CFLs - switching on and off
colin...@hotmail.com
off was not a good idea and that they should be switched on and left
on. Unlike filament lamps which do not seem to mind.
How do the modern CFLs compare/suffer etc etc?? I know that they can
take a minute or two to warm up and maximise their light output.
If they should be switched on and left on, then they begin to defeat
the very purpose of having energy saving CFLs fitted.
Marsbar
<colin...@hotmail.com> wrote in message
news:1187458833.4...@57g2000hsv.googlegroups.com...
fluorescent light on and off wasn't a good idea?
Andy Burns
> Switch on only when required. What made you think that switching a
> fluorescent light on and off wasn't a good idea?
It was either a common misapprehension, or used to be different with
older tubes. I certainly remember an "order" going around school in the
early 70's to NOT switch off lights during break-times as it took more
electricity to re-start them, than to leave them on for 20 minutes.
tra...@optonline.net
In the case of CFL's I recall that while they do take more energy on
start-up, the break even point comes after a matter of a few seconds.
So, for practical purposes, they should be turned off when not
needed. With all of the ones I've seen, the bigger problem is what
was already noted. They take a couple mins to reach full output.
Even worse, the output is terrible for the first 30-60secs. For that
reason, I leave them on more than I would a regular light, thinking
I'll need it again in maybe 10 mins. But overall, I'm pretty sure
I'm saving a good bit on energy.
Fred
"Andy Burns" <usenet....@adslpipe.co.uk> wrote in message
news:13ced8u...@corp.supernews.com...
It was a complete redherring then and still is. Yes the current is higher
in order to ignite the tub, but given it lasts for a second or so. As long
as the tube is switched off for over, say 5 seconds, there'll still be a net
saving.
Newshound
<tra...@optonline.net> wrote in message
news:1187460863.6...@w3g2000hsg.googlegroups.com...
the lamp from thermal cycling which shortens the life. I leave a CFL on in a
(very dark) hall and landing all day, and they last years.
meow...@care2.com
a popular myth
http://www.wiki.diyfaq.org.uk/index.php?title=Fluorescent_Lighting
NT
Derek Geldard
<news...@fairadsl.co.uk> wrote:
>> In the case of CFL's I recall that while they do take more energy on
>> start-up, the break even point comes after a matter of a few seconds.
>> So, for practical purposes, they should be turned off when not
>> needed. With all of the ones I've seen, the bigger problem is what
>> was already noted. They take a couple mins to reach full output.
>> Even worse, the output is terrible for the first 30-60secs. For that
>> reason, I leave them on more than I would a regular light, thinking
>> I'll need it again in maybe 10 mins. But overall, I'm pretty sure
>> I'm saving a good bit on energy.
>>
>I don't think the startup energy is the issue, it is the wear and tear on
>the lamp from thermal cycling which shortens the life.
It's not an issue of thermal cycling :
http://en.wikipedia.org/wiki/Fluorescent_lamp
It's during start up that most of the wear and tear is done to the
tube filaments. The filaments are coated with an emission mix which is
sputtered off causing blackening of the tube ends. When most of it has
gone the striking voltage of the tube rises until in the end stage the
tube will not strike any more. Some electronic ballasts can detect
this condition coming and shut down, other earlier/cheaper electronic
ballasts will continue to try to start the tube and end up being
damaged by overvoltage and fail. Simpler inductive ballasts with glow
switch starters do not fail but will typically run for months with the
tube flashing on but failing to start and then cyclically restarting
from scratch, this causes a nuisance and should not be allowed to
continue as eventually the starter can overheat melting it's mounting.
Derek Geldard
>
>"Andy Burns" <usenet....@adslpipe.co.uk> wrote in message
>news:13ced8u...@corp.supernews.com...
>> On 18/08/2007 18:59, Marsbar wrote:
>>
>>> Switch on only when required. What made you think that switching a
>>> fluorescent light on and off wasn't a good idea?
>>
>> It was either a common misapprehension, or used to be different with older
>> tubes. I certainly remember an "order" going around school in the early
>> 70's to NOT switch off lights during break-times as it took more
>> electricity to re-start them, than to leave them on for 20 minutes.
>
>It was a complete redherring then
When, and with fittings of what vintage? Tubes and control gear have
been improved continuously, changing out of all recognition since the
end of the war.
>and still is.
Sorry not the case, there is still an optimum way of operating the
tube.
This sort of advice was originally intended for industrial users who
had a machine shop or a weaving shed etc lighted with hundreds of
fluorescent fittings mounted on the ceiling over the machines. These
tubes were replaced on a planned preventative maintenance basis whilst
the plant was shut down for (annual ?) holidays. To have tubes fail
between times was very expensive, the electrician would have to work
above the machines (which would have to be stopped) and there was the
possibility that a tube would be dropped or broken contaminating the
workplace with broken glass, and if that was a loom would include
hundreds of feet of very expensive cloth.
>Yes the current is higher in order to ignite the tub,
That's not the issue (even if it's true, I've not seen it mentioned
elsewhere), shortening of the tube life is the issue. Modern control
gear can be a lot better than old stuff, but as always the best
equipment is more expensive and not always used.
>but given it lasts for a second or so. As long as the tube is
>switched off for over, say 5 seconds, there'll still be a net saving.
No.
http://en.wikipedia.org/wiki/Fluorescent_lamp
Nowadays domestic users of fluorescent tubes need not concern
themselves too much, but "Best Practice" is "Best Practice". If you
have a fitting that requires a lot of effort to get at (above the
stairs say) it makes sense to get the most out of the tube.
DG
Joseph Meehan
Actually it was that the total cost was more to turn them off than to
leave them on for periods of about 20 minutes or less. The largest factor
was the cost of replacing the lamps because cycling them reduced their life.
Note those figures were based on commercial applications and included the
cost of the maintenance man doing the replacement. This was one of the
studies we took apart in my statistics class while working on my economics
degree.
--
Joseph Meehan
Dia 's Muire duit
clot
I seem to recall, and emphasise seem, that Which back in the year dot
when I started to subscribe, then suggested that a tube was best left on
for 45 mins., if you were likely to re-enter that room. I still have
that habit, but have reduced it to about 15 - 20 minutes.
