Alliance for Lighting Information

Some Inconvenient Truths about CFLs

by David M. Keith

Compact Fluorescent Lamps (CFLs) have been around for quite a while and have been recently promoted as the best alternative to the soon-to-be-illegal incandescent lamps typically used in residences. While CFLs can be very effective in some lighting applications, there are significant and important issues about using low-cost CFLs in residences. The CFL industry and advocates for sustainability have been ignoring or dismissng such concerns, citing the critical importance of reducing electricity use to thereby reduce pollutions and contributions to greenhouse gas emissions and global warming. However, now that the US government ban on incandescent lamps is approaching, the suitability of CFLs for residential applications deserves closer scrutiny.

The primary issues with CFLs are:
- starting current
- energy use and power factor
- rated life
- anticipated cost savings
- disposal/recycling of mercury

The electrical current used to start CFLs is higher than operating current - but this is mosly an issue for the designer of the wiring. In general, in residential applications, this will not matter at all. The difference between starting and operating current is relatively large, but either is rather low compared to the power used by an incandescent. However, the power used by any CFL depends on the power factor built into that CFL.

Power factor is the difference between a "completely resistive electrical load" and all other electrical loads. Examples of a "completely resistive load" are toaster, incandescent light, electrical heater - anything that converts electricity to heat through resistance. For these resistive devices, the current and voltage are "in phase" with each other - the peaks and valleys of the alternating current occur at *exactly* the same time.

All other loads - computers, fans, motors, clocks, HID ballasts, medical equipment, battery chargers, LEDs and CFLS - can create a phase difference between current and voltage. This means that the wattage (W) may NOT equal the volt-amps (a.k.a. VA) for these devices - which means that the way the electricity is measured (and billed) is not well matched to the energy actually being consumed. The lower the power factor (pf), the greater the discrepancy between W and VA. The technical relationship is: W = VA * pf, with pf equal to or less than 1.0. In the terminology of the industry, "high pf" is ~0.9 and "normal pf" can be around 0.4 or even less.

This means that a CFL rated 15W may be using 17 or 38 VA, for "high" (0.9) or "normal" (0.4) power factor values respectively. That's more than a 2:1 difference. At present, the electrical utility does not get to charge differently, either - even though a CFL with "normal" pf use twice as much electricity compared to a CFL with "high" pf for the same rated wattage and lumen output.

Of course either CFL is saving energy compared to the 40 to 60 Watts (= 40 to 60 VA) that would be used by a comparable incandescent. However, the "normal" pf CFL at 38 VA is only marginally different from a 40W incandescent at 40 VA, and nearer to 60 VA for an 60W incandescent than to the 17 VA for a "high pf" CFL.

So the "promised energy savings" from CFLs - assuming that most sold are "normal" pf - is half or less of what would be predicted based on comparing wattages with incandescent lamps.

Furthermore the "promised money savings" from using CFLs is questionable too. This is because the predictions are based on switching them on just once a day for at least 3 hours - the "3-hour burn cycle" that is an unstated assumption in rating the "life" of a CFL. That "life" is typically in the "thousands of hours" - 6,000 to 12,000 to even 20,000 - and those numbers are all based on that on-once-a-day-for-3-hours use pattern.

When fluorescent lamps are switched on and off more frequently, the corresponding "life" decreases significantly - and when they are switched on and then off within a few minutes, they fail very quickly. This is because the failure of CFLs is usually caused by the failure of its cathode - meaning that the lamps stop working because they can not switch on. This means that the operating period of fluorescent lamps in residential applications has very little to do with the use pattern that predicts "thousands of hours of use" and for such situations is more realistically predicted by the number of times the lamp is turned on *and* left on long enough to heat up completely.

The right way to predict the performance of CFLs in situations where they are frequently switched on and off is by the "number of cycles" - disregarding the hours that they might be left on in the meantime. That is to say, CFLs "life" should be rated as "number of on/off cycles" instead of some number of hours that depends entirely on how long they are left on. For example, instead of a rating of 6,000 hours, a CFL should be rated ~2,000 cycles (using the 3-hour cycle to estimate.)

For anyone interested in investigating this for themselves, there is a suggested residential CFL research task to demonstrate how well CFLs perform compared to incandescents in "frequently-switched applications" such as residences.

For example, if someone turned on a closet CFL every workday morning and evening, that CFL might last 2 to 4 years - even though it burned less than 100-1,000 hours total - because the cathodes would fail after approximately 2,000 cycles even if that cycle was only 1 to 10 minutes long. That would certainly not provide the cost savings that the CFL industry has been advertising - in fact it probably would turn out to be much more expensive to buy that CFL than one incandescent lamp that might keep burning for years, perhaps a decade, under the same pattern of use.

And then comes the last step - disposal. For an incandescent lamp, the waste is composed of glass, filament, copper or aluminium base socket - period. The filament is a tungsten wire, the glass comes from sand - no contaminants to speak of at all, and potentially completely recyclable. By comparison, CFLs contain mercury = a known bio-accumulative toxin, partly mixed into the phospors. The CFL usually includes its ballast, a mix of eletronics with heavy metals and contaminants - hardly as benign as an incandescent lamp.

Recycling is expensive and not included in the purchase price. From Colorado, lamps have to be transported hundreds of miles to a site that can recycle fluorescent lamps. The CFLs being collected large-scale retaillers in "take-back" programs are probably being transported using otherwise-empty-return trucks. The Earth911 list for recycling includes CFLs for drop-off but does not say where the actual recycling occurs. While state and local jurisdctions are listed as collectors, the nearest recyler to my 80305 zip code is probably one in Phoenix or Missouri (coming soon to Texas and California). Obviously, such transporting of CFLs for recycling reduces - or even overwhelms - any remaining "energy savings" compared to comparatively-benign incandescent lamps.

Recycling should be part of the initial lamp cost - and then the costs would not be hidden or ignored. As of the middle of 2011, there are disposal arrangements that cost $1-2 per CFL. Another program accepts mixed lamps at around $2.50 per pound. Recyclers are using new equipment that is much more effective in recycling the components of the lamps and in particular in handling the mercury involved - but the fate of the recycled mercury is not clear.

Finally, there are the environmental impacts from the mining, processing and transportation of the materials that are used in the manufacture of CFL lamps and disposable ballasts. For this discussion we will consider that such impacts are honestly represented by the cost of the lamp and/or ballast itself - even though in such situations the "externalized" costs apparently constitute significant pollutions that are not represented in the price of the object being considered.

So, depending on their use, CFL lamps may not be much of a bargain - and may be an environmental disaster on a very small scale, repeated millions of times.

Considering these issues, it seems better to put up solar panels or windmills to produce cleaner electricity to power incandescent lamps. Making incandescent lamps illegal just shifts the problem and misses the critical point - which is that we should be producing cleaner electricity to power devices of our own choosing.

This site is provided as a public service by resodance publishing co. and no guarantee is implied or provided.