Re: Higher power CFL burn out

JohnR66 <nospam@xxxxxxx> wrote:
I'm new to using the higher power (23 watts or more) spiral type CFLs
and just had a respectable brand burn out early. It was a Sylvania 23
watt.

I had what I considered to be an early failure of a GE 42 W spiral CFL.
I bought it two or three years ago and installed it my garage - standard
porcelain screw-base socket on a 3.5" round box on the drywall ceiling,
base up. No shade or reflector, just the lamp in the socket. The
garage is attached but unheated. I don't recall exactly, but it seems
like it lasted less than a year of calendar time, which is probably less
than 1000 hours (more like 500) of lamp-on time. I don't know what it
looked like at failure - I just flipped the switch one day and nothing
happened - no flashes, blinks, noises, etc. I'm not the only person in
the house, so it's possible it had a more "exciting" failure that I
didn't see.

The nomenclature on the base is

(GE HELICAL 42W
logo) 120VAC 60Hz 620mA
FLE42HLX/2/XL/SW
(FCC logo)

(UL LISTED CAUTION: RISK OF ELECTRIC
C logo) US 6G49 CAUTION: SHOCK - DO NOT USE
WHERE DIRECTLY EXPOSED TO WATER.
NOT FOR USE IN LUMINAIRES CONTROLLED
BY A DIMMER.
- MADE IN CHINA -

(Hg)
A147

(same caution statement in French)

Taking it apart revealed the first obvious failure of the fuse or
fusible resistor in series with the center contact. It is labeled
"Fuse" on the circuit board. Physically it looks like an 0.5 watt
resistor, except with a grey body color rather than the usual tan.
Some of the color bands are damaged but I think it's yellow-purple-white-
gold-white. The matte grey coating has burned off most of the middle of
the body, exposing a more silvery metallic substance underneath. There
is a grey smudge on the inside of the white plastic housing next to the
resistor.

The only other obvious trouble is a slight scorching of the component
side of the circuit board around pin A1, which goes to one of the tube
filament leads. The trace on the solder side is also lifting, but this
could be from the heat or from the disassembly. A1 connects to one end
of two capacitors (?) - small gray boxes, one labeled 154K250 and the
other, somewhat larger, labeled

<||> MMKP82
123J1600

.. The logo in the first line is two triangles - think of the 'play'
symbol on a tape deck or MP3 player mirrored vertically.

Other interesting components include a Matsushita electrolytic
capacitor, 47 uF 200 V 130 degC, mounted off of the board about 0.3"
or 8 mm, with insulation on its legs. There's a transformer that has
six pins - maybe a center-tapped primary and center-tapped secondary? -
labeled "20 NO7(A15)". There is a thermistor labeled "PTC" on the
circuit board, and two "ST" brand IRF640 MOSFETs in TO-220. There are a
few surface-mount resistors and capacitors on the solder side. It
appears to be all discrete - no controller chip.

The "model" of the circuit board appears to be C55-1; this is
silkscreened in relatively large letters on both sides.

I haven't yet tried to reassemble it and fire it up. I know I can't
just bypass the fuse resistor, reassemble it, stick it back out in the
garage, and forget about it, but it might be interesting to see if it
will light up on the test bench from a current-limited AC source.

One difference I noted between this lamp and a Phillips "Marathon" CFL
was that the GE lamp has two small-gauge lead wires from the screw base
to the single circuit board, maybe 22 or 24 AWG and 3" long (less than
0.8 mm^2 and about 75 mm). The Phillips lamp had a small circuit board
right in the screw base with the fusible resistor and (IIRC) two
rectifier diodes on it, and three wires up to the main circuit board.
The Phillips lamp also had a Phillips control chip on the main circuit
board. This lamp was a 60 or 75 W equivalent with 3 U-shaped tubes in
parallel; it died of massive trauma to the glass, not because of
electrical failure. It was maybe about 6 or 7 years old.

Matt Roberds

.

Relevant Pages

  • Re: Live signals on non-existant tracks
    ... would run for years without failure. ... imagine that a lamp failure would see the lamp replaced to silence ... the lamp in the OP's photo is lit - clearly not a resistor. ...
    (uk.railway)
  • Re: Voltage converter circuit - Bosch / Siemens washing machine
    ... resistor (first two bands brown and black, ... from R2 ends up at a connector at the edge of the circuit board (e.g. ... the motor or pump and possibly control the speed depending on the complexity ... external connections to the motor and pump from the circuit board. ...
    (sci.electronics.repair)
  • Re: Basic resistor question
    ... at the body of the resistor and to bend out past the needle nose. ... If the lamp is a low voltage lamp, let's say 6 volts as an example. ... out that the resistance of the lamp when it is operating is 300 ohms. ...
    (rec.models.railroad)
  • Re: Tech Firepower problem
    ... This is usually caused by a failure of one or more components in the lamp ... Usually this is caused by a failed 6821 PIA chip for the ... also some good experience/skills with circuit board repair to properly ... board out to a qualified repair shop for professional repair- it's usually ...
    (rec.games.pinball)
  • Re: Resistor instead of ballast
    ... The PL-S type may not work reliably with a resistive ballast. ... because it is a lamp type which has a built-in glowbottle starter, ... With a simple resistor you don't get this same ...
    (sci.engr.lighting)