Derek Geldard
>colinst...@hotmail.com wrote:
>
>> I have understood that switching fluorescent lamps - tubes - on and
>> off was not a good idea and that they should be switched on and left
>> on. Unlike filament lamps which do not seem to mind.
>> How do the modern CFLs compare/suffer etc etc?? I know that they can
>> take a minute or two to warm up and maximise their light output.
>> If they should be switched on and left on, then they begin to defeat
>> the very purpose of having energy saving CFLs fitted.
>
>a popular myth
Which one?
Could you point out please the particular myth you refer to?
>http://www.wiki.diyfaq.org.uk/index.php?title=Fluorescent_Lighting
>
Could you expand on that? It's a rather large webpage but all it says
on lamp life is :
"Tube life depends on type of ballast (& starter where used), and how
often the tube is switched on and off."
Which is correct but not specially helpful.
BTW Mr. Meow. we had another CFL fail yesterday after 6 months service
in a cap down open fitting. It was a Feit electric 23 watt spiral
offered for sale promising a 5 year life. The phosphor is quite
significantly darkened and the top of the plastic end cap containing
the electronic ballast has been toasted brown, what happened to the
cheap Chi/Taiwa-nese pcb assy inside is anyone's guess.
DG
Fred
"Derek Geldard" <d...@miniac.demon.co.uk> wrote in message
news:3goec3987kj238ofd...@4ax.com...
starting of a fluorescent tube was equivalent to 1/2 hour running.
>>but given it lasts for a second or so. As long as the tube is
>>switched off for over, say 5 seconds, there'll still be a net saving.
>
> No.
>
> http://en.wikipedia.org/wiki/Fluorescent_lamp
>
> Nowadays domestic users of fluorescent tubes need not concern
> themselves too much, but "Best Practice" is "Best Practice". If you
> have a fitting that requires a lot of effort to get at (above the
> stairs say) it makes sense to get the most out of the tube.
>
I have read the article and it confirms what I know. I was answering a post
outlining a supposition put forward in the 70's. At that time the ballast
would be an inductor and the starter would be gas filled device.
It's a great shame that the article doesn't qualify "Lamps operated for
typically less than 3 hours each switch-on will normally run out of the
emission mix before other parts of the lamp fail". That is the most common
failure mechanism for lamps. I recall figures which suggested that whilst
the lamp lifetime when "on" was shorter, the act of switching it off when
not needed actually increased the "real" life time of the lamp as well as
saving energy!
Graham.
A five year life would be 5*365*24=45,800 hours
I_don't_think_so.
Why do GLS lamp manufacturers give an honest average
lifetime in hours (say 2,000) but with longer lived CFL lamps,
you have to find the small print that qualifies the headline figure.
One for the ASA IMHO.
--
Graham
%Profound_observation%
The Natural Philosopher
Classic case of insulated electronics failing after prolonged 'on' periods.
I remember having this issue once with some gear I designed: A probe
inside showed the unit took nearly ten hours to reach equilibrium - or
would have. It generally failed after 7..;-) ten holes.. 5 in the top
and 5 in the base made it totally reliable.
Don Klipstein
That is a common myth. Any extra power consumption surge during
starting amounts to the amount of energy consumed in a second or two of
steady operation - or less.
However, the cost of bulb wear from an extra start could require several
minutes of off-time in order for electricity savings to outweigh that.
How many minutes? This depends on the bulb cost, electricity cost,
starting method, and when the bulbs were made.
Modern fluorescents suffer less starting wear than older ones.
The "break even time" is less for ones 4 feet and longer than smaller
ones. Lower wattage bulbs cost even more than 4-footers, and lower
ratio of power consumption to bulb cost increases the "break even time".
The "break even time" varies with starting method because different
starting methods cause different amounts of starting-related wear:
"Program Start" - this is used in some CFLs. The bulb does not come on
at all until a fraction of a second to about a second after power is
applied, then turns on without blinking. It may have a "rapid fade-on"
during a fraction of a second.
This causes the least wear, and is often used in CFLs of Philips and
Sylvania brands (and some others but I can't remember who and I have not
tried them all).
"Rapid Start" - bulbs come on instantly very dim, usually slightly
flickery, stay dim for about half a second to a second, then quickly
brighten over a fraction of a second. This is next-best to "Program
Start" for minimizing wear from starting.
"Trigger Start" refers to a variation of "rapid start" used on bulbs
designed for "Preheat Start".
"Instant Start" - The bulbs are on instantly. Sometimes the brightness
makes a sudden slight upward jump a fraction of a second after starting
when the filaments achieve normal operating temperature. This is worse
than "program start" and "rapid start" for starting-related-wear.
"Preheat Start" - usually has a "glow switch starter" or "glow bottle
starter", rarely an electronic alternative. Bulbs usually blink a few
times before they start and stay started. Since each blink causes
starting-related-wear, this method is worst for starting-related-wear.
Fluorescent lamps using this starting method, especially with bulbs 22
watts or less, are likely to have break-even times in/near the 15 minute
to 1 hour ballpark, and should be left on rather than being turned off and
back on shortly later.
Electronic versions of starters that make the first starting attempt
successful greatly reduce the starting-related-wear. Electronic schemes
that make some determination when the filaments are properly preheated as
well as making the first starting attempt successful are at least
arghuably "program start" schemes.
===========================
Now, how bad is it to turn off and back on a fluorescent lamp?
Case 1: 4-footer, 32 watt T8, instant start costing $2. I am guessing
that a start with an instant start ballast takes 20 minutes off its 20,000
hour life. I am assuming also that this is with an electronic ballast
that improves energy efficiency and also mildly underpowers the lamp/bulb
(expect about 90% of "catalog" light output from the lamp/bulb), and
per-bulb power consumption could be about 30 watts, usually not exceeding
32 watts. Another assumption - electricity cost 11 cents per KWH, which I
believe is close to current USA average residential rate.
