Compact Fluorescent Light is One Crispy Critter

EDN’s PowerSource Blogger Margery Connor has complained about her frustrations with premature failure of CFLs (compact fluorescent lights) in down lights at her house. To prove her point, in a recent post titled “Compact fluorescent lights: Not always the best solution” she offered this photo of the burned-out innards of a failed CFL. (Note in particular the brown scoring on the plastic base.)

To also prove her point, she shot the temperature of a downlight CFL in her home with an IR gun and got a reading of 160F at the base, while the table lamp CFL was at 120F — a temperature difference that those finicky little transistors in particular might find a tad unbearable. 

Related Tech News


  1. Luis
    January 26, 2009 02:31

    I have a really weird problem… I just installed a new light fixture with 4 lights. When I put in 4 X 13W CFL lights, it trips the circuit braker… When I use incandecent lights, it works like a charm so I know I don’t have a short in the wiring. When I run two CFL’s and 2 incandecents, it also works… The only explanaton I can think of is the it is the CFL’s that are causing the short… Theoretically 4 X 13 W is only 52watts and these things are rated at only 200 mAmps… Have any of you run into this problem? Help!!!
    Thanks. LCC

  2. Peter Willis
    November 07, 2008 12:27

    I use the 2D style luminaries, 16w and 28w (not yet 38w), and convert bowl lamp ceiling lights with starter gear,from tungsten. I write the date on them and have years of life. However I find that the circular 22w fluoescent tubes are very short lived.

  3. D Valentine
    November 06, 2008 11:35

    Something to bear in mind. What about electronics failures caused by supply line transients? Some years ago I attended a presentation by Texas Instruments where semiconductors for this type of application were being introduced. They thought that 400V devices were fine on a UK 240V (340V pk) supply without additional protection. Not much margin for transients! If reputable companies like TI think a narrow margin is acceptable, what about the 10p manufacturers? For our designs, we always use 600V+ rated devices on mains supplies. If higher voltage devices are available, we spend the extra pennies to include them. Also we always spend the money on transient suppression components. But we have the luxury of making equipment where reliabiity is more important than saving a few pennies on the selling price.

  4. November 06, 2008 09:24

    Bought 45 of these compact type mini spiral from B&Q to replace candle bulbs in light fittings in a local club. They were all mounted at a 45 degree downward sloping angle. According to the box were supposed to have either a 5 or 7 year life (can’t remember which). On taking them out of the box,two were found to be not working when fitted. In the first month approx one per week failed and during the second month one every two weeks on average. Not sure what the long term failure rate will be, but it was a good job I kept the receipt, as so far they are changing them without question.
    I should add the old clear tungsten candle bulbs were failing at the rate of approx two per week continuously, so far it appears to be an improvement, but not as good as I had hoped.
    A plus point is they are producing more light (equiv 60 watt,old candle bulbs 40 watt) and use just over 75% less electricity.The latest generation appear to be a lot quicker in warming up to full brilliance than the older ones used to as well.

  5. John Goldsmith
    November 06, 2008 08:27

    Would Dr. Bose please explain why the harmonics in the waveform cause blackening of the tube ends? I look after flats with dozens of compact fluorescents that operate with conventional starter and ballast (ie 50Hz sinewave) and at failure they all have blackened ends. Surely it is due to the heater filament used for starting evaporating onto the glass?
    As for bulb life, my mother gets bombarded with so many free bulbs by her energy suppliers etc. that she cannot use them fast enough. I have no use for them – in winter at least – as they save no energy at all in my electrically heated flat: they just push more of the load onto the heating, while absorbing more of the earth’s resources (and energy) in manufacture.

  6. Dr. Suvendu Nath Bose
    November 06, 2008 05:52

    The temperature is not the only problem – I guess. The BLACKening at any end of the used CFL is an important feature also! This can happen because of the rich content of higher harmonics in the waveform generated by the power converter. I feel that CFL driven by high frequency sinusoidal waveform [without harmonics] should enhance the life of the CFL. Thus the design of the circuit used in the power converter is most important!

  7. colin burn
    November 05, 2008 21:41

    Thes bulbs are currently available for as little as 10p so there’s not much margin for quality control….
    I reckon they are good for only about 18 months. I have 2 bulbs from 3, fitted about 18 months ago which are now intermittently failing to turn on.
    Thee were some remarks about mecury content causing a disposability issue. Anything further on this?

  8. November 05, 2008 17:34

    We have had the same experience and come to the same conclusion. Some CFL’s have had a shorter life than conventional bulbs. We don’t use CFL’s with the electronics uppermost except in one outdoor metal/glass fitting where the lamp runs cooler.
    The problem seems common to many manufacturers but especially those offering cheaper products.

  9. Tom
    November 05, 2008 17:31

    160F is only 71C which is not *that* bad although maybe its hotter inside. I’d guess the electrolytic capacitor would be more of a reliability problem than the transistors/ICs.

  10. Graham
    November 05, 2008 17:28

    Just to add I have had several CFLs fail in the upright position so orientation maybe one issue but it is not the only reason for premature CFL failure.

  11. Ian Cooper
    November 05, 2008 17:24

    I agree with Margery. I’ve been using energy saving bulbs in my home for ten years or so and none of them has lasted anywhere near the boasted life of 8~10 years. They might save electricity but their life span doesn’t seem to be much longer than an old filament bulb. Quality control also seems to be a problem, several bulbs failed almost immediately.

Share your knowledge - Leave a comment