. It's not just the problem of mercury, but there's a whole printed circuit board of electronic parts that's also just thrown away, as well as the glass tube and plastic case. All of this material should be recycled, but it's not happening anywhere - at least not to the scale that it should. If (and that's a very big 'if' as most users are now aware) the CFLs lasted as long as claimed it may not be such a dreadful waste of resources, but the average life is usually well short of what's claimed. There are proper recycling facilities in some locations, but they should be everywhere, not just a few states in the US or small localities in Europe. Australia? - a couple of schemes that are mainly aimed at commercial users, but the rest of us seem to be out of luck for the most part. This is shameful!

Finally, there's the issue of CFLs and very low temperatures. This isn't a problem in most parts of Australia, but in many countries people have found that their CFLs either won't start at all, or are so dim for the first 5-10 minutes that they're unusable as a light source. CFLs do not like low temperatures, and like all fluorescent lamps, the tube itself needs to be at around 35°C or more before light output reaches its claimed level. That can be very hard if the outside temperature is -20°C or less.


This short article is intended to dispel some of the nonsense that surrounds electronic lighting in general, but CFLs in particular. Every type of artificial light source we use has advantages and disadvantages, and it's unreasonable to claim that CFLs are an unmitigated disaster compared to other technologies. Of course there are problems with them, just as there are problems with incandescent lamps and LEDs. However, it is fair to say that CFLs have more issues than competing technologies - as I have stated before, I consider them to be an interim solution to our lighting needs.

It's important to understand that unlike the incandescent bulb we got used to over the last 100 years or so, the term 'CFL' actually describes a class of product. Different brands don't necessarily perform the same, and even different power ratings by the same maker can be quite different from each other in several areas. The internal circuitry can change quite radically between different models, and so can the phosphor formulations and type of mercury (amalgam vs. 'free' mercury for example). Every change affects the way the lamp works, the ideal operating temperature and/or how fast it comes up to full brightness.

With incandescent lamps, the term 'GLS' (general lighting service) meant just that. You chose the lamp's power rating, and everything else was simply defined by very basic physics. With CFLs (and LED lamps as well), there is no such thing as GLS - you can't simply select the power rating you need and install it in any old light fitting that you come across. There are so many variables and there is a completely new requirement that you need to ensure there is adequate ventilation. Add to this that you need to be aware of dimmers and now have to worry about the particular type of dimmer as well. Anyone who claims that CFLs are a direct equivalent to pre-existing incandescent lamps is seriously misguided.

Many of the issues that plague users can be attributed to poor instructions and incorrect installation, in particular not providing adequate ventilation or assuming that a dimmer will be ok if set to maximum (it's not !). Some of the claims made on the Net have to be considered half-baked at best, while others are (probably) nothing more than scare-mongering for reasons that perplex me. It's easy to see that at least a few of the scare campaigns are aimed at selling you something you don't need and that probably doesn't work, but others are crying foul for no apparent reason. Mercury is a real problem, but some of the statements you read are just plain silly. As for a claim I saw recently that the "mercury will be phased out", this is complete nonsense! By definition, fluorescent lamps of all kinds are mercury discharge lamps and if you take the mercury away they don't work.

For the most part, I don't respect or give any credence to most of the 'green' websites, not because I think saving energy isn't a good idea, but because the majority completely fail to provide all the information that is needed, such as CFL suitable luminaires, provision for ventilation, etc. Many government websites are no better! Outlandish claims are common, but few even consider all the material that's thrown away when a CFL fails, and seem to take manufacturers' claims as gospel, with nary a test ever carried out. I do perform tests, and am more than happy to publish results which can be found throughout the Lamps & Energy section of my website.

Like anything else, how to light your home is a compromise. The best way to ensure that you select the best compromise for your situation is to be well informed, and know as much about the topic as possible. A glowing reference here, a damning report there, or a paper issued by a government agency that fails to consider all the variables is not balance, and doesn't help anyone. Sometimes it's hard to find good, objective information that gives the whole story, and that's one of the things I've been doing for some time.

I won't always get everything right, but if errors are found they are fixed as soon as I'm made aware of them.

  1. Incandescent Lamps - ESP Article (very long!)
  2. Aust. Government - Details of the phase-out
  3. The Last Word On Nothing - dirty, dirty electricity
  4. CFL Light Output In Different Positions
  5. Performance of CFLs at Different Ambient Temperatures


ArticlesLamps & Energy Index
IndexMain Index

Copyright Notice. This material, including but not limited to all text and diagrams, is the intellectual property of Rod Elliott, and is Copyright © 2013. Reproduction or re-publication by any means whatsoever, whether electronic, mechanical or electro-mechanical, is strictly prohibited under International Copyright laws. The author / editor (Rod Elliott) grants the reader the right to use this information for personal use only, and further allows that one (1) copy may be made for reference. Commercial use in whole or in part is prohibited without express written authorisation from Rod Elliott.
Page created and copyright © 08 Jan 2013./ 27 Jan 13 - added 'utter nonsense at large' section.