Starting wear taking 20 minutes off the life of a $2 20,000 hour bulb
costs .003333 cent. (Actually slightly less, since these bulbs are rated
to last 20,000 hours with one start using rapid-start-method every 3
hours, and will last slightly longer than 20,000 hours if used
continuously.)
To consume .003333 cent worth of electricity at 30 watts and 11 cents
per KWH (.33 cent per hour) only takes .0101 hour, about 36 seconds.
Case 2: 15 watt spiral CFL purchased at a higher-side price of $7,
instant-start. Assuming that the filament here is optimized a bit more
for enduring starts, so I guess 15 minutes of life lost per start. (It
could easily be 10 minutes or less.) Also, rated life expectancy 6,000
hours. (I know, now they make ones rated 7500 or 10,000 hours. But I
want to be a little conservative here!)
The life rating is with a start every 3 hours. So if it lasts 6,000
hours with a start every 3 hours and a start costs 15 minutes, then
continuous operation avoids 2,000 starts over 6,000 hours and would add
500 hours to the 6,000 hour figure and make it 6,500 hours.
Assuming that all my numbers here are good including ones that I am
halfway pulling out of a hat, a start costs about .027 cent. If you get
these bulbs in a $10 4-pack, then a start costs about .0096 cent. If you
get these bulbs in a promotional $10 6-pack, then a start costs about
.0063 cent.
Now, to balance against 15 watts of power consumption at 11 cents per
KWH (.165 cent per hour):
.027 cent per start ($7 bulb) means "break-even" at 10 minutes
.0096 cent per start ($2.50 bulb) breaks-even at about 3.5 minutes
.0063 cent per start ($1.67 bulb) breaks-even at about 2.35 minutes
Lower wattage CFLs will tend to have longer "break-even" times, higher
wattage ones will tend to have shorter "break-even" times.
Break-even time is also inversely proportional to electricity cost. It
will be a little shorter in the metro areas of Chicago, Philadelphia and
NYC.
===================================================
Where starting wear on lower wattage CFLs is a big issue, consider cold
cathode ones. Those do not suffer any significant starting wear, and are
often even rated for heavy blinking duty. They also have longer life
expectancy than hot cathode ones of same wattage even without starting
wear.
Downsides:
1. Ones over 3 watts are mostly available from online lightbulb sellers
such as bulbs.com. Even there, they are largely limited to about 8 watts
or so as of last time I checked.
(A 3 watt cold-cathode model I have seen at Home Depot - of the N:Vision
brand. Look for it being rated for dimming, probably also blinking, and
life expectancy 20,000 or 24,000 hours or so.)
2. Cold cathode ones produce somewhat less light than hot cathode ones of
same wattage.
An 8 watt cold-cathode one only slightly outshines a 25 watt
incandescent, while a 9 watt hot cathode one is about as bright as most 40
watt incandescents.
- Don Klipstein (d...@misty.com)
Don Klipstein
wrote:
In my small smapling of Feit Electric, specifically a 23 watt spiral, I
had better luck than that. I replaced it with a 3500K one of a different
brand while the Feit was still working, because I wanted the
whiter-but-still-warm color of 3500K.
Meanwhile, there is some tendency for longer life for ones of the "Big
3" brands: GE, Philips and Osram-Sylvania. Among others, there is at
least some chance of better performance if the CFL has the Energy Star
"seal of aproval" (my words).
- Don Klipstein (dkli...@rcn.com, d...@misty.com)
Don Klipstein
In the "States", many CFLs have limited warranties for 6 or 7 years or
whatever in "normal home use". The packages also clearly enough state
actual operating hour life expectancy figures - usually 6,000 to 10,000
hours. That is in lab conditions including 3 hours per start at 25 degree
C ambient and average figure, so I suspect some significant number to burn
out at or before 3,000 hours due to randomly fairing worse than average,
running less than 3 hours per start, or running where it gets a lot hotter
than 25 degrees C. I suspect based on some experience that averages may
be 4,000-6,000 hours in most home use.
The so-many-years figures could get more conservative if more users
request the manufacturers to make good on the warranties, which I suspect
few do.
- Don Klipstein (d...@misty.com)
Edwin Pawlowski
"Graham." <m...@privacy.com> wrote in message
>
> A five year life would be 5*365*24=45,800 hours
> I_don't_think_so.
>
> Why do GLS lamp manufacturers give an honest average
> lifetime in hours (say 2,000) but with longer lived CFL lamps,
> you have to find the small print that qualifies the headline figure.
You have to read the fine print. Under that bit Five Years is the small
print that bases the life on X number of hours per day.
Feit states 8,000 hours. From my experience with a few of them, I'd say
they are close and have even exceeded that. I use them in some places in our
warehouses for night lights and security lights that burn 24/7. Not a bad
deal for a bulb that sells for 74ข.
John Rumm
> It's very much the issue! I recall stories as the poster had where the
> starting of a fluorescent tube was equivalent to 1/2 hour running.
Analyse that statement logically and you will see it makes no sense...
How much current would be drawn by say a single 56W tube? 230mA.
How long does it take to start? say 3 secs
So the current drawn in those three seconds would need to be 600 times
(i.e. 1800 secs over 3) the nominal current so as to consume the same
amount of energy, or 138A.
Any guesses as to how many tubes with that sort of inrush current you
could start on a circuit protected by a 6A breaker without tripping it? ;-)
--
Cheers,
John.
/=================================================================\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\=================================================================/
mm
<see.my.s...@nowhere.null> wrote:
>Fred wrote:
>
>> It's very much the issue! I recall stories as the poster had where the
>> starting of a fluorescent tube was equivalent to 1/2 hour running.
>
>Analyse that statement logically and you will see it makes no sense...
First off, it's true, and it may be the basis for the original post,
so it's worth discussing.
Secondly, I would say that it does make sense**, but it's not accurate
and for someone who knows anything about the topic, it's not
believable.
Something that makes no sense, to me, would be something whose
intended meaning I can't discern.
**A lot of things use more electricity on start-up, so the sentence is
not illogical. But the numbers are wrong.
mm
>I have understood that switching fluorescent lamps - tubes - on and
>off was not a good idea and that they should be switched on and left
>on. Unlike filament lamps which do not seem to mind.
