Connect with us

Compact Fluorescent Failures

Discussion in 'Electronic Repair' started by Matthew Smith, Apr 5, 2005.

Scroll to continue with content
  1. Greetings

    Over the last couple of months, I have had a spate of failures of
    compact fluorescent lamps (electronic ballasts). (All of the lamps in
    our house, barring the ones that aren't kept on for any time, are CFLs.)

    Whilst a couple have been generic Far-Eastern ones, I've had a couple of
    failures of Philips units as well. (Back in England, one of the first
    Philips electronic CFLs was still running well after seven years.)

    One failure was within a week of installing a replacement - that went
    straight back to the shop for replacement. (I've taken to writing the
    installation date on the base.) In another, the electrolytic capacitor
    had failed (end blown out); I replaced it and it runs fine, although I
    have yet to replace the thermal fuse.

    Of the others, it's a bit of a mystery; I've performed the following tests:
    * Check appropriate DC voltage exists after bridge rectifier.
    * Check electrolytic capacitor out of circuit.
    * Check switching transistors out of circuit.

    I'm now painstakingly tracing the circuits of two of the Philips units
    so that I can mark up some voltage readings against "healthy" units. I
    was rather surprised to find that even the latest units are constructed
    with through-hole, discrete components. (What, no integrated switcher?)

    None of the fittings are enclosed that the units would be getting
    particularly hot.

    We are on a single-wire (19kV SWER - read "highly unreliable power
    supply") HV system - I don't know if this might have any bearing on the
    problem (spikes, etc).

    Some questions:

    1) Does anyone have any figures for commonest causes of failure? (In
    other words, where do I look first?)
    2) Are the thermal fuses essential? Not all units seem to have them -
    at least that I can identify. Whilst I'm not one to go removing safety
    devices willy-nilly, more recent units do seem to have slightly lower
    component counts, fuses included in some. Should I fit thermal fuses to
    ones that don't have them?
    3) Are these devices sensitive to spikes/surges and, if so, should I put
    MOVs in any that I repair?

    I know that these only cost $5 AUD each, but there's a principle at
    stake here...

  2. If they don't have a thermal fuse, they probably have a fusable resistor.
    And, yes, you should replace the safety devices!

    --- sam | Sci.Electronics.Repair FAQ Mirror:
    Repair | Main Table of Contents:
    +Lasers | Sam's Laser FAQ:
    | Mirror Sites:

    Note: These links are hopefully temporary until we can sort out the excessive
    traffic on

    Important: Anything sent to the email address in the message header above is
    ignored unless my full name is included in the subject line. Or, you can
    contact me via the Feedback Form in the FAQs.
  3. I have seen a few with really bad soldering. The switcher transformer
    almost fell out of its holes. Also something blew the FETs on a few.
    They haven't failed again since I changed the FETs and fuse.
  4. Mark

    Mark Guest

    I had one with bad solder joints that I repaired and is still working.

    I had a few where the starter filament on the CF tube went. I have
    not had one last anywhere near 7 years. They are used outside and on
    all night every night.

    Anybody know how much it matters if they are mounted base up or base
    down regarding both light output and relibability. Base up seems to
    work better but I don't have any real data.

  5. NSM

    NSM Guest

    I've heard it's a large electro (often in the base of the lamp) which is
    often the fault.
  6. NSM wrote:
    I can vouch for that being the case in one of the failures I've observed
    - end actually blown out.

    After the posting regarding the poor soldering, I've found some very
    dodgy looking joints on one of the units; I'll re-flow them and see if
    that works - before I continue testing the components one by one.


  7. Edd Whatley

    Edd Whatley Guest


    My number one common failure fault finding is the units raw B+ filte
    cap(s). They inherently have to be mechanically small to fit within th
    cramped casing restraints. [Seems like someone mentioned one venting o
    them...hopefully with a....BANG!!!... like mine did.] Therefore, tha
    tends to have them at woefully smaller values than could be optimal
    They really take a hammering from the ~40 khz operating frequenc
    repetitions of squarewaves. If you have facilities/capabilities o
    doing an ESR test on some used units you should find them onsetting t
    high ESR levels in a hurry. Not having that, just a running of a uni
    (open cased for access) for a warm up period of 10 minutes and then
    totally disconnecting from AC power and subjecting to a finger ti
    temp/warmth test should reveal units that are in ESR excess. (Will b
    quite warm or hot/versus just cool.) Also one could make connectio
    across a cap at a time with a clip leaded AC voltmeter for a rippl
    test, bet you will find it quite high with their selecte
    (necessitated?) values of caps.
    Some units that I did enact a repair upon , ended up in a location tha
    physical width provided no restraint, so I utilized healthier cap value
    as well as 105deg temp ratings (for sure). The new caps then extende
    thru 2 holes of the plastic housing with epoxied plastic insulativ
    "warts" encasing their slight protrusions. 4 years and counting now.

