Switches intended for use with lighting (timers, PIR, photocells) warn against use with CFLs, and state that filament lamps should be used. Why? It shouldn't make any difference whether a relay or triac (with ZCD circuitry) is used to switch the lamp on/off the lamp. I've used a couple of timers with a CFL for years without any apparent problem.
And if there is a problem I am unaware of, what will we be able to use when the filament lamps are taken off the market in the EU in a few years time?
-- Jeff (cut "thetape" to reply)
A CFL starts with a huge current pulse charging the main supply cap inside. Not all switches like that.
Maybe but I doubt it. Cold filament lamps start with a high inrush peak current too. For all I can imagine switching off a CFL with a classic ballast may cause problems. Switching on/off modern "energy saving" CFLs should not do worse then filament bulbs. Dimmers are different and I think the warning not to use CFLs in timers and the like comes from that dimmers. The manufacturers simply stay on the safe side to avoid claims. After all the behavior of filament bulbs is well known but they can't be sure of the electronics that are put in CFLs. As for abandonning filament bulbs, I expect LED lights to push aside both CFLs and filament lamps in next years.
petrus bitbyter
Cold filaments draw peak current something like 20 times their RMS current, depending on when in an AC half-cycle they are switched on. CFLs often draw much more than that, with peak current limited by line impedance and the impedance of the bridge rectifier, filter capacitor, and a small RF choke. Sounds to me like an ohm or two.
Probably a good idea for consumers also. I would not like to have to explain to a fire insurance company why I was using electrical equipment other than as directed, even if the fire started for a reason other than the fact that the equipment was being used in a way that was warned against.
- Don Klipstein ( snipped-for-privacy@misty.com)
There is also the observed fact that for many CFLs, any small leakage current through or across the switch will cause the CFL's reservoir capacitor to charge until there is enough voltage for the inverter circuit to start, thus the lamp tends to flash. I have seen this happen with just the capacitance between two parallel wires (in the same cable) in stairway lighting circuits. I wonder what effect on the safety and reliability of the lamp many thousands of repeated starts a day has?
Failure modes for a normal lightbulb tend to be short (often with a momentary arc) - usually protected by an integral fuse, open, and envelope rupture. In a normal operating environment, none are particularly hazardous. Perhaps CFLs have a failure mode that involves dangerous overheating on startup so require a human present to check they have turned on safely?
Some interesting comments.
I'm not sure about the peak current taken by a CFL at switch-on, but according to the circuits at
Out of interest once I connected a CFL up to a standard light dimmer. Waste of time - the CFL was basically on or off, and flickered badly at any intermediate setting.
-- Jeff
Mine made a gentle pop, and ceased to function thereafter. Can't remember what type of dimmer switch it was.
-- http://www.petersparrots.com http://www.insanevideoclips.com http://www.petersphotos.com A construction worker goes to the doctor and says, "Doc, I'm constipated." The doctor examines him for a minute and then says, "Lean over the table." The construction worker leans over the table, the doctor whacks him on the ass with a baseball bat, and then sends him into the bathroom. He comes out a few minutes later and says, "Doc, I feel great. What should I do?" The doctor says, "Stop wiping with cement bags."
what type of dimmer switch it was.
There is/was some good info on flouro lights,and CFL's here:
I am finding some flawed measurements, such as in:
That page claims that an LED tube light produces more than twice as much light per watt than fluorescents that it replaces, on basis of power consumption and lux on a benchtop under the light. However, the LED tube light is probably illuminating a smaller area than the fluorescents did.
If you find benefit from making a light source directional, there is such a thing as reflectors. With a suitable reflector, a decent fluorescent will have the same directional pattern as the LED tube light and provide similar (more likely greater) lux per watt than the LED tube light.
I am also finding some flawed data, as in:
That page has a table showing supposed overall luminous efficacies of various light sources.
It shows 17.5 lumens/watt for 100 watt incandescent - that is top end for 100W non-halogen 120V incandescent. And top end 230V 100W ones do not achieve that, due to economies of scale on the thinner filament.
A few paragraphs later that page says a 100 watt incandescent produces
1800 lumens but clams CFL manufacturers are claiming 1246 lumens as equivalent to 100W. In what country? Many European and Australian 100W incandescents do only produce 1200-1400 lumens. 100W incandescents that get referred to as GLS lamps (a term used on that page and where household voltage is 230V) do not produce 1750-1800 lumens, and even USA ones don't produce 1800 lumens.Same story for the 40 watt incandescent.
It also shows 93 lumens/watt as "max, typical" for 36 watt T8 fluorescent. Keep in mind a couple things:
The same table says 24 lumens/watt for halogen. Most rated at least
2,000 hours and drawing 2.5 amps or less achieve 21 or less.That page also mentions some fixtures to not use any CFLs in. However, some CFLs, such as Philips "triple arch" 15, 20 and 23 watt non-dimmable, are specifically rated to be OK in recessed ceiling fixtures.
That page claims that CFLs emit significantly more UV than incandescents. I have found that to hardly be true. And that page claims that the UV increases risk of *breast* cancer?
That page also mentions problems from CFLs taking 5 minutes to warm up. Ones without outer bulbs generally take more like 1 minute to be most of the way warmed up.
I sense anti-fluorescent bias.
- Don Klipstein ( snipped-for-privacy@misty.com)
Or just confusion. There are far too many ways to measure light.
LEDs WILL take over our lighting. They may need to get more efficient first though. They already have advantages in things like traffic light5s and cars, as they don't need changing so often, and are instant on/off.
-- http://www.petersparrots.com http://www.insanevideoclips.com http://www.petersphotos.com Never dive into deep concrete.
Peter Hucker wrote: (snip)
I am sure that they will. But I'm not so sure about their predicted life (compare what is still being said about CFLs with real-life experience). The high-power LEDs do run really hot, and need a good-sized heatsink to deal with the waste heat. No good adding fans to cool them, as they aren't reliable enough for really long-term usage.
Of course it may be that semiconductor LEDs will be superseded by OLEDs. Not for every use, perhaps, but where one needs diffuse lighting they will be OK. I can foresee the time when we don't have luminaires as such - the whole ceiling (and perhaps walls) will be a light emitter.
-- Jeff
When they become more efficient, there will be less heat output.
What about a heat pipe like thy use in computers?
From wikipedia: "The biggest technical problem for OLEDs is the limited lifetime of the organic materials. In particular, blue OLEDs historically have had a lifetime of around 14,000 hours"
-- http://www.petersparrots.com http://www.insanevideoclips.com http://www.petersphotos.com What do you call a dwarf who throws the discus? A compact disc player.
Too expensive.
But see
True, the life isn't that good yet (although 5000+ hours isn't bad), but look at the lm/W figure, and what is projected for next March. ISTM that OLEDs are in the stage of development that LEDs were in a couple of years ago, but are catching up fast.
-- Jeff
Are you sure? They're included on some quite cheap graphics cards.
The lm/W next March is no better than normal LEDs and CFLs.
-- http://www.petersparrots.com http://www.insanevideoclips.com http://www.petersphotos.com REALITY.EXE corrupt. Reboot universe (Y/N)?
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.