Why do you say they don't mind? Haven't you noticed that they almost
always burn out at the moment they are turned on?
That's when they heat up, the filament expands (maybe faster than when
it contracts on turning it off) and stresses on the filament are
greatest.
If you leave a filament bulb on, it will generally last much longer
than if you turn it on and off.
Mike Ruskai
<NOPSAM...@bigfoot.com> dribble thusly:
>On Sat, 18 Aug 2007 10:40:33 -0700, colin...@hotmail.com wrote:
>
>>I have understood that switching fluorescent lamps - tubes - on and
>>off was not a good idea and that they should be switched on and left
>>on. Unlike filament lamps which do not seem to mind.
>
>Why do you say they don't mind? Haven't you noticed that they almost
>always burn out at the moment they are turned on?
I was amazed no one had pointed out the error in that statement, and
was about to comment. Sure enough, last message in the thread, and
I'm scooped.
--
- Mike
Ignore the Python in me to send e-mail.
meow...@care2.com
AIUI this old advice was based on an elementary error, as it
overloooked the fact that although the tubes may last more hours if
left on, they will in fact last less days.
This whole subject is filled with myth and bad advice. Today for
domestic installs, the only sensible thing to do is turn off when not
wanted, however long or short that may be.
NT
Alan
Philosopher <a@b.c> wrote
I've just had a Pro-life 25W spiral bulb fail in a spectacular way (very
loud bang followed by lingering burnt smell). It's been fitted
approximately 6 months.
Photos
http://www.amac.f2s.com/bulb/index.htm
--
Alan
news2006 {at} amac {dot} f2s {dot} com
dennis@home
There are a lot of "fools" about that believe anything they are told/hear
and lack any sort of understanding to know its rubbish.
Just look at Lenny in uk.t.b and you will know what I mean.
I bet there are loads of people about that think you should leave the CH on
at night as it takes more energy to warm the house up in the morning than it
does to keep it warm.. another common misconception doing the rounds..
again!!
dennis@home
"Newshound" <news...@fairadsl.co.uk> wrote in message
news:5iovu3F...@mid.individual.net...
> I don't think the startup energy is the issue, it is the wear and tear on
> the lamp from thermal cycling which shortens the life. I leave a CFL on in
> a (very dark) hall and landing all day, and they last years.
>
I turn mine on and off and they last years (how long I don't know as they
are still going after a few years now).
What does it prove?.. only that a sample of one or two is meaningless.
meow...@care2.com
Don is a lighting expert with a very informative and detailed site
last time I looked. However readers should bear in mind that tube
types, starting types, costs, common practices and terminology are all
different here in UK to the US.
Also I think somehting was missed in your calculations of switch off
break even time. If keeping the tube on for 20 minutes gains you 20
minutes extra tube life, you have in fact gained absolutely nothing.
The only difference is that 20 minutes of electricity have been
wasted. You wont get a single extra day of service time out of the
tube this way. I dont know why but this is so often overlooked in
these calculations.
NT
tra...@optonline.net
> On Sun, 19 Aug 2007 02:54:32 +0100, John Rumm
>
> >Fred wrote:
>
> >> It's very much the issue! I recall stories as the poster had where the
> >> starting of a fluorescent tube was equivalent to 1/2 hour running.
>
> >Analyse that statement logically and you will see it makes no sense...
>
> First off, it's true, and it may be the basis for the original post,
> so it's worth discussing.
>
> Secondly, I would say that it does make sense**, but it's not accurate
> and for someone who knows anything about the topic, it's not
> believable.
>
> Something that makes no sense, to me, would be something whose
> intended meaning I can't discern.
>
> **A lot of things use more electricity on start-up, so the sentence is
> not illogical. But the numbers are wrong.
>
Well, Duh! His obvious point was that if you look at the real facts,
which he clearly presented, then it makes no sense, because the
current required in the few secs of startup would be huge. And I
think your definition of "makes no sense" equals "can't discern
intended meaning" isn't exactly mainstream. For example, if someone
said that Mars revolves aroung the Earth, the meaning is quite clear,
yet any reasonable person would say that makes no sense.
>
> >How much current would be drawn by say a single 56W tube? 230mA.
>
> >How long does it take to start? say 3 secs
>
> >So the current drawn in those three seconds would need to be 600 times
> >(i.e. 1800 secs over 3) the nominal current so as to consume the same
> >amount of energy, or 138A.
>
> >Any guesses as to how many tubes with that sort of inrush current you
> >could start on a circuit protected by a 6A breaker without tripping it? ;-)- Hide quoted text -
>
> - Show quoted text -
Smitty Two
tra...@optonline.net wrote:
> On Aug 19, 3:03 am, mm <NOPSAMmm2...@bigfoot.com> wrote:
> > On Sun, 19 Aug 2007 02:54:32 +0100, John Rumm
> >
> > <see.my.signat...@nowhere.null> wrote:
> > >Fred wrote:
> >
> > >> It's very much the issue! I recall stories as the poster had where the
> > >> starting of a fluorescent tube was equivalent to 1/2 hour running.
> >
> > >Analyse that statement logically and you will see it makes no sense...
> >
> > First off, it's true, and it may be the basis for the original post,
> > so it's worth discussing.
> >
> > Secondly, I would say that it does make sense**, but it's not accurate
> > and for someone who knows anything about the topic, it's not
> > believable.
> >
> > Something that makes no sense, to me, would be something whose
> > intended meaning I can't discern.
> >
> > **A lot of things use more electricity on start-up, so the sentence is
> > not illogical. But the numbers are wrong.
> >
>
> Well, Duh! His obvious point was that if you look at the real facts,
> which he clearly presented, then it makes no sense, because the
> current required in the few secs of startup would be huge. And I
> think your definition of "makes no sense" equals "can't discern
> intended meaning" isn't exactly mainstream. For example, if someone
> said that Mars revolves aroung the Earth, the meaning is quite clear,
> yet any reasonable person would say that makes no sense.
