One year ago I replaced my mother's kitchen fan lights with 4 CFLs.

Two were good quality OSRAM DULUX 21W and two were cheapo chinese 11W brands,
all of 2700K cct.

Yesterday one of the chinese ones failed (would not light).

Can anyone give me a short course on what the most popular mode of failure for
cheapo CFLs is?

In article <1207759286.123190@athprx04>,
"I.N. Galidakis" <morpheus@olympus.mons> writes:
One year ago I replaced my mother's kitchen fan lights with 4 CFLs.

Two were good quality OSRAM DULUX 21W and two were cheapo chinese
11W brands, all of 2700K cct.

Yesterday one of the chinese ones failed (would not light).

Can anyone give me a short course on what the most popular mode of
failure for cheapo CFLs is?

I added a section in wikipedia on this some time back
http://en.wikipedia.org/wiki/Fluorescent_lamp#End_of_life
giving 4 common causes, after having posted it on usenet.

It doesn't look like it's been hacked around much since I
wrote it, and someone's added a nice picture.

Andrew Gabriel wrote:
In article <1207759286.123190@athprx04>,
"I.N. Galidakis" <morpheus@olympus.mons> writes:

Can anyone give me a short course on what the most popular mode of
failure for cheapo CFLs is?

I added a section in wikipedia on this some time back
http://en.wikipedia.org/wiki/Fluorescent_lamp#End_of_life
giving 4 common causes, after having posted it on usenet.

It doesn't look like it's been hacked around much since I
wrote it, and someone's added a nice picture.

Thanks Andrew. Can you provide some more specific details on exactly what fails
in the electronic ballast failure mode? Is the failure due to some specific
damage within the circuit, such as capacitors, resistors, etc?

It looks to me as though there has been a serious compromise between CFLs and
magnetic ballast fluoros. While CFLs are more convenient, the added electronics
seem to have introduced an additional failure vector which did not exist in the
magnetic ballast fluoros.

In article <1207765922.55820@athprx04>,
"I.N. Galidakis" <morpheus@olympus.mons> writes:
Andrew Gabriel wrote:
In article <1207759286.123190@athprx04>,
"I.N. Galidakis" <morpheus@olympus.mons> writes:

Can anyone give me a short course on what the most popular mode of
failure for cheapo CFLs is?

I added a section in wikipedia on this some time back
http://en.wikipedia.org/wiki/Fluorescent_lamp#End_of_life
giving 4 common causes, after having posted it on usenet.

It doesn't look like it's been hacked around much since I
wrote it, and someone's added a nice picture.

On a reread, someone's added the sentence
"Integral electronic ballasts suffer from shortened lifespans
in high humidity applications."
which I would question.

Thanks Andrew. Can you provide some more specific details on exactly
what fails in the electronic ballast failure mode? Is the failure
due to some specific damage within the circuit, such as capacitors,
resistors, etc?

I don't claim any particular expertise here, but I've seen
two types of failure. Semiconductor failure is instant death.
Sometimes the case of the switching transistor cracks open too,
and sometimes it also takes out a fusible resistor in the supply.

I've also seen failure of the electrolytic storage capacitor
(after the mains rectifier in the circuit). This just causes
slow starting and dimmer operation in the cases I've seen. I
suspect you would see 100Hz (120Hz) flicker too, but I've not
checked for that.

I have observed overheating and deteroriation of the voltage
dependant resistor which is used in some electronic ballast
designs in much the same way the glow starter used with magnetic
ballasts. I don't think I've seen a complete failure of this
though. If it did happen, the CFL might either fail to start
(if the ballast open circuit voltage is lower than the tube's
cold cathode striking voltage), or might start without preheating,
shortening tube life if the CFL is subjected to frequent switching.

It looks to me as though there has been a serious compromise between
CFLs and magnetic ballast fluoros. While CFLs are more convenient,
the added electronics seem to have introduced an additional failure
vector which did not exist in the magnetic ballast fluoros.