    73's de Ed
  8. James Sweet

    James Sweet Guest

    Aside from what you've already checked, I've had a number of lamps with
    shorted mylar caps in them, other times one of the cathodes will open so
    check those for continuity.
    Some have fusible resistors instead, if there's any sort of protection,
    replace it for sure. If there isn't any it wouldn't hurt to add a low value
    fusible resistor but they should all have something.

    Not worth it, the MOVs will cost more than the lamp generally.
    Exactly, I *always* try to repair them when they fail, it's a simple
    excercise if nothing else.
  9. James Sweet

    James Sweet Guest

    Most of them are rated for 6K-9K hours. The 7 year figure is assuming a few
    hours a day as you'll see in the fine print. Even so, a standard lamp only
    lasts about 1K.
    This will depend on the design of the lamp, base up is likely to work better
    at least on the spiral lamps because the dimple at the end to control vapor
    pressure will be the coolest point as designed, base down it'll be hotter.
  10. NSM

    NSM Guest

    I've bought them at the (US) dollar store (rather a cold phosphor). At a
    buck each I don't fix them.
  11. I have gutted a couple Philips lamps with an LBA202(?) SMT IC.
    Little else of interesting components in such a lamp...
    All the rest (far Eastern, as well as European made Osram and Sylvania)
    using a couple transistors in a power oscillator circuit.
    Of the several (must be more than twenty!) I have gutted the
    electrolytic most often has failed. I changed it in one of the lamps,
    which held more than six months afterwards. The lamps which show
    darkening of the FL tubes are nothing to keep, they probably are at the
    end of life.
    I have seen exploded transistors, but only in a couple of the lamps.
    They are small video output transistors in a 'long TO92' case in the
    far-East imports.
    I collect the ferrite toroids and the hi-voltage caps after measuring
    them. Some are nice replacements in tube gear..
    The caps are marked with values and are nearly always within tolerances.
    Electrolytic. It dries out from high ripple current/high temp in the lamp.
    Some have only a glass-cased fuse in one of the socket wires, some have
    fusible resistors. I have seldom seen temp fuses.

  12. James Sweet

    James Sweet Guest

    I've really grown to like those cold phosphors, especially the daylight ones
    but most everything around here is yellow 2700K.

    Yeah they're not really economical to fix from that standpoint but it is
    kinda fun, I always at least take a crack at it when one dies.
  13. James Sweet

    James Sweet Guest

    It's funny, I'm always hearing about this problem but somehow I have yet to
    ever run across a single one in which the lytic checked bad (ESR meter) or
    blew, I know it's coming but so far never had that happen.
  14. Ken Weitzel

    Ken Weitzel Guest


    Durn, I've been using them for about 10 years now - don't
    have a single incandescent left in the place - but never yet
    had one fail. Mostly Sylania; but a few Globe badged.
    Did throw one away because the light output had dropped too

    Waiting to fix one :) Downside is that looking at them, I
    can't imagine how I'd getting one apart without destroying
    it. Is there a trick I can't see?

  15. James Sweet

    James Sweet Guest

    The name brand ones do tend to be more reliable, as you say the lumen
    depreciation will nessessitate replacement before a really well made one
    completely fails. The problem is the name brands are rather expensive so the
    cheap Chinese junk sells, admittedly my house is full of them.

    Some are easy to open, some aren't. Most of the ones I have will pop apart
    with a scredriver at the seam between the upper and lower section of the
    ballast housing.
  16. Haven't had one yet that can't be opened with a screwdriver like James
    says; of course, you do need to pick the right screwdriver ;-)


    Matthew Smith
    South Australia
  17. James Sweet

    James Sweet Guest

    They may be quite different in Australia, dunno. Most of the ones I've
    opened were either Feit or Commercial Electric. I have one right here
    awaiting a look, it's a TCP #18814/31, which is a somewhat hard to find
    3100K lamp which I like in the bedroom lamp.
  18. NSM

    NSM Guest

    Start by cracking off the metal base - should be easy.
  19. I have found _some_ good electrolytics in the lamps,maybe 10 percent of
    I measure ESR (some caps measure about right cap. value, but has high
    ESR (>20 ohms at 50kHz)
  20. Mike

    Mike Guest

    If you're really lucky, failed ones will pop themselves apart along the
    seam, to facilitate you repairing them. It's a design feature, I think :)

    A couple of ones I've had have failed like that: A loud bang, a smell
    of burnt toffee/toast, which seems to come from the epoxy-like substance
    used to seal inside them. Yum. Obviously the crack in the side let the
    smoke out ... not opened one yet to see what went.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day