>
mm's definition may not be mainstream, but it's correct. Colloquialisms
aside, the liberties we take with our (mis-)use of the English language
are staggering. It's surprising we understand half of what's said or
written. I try to bite my tongue when it comes to playing grammar and
spelling cop on usenet, (and I'm not even close to perfect myself, and
well aware of it) but there's a hell of a lot of people wandering around
the planet that barely qualify as literate. The significance and
implications of our collective ignorance are widely underestimated.
meow...@care2.com
> On Sat, 18 Aug 2007 10:40:33 -0700, colin...@hotmail.com wrote:
> >I have understood that switching fluorescent lamps - tubes - on and
> >off was not a good idea and that they should be switched on and left
> >on. Unlike filament lamps which do not seem to mind.
>
> Why do you say they don't mind? Haven't you noticed that they almost
> always burn out at the moment they are turned on?
>
> That's when they heat up, the filament expands (maybe faster than when
> it contracts on turning it off) and stresses on the filament are
> greatest.
>
> If you leave a filament bulb on, it will generally last much longer
> than if you turn it on and off.
Do you have further information on this? I understood that avoidance
of switch-on surges gave only minimal extra life to GLS lamps, though
considerable extra life to halogens.
NT
mm
>
>>>Yes the current is higher in order to ignite the tub,
>>
>> That's not the issue (even if it's true, I've not seen it mentioned
>> elsewhere), shortening of the tube life is the issue. Modern control
>> gear can be a lot better than old stuff, but as always the best
>> equipment is more expensive and not always used.
How does that contradict what he said? And it is the issue, like he
says below.
>>
>It's very much the issue! I recall stories as the poster had where the
>starting of a fluorescent tube was equivalent to 1/2 hour running.
They said the same thing about lncandescent lights, btw, and I
believed it until I thought about it. My line of thinking was. If
turning the light on was equivalent to using 10 minutes of
electricity, imagine the time it takes for the bulb to get to full
brightness and imagine running 10 minutes worth of electricity through
the filament in less than a second. That's a half amp or so times
36,000 (sec./10 minutes). That's 18,000 amps through my lightbulb.
Wow. Maybe I'll rejuvenate a cadaver next time.
I haven't read the url below, so I'll leave that for later.
>
** Frank **
<colin...@hotmail.com> wrote in message
news:1187458833.4...@57g2000hsv.googlegroups.com...
>I have understood that switching fluorescent lamps - tubes - on and
> off was not a good idea and that they should be switched on and left
> on. Unlike filament lamps which do not seem to mind.
Not good for filament lamps too. In fact, too many switching operations are
not good for most things like computers, motors, TVs, etc. Switching
transients (both switching on and off) could have many thousand volts and
could draw 10x or more rated current across the device which could results
in electrical, mechanical and thermo shock. Note many lamp failures are at
the moment of switching either on or off.
My own experience is CFLs (and Circle Lines) are much more susceptible to
switching than incandescent or the 4' and 8' fluorescent lamps. YMMV.
Derek Geldard
wrote:
>On Sat, 18 Aug 2007 10:40:33 -0700, colin...@hotmail.com wrote:
>
>>I have understood that switching fluorescent lamps - tubes - on and
>>off was not a good idea and that they should be switched on and left
>>on. Unlike filament lamps which do not seem to mind.
>
>Why do you say they don't mind? Haven't you noticed that they almost
>always burn out at the moment they are turned on?
>
>That's when they heat up, the filament expands (maybe faster than when
>it contracts on turning it off) and stresses on the filament are
>greatest.
>
>If you leave a filament bulb on, it will generally last much longer
>than if you turn it on and off.
>
During the time it is energised the filament loses metal and gets
thinner, not necessarily uniformly along it's length. Eg. It is
cooled where it is supported and less metal is lost at these points.
Where it is thinnest is also where it is mechanically weak and it's
electrical resistance is greatest. At switch on the whole filament has
a low resistance and takes a big surge of current. The thin weak
points are heated up disproportionately to the rest of the filament
and eventually the filament will fail at one of these points.
I don't think that routine switching on and off shortens the life of
the lamp that much, but rather that a lamp at the end of it's life
which has become frail will tend to fail when it's switched on.
DG
Grimly Curmudgeon
drugs began to take hold. I remember meow...@care2.com saying something
like:
>Do you have further information on this? I understood that avoidance
>of switch-on surges gave only minimal extra life to GLS lamps, though
>considerable extra life to halogens.
As anecdotal evidence only...
I've noticed (as have others) that side-mounted filament lamps don't
last long, typically a few months of normal useage. Since powering a
desktop lamp through a push-button dimmer and using that dimmer as the
only on/off switch the same bulb has been in that lamp for the past 4
years. Ordinary 60W bulb, btw.
--
Dave
willshak
> <colin...@hotmail.com> wrote in message
> news:1187458833.4...@57g2000hsv.googlegroups.com...
>
>> I have understood that switching fluorescent lamps - tubes - on and
>> off was not a good idea and that they should be switched on and left
>> on. Unlike filament lamps which do not seem to mind.
>>
>
> Not good for filament lamps too. In fact, too many switching operations are
> not good for most things like computers, motors, TVs, etc. Switching
> transients (both switching on and off) could have many thousand volts and
> could draw 10x or more rated current across the device which could results
> in electrical, mechanical and thermo shock. Note many lamp failures are at
> the moment of switching either on or off.
>
> My own experience is CFLs (and Circle Lines) are much more susceptible to
> switching than incandescent or the 4' and 8' fluorescent lamps. YMMV.
>
Yes, MMDV (no, not 2505. My Mileage Did Vary)
I've had the spiral CFL bulbs installed in all of my formerly
incandescent bulb lamps, wherever the lamp took a regular bulb. The
exceptions are mini spots, and decorative candelabra bulbs.
The CFLs in my basement stair lights have been operated for the past 4
years, and at least 4X a day, and more.
The others have been changed over the past couple of years
None have not failed yet. They only take a few seconds to full
brightness, and when first turned on, they are about 80% bright.
They are GE, if that makes a difference.
>
>
>> How do the modern CFLs compare/suffer etc etc?? I know that they can
>> take a minute or two to warm up and maximise their light output.