In most cases I've come across, CFL failure has not been due
to the ballast, but due to failure of the tube electrodes
(complete loss of emission material, which will be quickly
followed by destruction of the eectrode filaments if the
control gear continues operating the lamp in this state).

Your comments about electronic ballasts are probably more
relevant where the ballast is part of the fixture rather than
the lamp. They don't have anything like the same life as older
magnetic ballasts (certainly the simple series switchstart
magnetic ballasts used on 220-240V supplies, which you never
normally have to replace).

One year ago I replaced my mother's kitchen fan lights with 4 CFLs.

Two were good quality OSRAM DULUX 21W and two were cheapo chinese 11W brands,
all of 2700K cct.

Yesterday one of the chinese ones failed (would not light).

Can anyone give me a short course on what the most popular mode of failure for
cheapo CFLs is?

Additional online circuit diagrams would be nice too.

Many thanks in advance.

On 09 Apr 2008 18:23:43 GMT Andrew Gabriel <andrew@cucumber.demon.co.uk> wrote:
| I added a section in wikipedia on this some time back
| http://en.wikipedia.org/wiki/Fluorescent_lamp#End_of_life
| giving 4 common causes, after having posted it on usenet.
|
| It doesn't look like it's been hacked around much since I
| wrote it, and someone's added a nice picture.

Since the article was about the lamp, I guess the failure modes have to be
focused on the lamp, and not the fixture. The ballast in the fixture can
fail. Also, the starter can fail, but you wouldn't know about those things
over in the UK.

I'd like to see a list of all fluorescent lighting failure modes titled by
the observed symptoms. Last week, a friend's 2-tube kitchen light failed.
The mode is that both tubes just barely glow when power is applied, and
there is some flickering effect in that glow. I didn't actually see it
myself. My father saw it, and he thinks it must be the ballast since both
tubes went out at the same time. They scheduled an electrician to come
check the ballast.

I've seen a variety of other failure modes, some boring and some quite
interesting (like light pulses crawling down the tube, or a narrow band
of light twisting inside the tube).

On Wed, 9 Apr 2008 19:41:26 +0300, "I.N. Galidakis"
morpheus@olympus.mons> wrote:

One year ago I replaced my mother's kitchen fan lights with 4 CFLs.

Two were good quality OSRAM DULUX 21W and two were cheapo chinese
11W brands, all of 2700K cct.

Yesterday one of the chinese ones failed (would not light).

Can anyone give me a short course on what the most popular mode of
failure for cheapo CFLs is?

Additional online circuit diagrams would be nice too.

Many thanks in advance.

The most common point of failure for any electronic system,
and especially one operated at high temperature, is the
electrolytic capacitor.

Low cost CFLs also suffer from the power transistors or FETs
getting too hot due to inadequate heat sinking. There are
also designs in which certain drive resistors may overheat.

The main problem with CFLs is HEAT in the ballast
compartment.

Victor Roberts wrote:
On Wed, 9 Apr 2008 19:41:26 +0300, "I.N. Galidakis"
morpheus@olympus.mons> wrote:

One year ago I replaced my mother's kitchen fan lights with 4 CFLs.

Two were good quality OSRAM DULUX 21W and two were cheapo chinese
11W brands, all of 2700K cct.

Yesterday one of the chinese ones failed (would not light).

Can anyone give me a short course on what the most popular mode of
failure for cheapo CFLs is?

Additional online circuit diagrams would be nice too.

Many thanks in advance.

The most common point of failure for any electronic system,
and especially one operated at high temperature, is the
electrolytic capacitor.

Low cost CFLs also suffer from the power transistors or FETs
getting too hot due to inadequate heat sinking. There are
also designs in which certain drive resistors may overheat.

The main problem with CFLs is HEAT in the ballast
compartment.

Thanks. Searching Google, I found what looks like an interesting page that has
somewhat detailed descriptions (in addition to arguing against the inandescent
ban). At least as far as "failure modes" are concerned, this guy seems to know
what he is talking about.

http://sound.westhost.com/articles/incandescent.htm

On Thu, 10 Apr 2008 20:20:06 +0300, "I.N. Galidakis"





morph...@olympus.mons> wrote:
Victor Roberts wrote:
On Wed, 9 Apr 2008 19:41:26 +0300, "I.N. Galidakis"
morph...@olympus.mons> wrote:

One year ago I replaced my mother's kitchen fan lights with 4 CFLs.