>> If they should be switched on and left on, then they begin to defeat
>> the very purpose of having energy saving CFLs fitted.
>>
>>
>
>
>
>
--
Bill
In Hamptonburgh, NY
To email, remove the double zeroes after @
John Rumm
>>> It's very much the issue! I recall stories as the poster had where the
>>> starting of a fluorescent tube was equivalent to 1/2 hour running.
>> Analyse that statement logically and you will see it makes no sense...
>
> First off, it's true, and it may be the basis for the original post,
> so it's worth discussing.
I accept that the claim has been made - I have heard people making it as
well.
I also expect that is is a corruption of the original research that was
looking at overall costs including lamp life and not just running costs.
It has just that much of the detail has been lost in the Chinese
whispers along the way and it has been reduced to an absurdity
concerning just energy costs.
John Rumm
> This whole subject is filled with myth and bad advice. Today for
> domestic installs, the only sensible thing to do is turn off when not
> wanted, however long or short that may be.
Unless you have one of those pesky CFLs that take 10 mins to achieve a
worthwhile light output! ;-)
Bob Eager
<see.my.s...@nowhere.null> wrote:
> meow...@care2.com wrote:
>
> > This whole subject is filled with myth and bad advice. Today for
> > domestic installs, the only sensible thing to do is turn off when not
> > wanted, however long or short that may be.
>
> Unless you have one of those pesky CFLs that take 10 mins to achieve a
> worthwhile light output! ;-)
Nah, don't have that problem...they *never* achieve a worthwhile light
output!
--
The information contained in this post is copyright the
poster, and specifically may not be published in, or used by
http://www.diybanter.com
David Hansen
meow...@care2.com wrote this:-
>> If you leave a filament bulb on, it will generally last much longer
>> than if you turn it on and off.
>
>Do you have further information on this? I understood that avoidance
>of switch-on surges gave only minimal extra life to GLS lamps, though
>considerable extra life to halogens.
It is the same with valves. Left on they can last for a very long
time, turned on and off they can be unreliable. See Tommy Flowers,
1930s telecommunications equipment and the genesis of Colossus for
the evidence.
--
David Hansen, Edinburgh
I will *always* explain revoked encryption keys, unless RIP prevents me
http://www.opsi.gov.uk/acts/acts2000/00023--e.htm#54
David Hansen
<will...@00hvc.rr.com> wrote this:-
>I've had the spiral CFL bulbs installed in all of my formerly
>incandescent bulb lamps, wherever the lamp took a regular bulb. The
>exceptions are mini spots,
There are now some good energy saving bulbs for these.
>and decorative candelabra bulbs.
While there is no equivalent of clear bulbs there are now a number
of "candle" shaped energy saving bulbs that are very similar to
pearl "candle" bulbs.
It is becoming difficult to find an indoor application where there
is not a suitable energy saving bulb.
Don Klipstein
Most incandescents do not suffer significant wear from starting. What
happens is that an aging filament becomes unable to survive a cold start a
little before it becomes unable to survive continuous operation.
An incandescent burnout is generally from melting of a thiner hotter
section of the filament. Such a "thin spot" has a temperature overshoot
during a cold start.
Once a filament has a hot-running thin spot that becaomes unable to
survive a cold start, its hours are numbered. The thin spot suffers worse
evaporation because it runs hotter, and this condition accelerates worse
than exponentially, so the filament's days/hours are numbered once it is
in bad enough shape to be unable to survive a cold start.
- Don Klipstein (d...@misty.com)
Don Klipstein
>meow...@care2.com wrote:
>
>> This whole subject is filled with myth and bad advice. Today for
>> domestic installs, the only sensible thing to do is turn off when not
>> wanted, however long or short that may be.
>
>Unless you have one of those pesky CFLs that take 10 mins to achieve a
>worthwhile light output! ;-)
I see few taking that long, and they tend to be outdoor types in colder
conditions.
If you want ones that warm up faster, I have found in general that ones
without outer bulbs start brighter and warm up faster than ones with outer
bulbs. Ones with outer bulbs have the tubing getting hotter, and are
designed to work optimally at such a higher tubing temperature.
- Don Klipstein (d...@misty.com)
tra...@optonline.net
> In article <1187526585.974797.155...@r29g2000hsg.googlegroups.com>,
According to whom? Here's a dictionary definition of the word sense,
when used as a noun:
The American Heritage® Dictionary of the English Language: Fourth
Edition. 2000
"sense 4c. Something sound or reasonable: There's no sense in waiting
three hours."
Or applied in this case:
>Analyse that statement logically and you will see it makes no sense...
or my example:
>For example, if someone said that Mars revolves aroung the Earth, the meaning is >quite clear, yet any reasonable person would say that makes no sense.
Both of those are perfectly correct usage as defined by the
dictionary, not colloquialisms or misuse of the English language. So
maybe you should check the dictionary before you start spouting about
collective ignorance.
Colloquialisms
> aside, the liberties we take with our (mis-)use of the English language
> are staggering. It's surprising we understand half of what's said or
> written. I try to bite my tongue when it comes to playing grammar and
> spelling cop on usenet, (and I'm not even close to perfect myself, and
> well aware of it) but there's a hell of a lot of people wandering around
> the planet that barely qualify as literate. The significance and
> implications of our collective ignorance are widely underestimated.- Hide quoted text -
Don Klipstein
>I've just had a Pro-life 25W spiral bulb fail in a spectacular way (very
>loud bang followed by lingering burnt smell). It's been fitted
>approximately 6 months.
I avoid 25 watt spirals, especially of brands that are neither "Big 3"
nor home center mainstays.
It appears to me that a big run of bad 25 watt spirals was made around
2000-2001 or so. I bought one of the Lights of America brand (and that
brand I often had trouble with) and 2 of the GE brand (GE is one of the
"Big 3" and normally does well). All 3 burned out in only a few hundred
operating hours, but quietly.
I have seen only a year or two ago 25 watt spirals at Walgreens, of a
brand that I cannot remember, that appeared to me to be of similar
vintage. So I am suspicious that there are businesses that bought some of
that boatload of 2000-2001 or whatever garbage and hope to make money
reselling it under different brands.
========================
As for CFLs failing with a bang: Sadly, that was somewhat normal.
Two ways for a CFL to make a loud pop and what the manufacturers have
done about it (or should be doing):
1. A usual screw base CFL with internal electronic ballast has a filter
capacitor after the rectifier. This capacitor has limited life
expectancy, especialy at elevated temperatures. It also contains a water
solution of electrolyte, since it is an electrlytic capacitor.
If this capacitor gets too hot, the electrolyte can boil and make the
capacitor burst. The capacitor's housing is normally designed to break
without producing shrapnel of the housing.
A few years ago, quite a few people were disturbed by CFLs going POW and
occaisionally dripping electrolyte. Usually, at least one of the
following is usually the case:
* The CFL was an off-brand one
* The CFL was operated in a higher temperature environment than the
manufacturer anticipated, often in a downlight or a small enclosed
fixture
* The capacitor was not as good as the CFL manufacturer thought
What manufacturers have done about this: They have gotten better at
using capacitors that are up to the task of CFL duty.
I'm sure there will still be some capacitors popping in the future, but
I am already hearing less about capacitors popping than I heard earlier
this decade.
2. The electronic ballast shorts and a wire or a part acts as a fuse,
sometimes with a loud pop or bang. Sometimes part of the ballast gets
scorched or discolored by smoke.
If the CFL is UL listed and production units conform to units tested by
UL, then the ballast and ballast housing materials are sufficiently flame
retardant for the CFL to be reasonably safe from starting a fire.
However, I hope the manufacturers are aware that a light bulb going out
with a bang, smoke output or getting a visible scorch mark in the process
appears scary and does not make good press. I would hope they now put in
fuses to make semiconductor failures/malfunctions leading to the CFL dying
less spectacularly.
I expect less scary failures from manufacturers that hope to still be in
the CFL business 10 years from now. I think "Big 3" (Philips, GE,
Sylvania) would want to avoid bad press, so I think they mostly make
better CFLs.
- Don Klipstein (d...@misty.com)
Don Klipstein
Oops, I was only calculating break-even points in whether cost of
operating the lamp is increased or decreased, without regard to
considering leaving-the-lamp-on causes some of the lamp's life to be
wasted in addition to the electricity being wasted.
I expect that adding consideration of wasting lamp operating life will
change break-even-time calculations and make break-even times shorter.
I expect this correction will only make break-even times slightly
shorter if electricity consumption during the life of the lamp (bulb)
costs a lot more than the lamp (bulb) does, which is often the case.
With lower wattage lamps, lamp cost becomes a more significant fraction
of the total cost, so this correction gets less minor.
- Don Klipstein (d...@misty.com)
Don Klipstein
Many dimmers don't let you achieve full brightness. The bulb may
randomly have longer-than-average life. Some "ordinary" bulbs are
long-life versions (with slightly less light output).
So I am not surprised.
- Don Klipstein (d...@misty.com)
colin...@hotmail.com
ahead is much brighter now. I amazed that this thread has not
degenerated into the usual flaming after a few posts - I can only
assume that a better class of person inhabits the a.h.r and uk.diy
threads. Best regards.
Grimly Curmudgeon
<see.my.s...@nowhere.null> saying something like:
>I also expect that is is a corruption of the original research that was
>looking at overall costs including lamp life and not just running costs.
>It has just that much of the detail has been lost in the Chinese
>whispers along the way and it has been reduced to an absurdity
>concerning just energy costs.
Ding!
The first time I read about this was nearly 30 years ago in New
Scientist, where the article made the point it was about overall
lifetime of the lamp and fittings, rather than electricity consumption.
In that original article there was bare mention at all of the start-up
energy cost, probably because it was insignificant. The piece
illustrated how the lifetime of a lamp was reduced by multiple starts,
and showed that it was more economical *at that time* to reduce the
number of starts, or once started, leave the lamp on for a while.
At that time, with the fluorescent lamps and fittings available and the
energy cost of the day, the break-even point was 20 minutes.
It's usenet, so it's not unknown for the ill-informed start an argument
without knowing the full story.
--
Dave
Grimly Curmudgeon
saying something like:
>
> Many dimmers don't let you achieve full brightness. The bulb may
>randomly have longer-than-average life. Some "ordinary" bulbs are
>long-life versions (with slightly less light output).
>
> So I am not surprised.
Good for you. Have a coconut.
--
Dave
Andrew Gabriel
"** Frank **" <noe...@xyz.net> writes:
>
> Not good for filament lamps too. In fact, too many switching operations are
Actually, regular mains filament lamps don't care about switching,
and it doesn't shorten their lives. (This might not apply to high
current and halogen lamps, for which I haven't seen figures.)
> not good for most things like computers, motors, TVs, etc. Switching
> transients (both switching on and off) could have many thousand volts and
> could draw 10x or more rated current across the device which could results
> in electrical, mechanical and thermo shock. Note many lamp failures are at
> the moment of switching either on or off.
That's why people think that switching shortens their life, but
it's a misunderstanding of what's happening. At the end of life,
lamps can continue operating for a few hours past the point where
they won't survive another switch-on. If you switch them off
during this period, they will blow at next switch-on, and in the
case of a lamp which isn't normally switched on for an hour or
more, it's pretty certain to blow at a switchon rather than whilst
running. However, this is independant of the number of times the
lamp has been switched on in the past and depends only on burning
hours. There are a number of applications where this effect can be
measured, such as continuously flashing signs which use regular
lamps, where life can be seen to depend on total burning hours
and not frequency of switching.
> My own experience is CFLs (and Circle Lines) are much more susceptible to
> switching than incandescent or the 4' and 8' fluorescent lamps. YMMV.
Assuming the most common fluorescent tube failure mode (electrode
emission mix all sputtered off), it depends on the design of the
control gear, and not much on the type of tube.
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
meow...@care2.com
> On Sun, 19 Aug 2007 08:34:27 -0700 someone who may be
> meow...@care2.com wrote this:-
> >> If you leave a filament bulb on, it will generally last much longer
> >> than if you turn it on and off.