Two were good quality OSRAM DULUX 21W and two were cheapo chinese
11W brands, all of 2700K cct.

Yesterday one of the chinese ones failed (would not light).

Can anyone give me a short course on what the most popular mode of
failure for cheapo CFLs is?

Additional online circuit diagrams would be nice too.

Many thanks in advance.

The most common point of failure for any electronic system,
and especially one operated at high temperature, is the
electrolytic capacitor.

Low cost CFLs also suffer from the power transistors or FETs
getting too hot due to inadequate heat sinking.    There are
also designs in which certain drive resistors may overheat.

The main problem with CFLs is HEAT in the ballast
compartment.

Thanks. Searching Google, I found what looks like an interesting page that has
somewhat detailed descriptions (in addition to arguing against the inandescent
ban). At least as far as "failure modes" are concerned, this guy seems to know
what he is talking about.

http://sound.westhost.com/articles/incandescent.htm

Interesting page, but a bit over the hill in certain areas.
For example, there is a scope trace of the input current
distortion produced when a low PF CFL is used with a phase
controlled dimmer, but no trace showing that even the
"perfect" incandescent lamp creates terrible line current
distortion when used with a phase control dimmer.

--
Vic Robertshttp://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.- Hide quoted text -

- Show quoted text -

The main problem with CFLs is HEAT in the ballast
compartment.

After reading a couple of pages on the electronics contained in the
"ballast" compartment of modern CFLs, I am still stumped at the
simplicity of this chinese small nightlight I have on my webpage:

http://ioannis.virtualcomposer2000.com/spectroscope/collection.html

In article <1208272838.643029@athprx04>, I.N. Galidakis wrote:

After reading a couple of pages on the electronics contained in the
"ballast" compartment of modern CFLs, I am still stumped at the
simplicity of this chinese small nightlight I have on my webpage:

http://ioannis.virtualcomposer2000.com/spectroscope/collection.html

Your nightlight has a short fluorescent lamp that instant-starts at
230 volts AC. The ballast is a simple capacitor ballast.

A fluorescent lamp with a capactor ballast usually requires an
inductor to be added in series with the capacitor, or else the
current waveform will be excessively spiky.

In CFLs of several watts, a capacitor ballast may well cost more
than an electronic ballast if the capacitor is sufficiently large in
size and has a suitable dielectric to withstand continuous use with
line voltage AC.

I found out the hard way about 25 years ago that not just any
non-electrolytic with a DC voltage rating over twice that of the line
voltage can take line voltage AC for long. I once had an 8
microfarad 600 volt oil capacitor burst from overheating after
something like an hour with about 260-280 volts AC across it.

- Don Klipstein (don@misty.com)
--

In article <1208272838.643029@athprx04>, I.N. Galidakis wrote:

After reading a couple of pages on the electronics contained in the
"ballast" compartment of modern CFLs, I am still stumped at the
simplicity of this chinese small nightlight I have on my webpage:

http://ioannis.virtualcomposer2000.com/spectroscope/collection.html

Your nightlight has a short fluorescent lamp that instant-starts at 230
volts AC. The ballast is a simple capacitor ballast.

A fluorescent lamp with a capactor ballast usually requires an inductor
to be added in series with the capacitor, or else the current waveform
will be excessively spiky.

In CFLs of several watts, a capacitor ballast may well cost more than an
electronic ballast if the capacitor is sufficiently large in size and has
a suitable dielectric to withstand continuous use with line voltage AC.

I found out the hard way about 25 years ago that not just any
non-electrolytic with a DC voltage rating over twice that of the line
voltage can take line voltage AC for long. I once had an 8 microfarad 600
volt oil capacitor burst from overheating after something like an hour
with about 260-280 volts AC across it.