> >
> >Do you have further information on this? I understood that avoidance
> >of switch-on surges gave only minimal extra life to GLS lamps, though
> >considerable extra life to halogens.
>
> It is the same with valves. Left on they can last for a very long
> time, turned on and off they can be unreliable. See Tommy Flowers,
> 1930s telecommunications equipment and the genesis of Colossus for
> the evidence.
But the 2 situations arent comparable afaics. Filament failure vs
oxdide contamination on a filament heater running only red hot.
NT
meow...@care2.com
meow...@care2.com
> On Sun, 19 Aug 2007 12:55:17 -0400 someone who may be willshak
> <will...@00hvc.rr.com> wrote this:-
> >I've had the spiral CFL bulbs installed in all of my formerly
> >incandescent bulb lamps, wherever the lamp took a regular bulb. The
> >exceptions are mini spots,
>
> There are now some good energy saving bulbs for these.
>
> >and decorative candelabra bulbs.
>
> While there is no equivalent of clear bulbs there are now a number
> of "candle" shaped energy saving bulbs that are very similar to
> pearl "candle" bulbs.
>
> It is becoming difficult to find an indoor application where there
> is not a suitable energy saving bulb.
CFL candle lamps will fit chandeliers etc, but imho they dont come
anywhere near being a replacement in visual terms. The appearance is
bulkier, ungainly, and they have no sparkle at all, unlike clear
filament candles. Also with chandeliers the splitting of colours
depends on a small light source, so use of CFLs does this no favours
either. Chandeliers are one app where filaments still rule. LV
filaments can at least gain a bit more efficiency over mains.
NT
meow...@care2.com
I would think it would make quite a big difference, but havent calced
it yet.
NT
Fred
"mm" <NOPSAM...@bigfoot.com> wrote in message
news:f5ufc3p8du9kmmqn4...@4ax.com...
> On Sun, 19 Aug 2007 02:54:32 +0100, John Rumm
> <see.my.s...@nowhere.null> wrote:
>
>>Fred wrote:
>>
>>> It's very much the issue! I recall stories as the poster had where the
>>> starting of a fluorescent tube was equivalent to 1/2 hour running.
>>
>>Analyse that statement logically and you will see it makes no sense...
>
> First off, it's true, and it may be the basis for the original post,
> so it's worth discussing.
>
> Secondly, I would say that it does make sense**, but it's not accurate
> and for someone who knows anything about the topic, it's not
> believable.
>
Many thanks for your support. I was trying to make the point, admittedly
badly, that in the 70's that there was the misapprehension that starting a
fluorescent tube took an inordinate amount of energy. I think we both agree
that this is very untrue.
Dan_Musicant
<d...@miniac.demon.co.uk> wrote:
:Nowadays domestic users of fluorescent tubes need not concern
:themselves too much, but "Best Practice" is "Best Practice". If you
:have a fitting that requires a lot of effort to get at (above the
:stairs say) it makes sense to get the most out of the tube.
:
:DG
Yes, DG has it right. The issue is NOT energy usage here but the life of
the bulb. Turning a CF on and off a lot DOES shorten the life. It may
have 10,000 hours MTBF, say, but if you turn it on and off 10,000 times
the life isn't apt to be 10,000 hours! I don't know if the bulbs have
gotten better that way, but what HAS been getting better is the cost of
the bulbs. They still aren't nearly as cheap as incandescents, but the
economy of the situation has incandescents out of the picture. I haven't
bought one in quite a few years.
If I only need a CF on for a few seconds, in my workroom, say, I
generally use a flashlight rather than turn on the overhead light for
10-15 seconds in order to find what I need. At the ceiling are two CF's,
and I don't want to wear them out. I've heard that nowadays the life is
only shortened maybe 5 minutes, but I suspect that's quite inaccurate. I
have had several CF's fail way before they were supposed to. There's a
circular one in my kitchen that would probably cost me over $10 to
place, and the one it replaced lasted maybe 10-15% of the supposed life
expectancy.
Dan
Dan_Musicant
Klipstein) wrote:
:"Program Start" - this is used in some CFLs. The bulb does not come on
:at all until a fraction of a second to about a second after power is
:applied, then turns on without blinking. It may have a "rapid fade-on"
:during a fraction of a second.
: This causes the least wear, and is often used in CFLs of Philips and
:Sylvania brands (and some others but I can't remember who and I have not
:tried them all).
Yes, IIRC Philips and Sylvania are among the handful of brands that I
have seen recommended, and consequently I have bought several. I have
had very spotty luck with off-brand CFL's lasting anything near the
advertized MTBF. In recent years I have seen some pretty good deals on
CFL's and bought some off brands anyway.
Dan
Don Klipstein
My suspicions:
1. Heat - see if it gets unduly warm where that light is. Rated life is
with ambient temperature 25 degrees C (77 degrees F).
Better ones should have only slight incidence of early failures if it
gets a fair amount warmer. But if you have an enclodure around it, try
removing the enclosure.
2. Was it a brand other than GE, Sylvania or Philips? Most "circline"
lamps that I have seen to go into screw sockets have been by Lights of
America so far, and I have had a disproportionate share of LOAs die young.
(However, I have only bought one LOA since 2001 so they may have
improved.) I have also generally experienced LOAs (as well as Maxlites)
to be a little dimmer than others of same claimed light output.
- Don Klipstein (d...@misty.com)
David Hansen
meow...@care2.com wrote this:-
>> >Do you have further information on this? I understood that avoidance
>> >of switch-on surges gave only minimal extra life to GLS lamps, though
>> >considerable extra life to halogens.
>>
>> It is the same with valves. Left on they can last for a very long
>> time, turned on and off they can be unreliable. See Tommy Flowers,
>> 1930s telecommunications equipment and the genesis of Colossus for
>> the evidence.
>
>But the 2 situations arent comparable afaics. Filament failure vs
>oxdide contamination on a filament heater running only red hot.
The comparison is one of whether leaving on equipment involving hot
bits of metal has advantages in terms of reliability. It does in
many fields, but this has to be weighed against the energy
consumption of doing so.