This morning I put two $20 Kill-a-Watt meters on two power strips with
a 100 W bulb screwed into one plug socket 80 cm from a "100 W equivalent"
23 W 10,000-hour Commercial Electric compact fluorescent with a 9-year
guarantee ($8.97 for 4 from Home Depot) and compared the outputs with
a Bunsen grease-spot photometer (a drop of oil on a piece of white paper :-)
Robert Bunsen (1811-1899) also invented the Bunsen burner. He was known as
an inept experimentalist with radical theories who isolated a foul-smelling
compound which he named cacodyl oxide and a whole series of related compounds
which turned out to be highly explosive. At one point, Bunsen accidentally
blew up his lab and was laid up in bed for a long time.
The grease spot disappeared (indicating equal illumination on both sides) when
the paper was 42.4 cm from the incandescent bulb, so it had (42.4/(80-42.4))^2
= 1.27 times the CF light output. After a minute or so, the 100 watt bulb
consumption dropped from 100 to 99 watts and the CF rose from 22 to 24, so
the CF was 99/(1.27x24) = 3.24 times more efficient, with 3.24 times more
lumens per watt.
After warmup, a "150 W equivalent" 42 W CF ($5.97 from Home Depot) used
35 watts and made the spot disappear 36.2 cm from the 100 W bulb when
it drew 98 watts, so it was (36.2/(80-36.2))^2 = 0.683 times brighter
than the CF, which was 98/(0.683x35) = 4.10 times more efficient.
Nick, this is much more useful than a lot of your pie in the sky
calculations. Well done.
BTW, if the CFs were in a pack of four, did you test for variations?
That would be useful information as well.
Also, did you check the lumen output either by using a standard
candle, or a photometer (perhaps one in a camera?). Incandescent
lamps dim with age, so using an older 100 watt lamp might have
affected the results.
This part may not have been vital - but it's the sort of thing I love to
I'm sure many of his calculations are useful to people who want to be
frugal - but I'm stunned that Nick managed this without a line of Basic
The numbers are interesting - the 23W CF really was approximately 23W but
the 42W CF was much less. I'm not at all surprised that the 23W bulbs
really aren't 100W-equivalent - typical marketing hype - but those results
are acceptable to me.
Don't CFs dim with age? Can we then expect the relative efficiency to
improve over time?
- also, keep in mind that the lumen ratings for
traditional fluorescents are the initial output,
but... it's _after_ 100 hours of burn in.
When first lit, they tend to be a modest amount
brighter than spec. There's a rapid 'burn off" (for
want of a better description, and then, at a nominal 100 hrs,
they stabilize (to a slight long term downward slope).
So if you used brand new CFLs, you're seeing a bit
of an artificial peaking...
Knowledge may be power, but communications is the key
No, not typically. Very few Gas Discharge lights dim with age unless
the excitation voltage drops significantly. The Gas doesn't wear out,
but the electronics that fire the voltage may.....
No, again not typically. Very few electronic systems get better as
the components age, so efficency should slowly drop a bit with age....
They why does my employer change out the fluorescent lighting about every 5
or 6 years even when they are all burning good? When they do, the rooms are
so much brighter. Makes a huge difference.
I'm betting the replacement was wasted. That's because the lamps were
changes individually as they failed. In other words, the average age of the
tubes they were changing out was far less than 5 years!
I have 14ea 48" tubes in my home, shop, and garage, and I'll bet the average
age is less than 2 years.
No, that is the reason to change out all the bulbs. At home we usually
change a bulb when it is gone. No big deal. In a factory or large office,
getting the lifts, scaffolding or whatever together, two employees +
overhead costs, they often find it cheaper just to sweep through an area and
change 100% of the bulbs at one time. That way, there are few, if any,
changes between, thus the average is still very high at 5 years.
If one bulb goes, all the others at that age are probably not far behind.
At work, we don't change the entire plant, but will do a section at a time.
When you have 20' ceilings, it is cheaper to change what you can get at once
the equipment is in place.
Agreed. The use of 'group relamping' (replacing all bulbs at once even
though many are still burning) has some advantages even though some bulbs
are replaced prematurely. Since bulb life is pretty predictable, most of
the replacements will not burn out too soon. All the bulbs are from the
same stock (usually), so it avoids mixing up different 'colors' such as
'cool white', 'softwhite' and 'daylight' in an area. And as you mentioned,
the 'overhead' of getting crew, tools, and equipment scheduled. If 'group
relamping' is done on a performance basis (for example, whenever 10% of the
bulbs have burned out), then large areas of burned out bulbs are usually
'Spot relamping' has the advantage that maximum use of each bulb is attained
and only a small quantity of replacements are needed to keep on hand at any
It is NOT out of the ordinary in commercial and institutional use, for
QUALITY tubes to last more than 3 years, and changing en mass before
failure is standard maintenance procedure. If maintenance needs to
change more than a few tubes before scheduled change-out, they either
modify the changeout schedule or find a new supplier!!!!!!!
5 years MIGHT be stretching it, but I know on office where in the last
5 years only 2 or 3 bulbs had been replaced (along with about 4
ballasts) before a mass relamping and reballasting took place.
Posted via a free Usenet account from http://www.teranews.com
I think more likely the phosphor degrades as the bulbs are used and the
replacement ones simply have full brightness.
There are many 4-foot fluorescent lamps now rated to last 20,000 or
24,000 hours, and that is with 3 operating hours per start. Expect a bit
more with one start per workday.
This means to me expect half to burn out in 5.5-6.5 years if used 10
hours a day 5 days a week and lasting no longer at 10 hours per start than
at 3 hours per start. Some installations will experience worse and some
will experience better.
Meanwhile, light output will decrease over the life of a fluorescent
lamp. It appears to me that a 32 watt T8 (1 inch diameter 4-footer) is
rated to be typically down about 5% in light output when 40% of the way
through its rated life, and I would expect about 10% down when getting
close to rated life. And it appears to me that typical phosphor
degradation is about double that for T12 (34 and 40 watt 1.5 inch
diameter) with older-tech phosphors.
I have known low-mercury ones to sometimes degrade faster due to the
mercury supply getting stuck in parts of the bulb from chemical reactions
and/or maybe mercury ions getting embedded into whatever, maybe in the
- Don Klipstein ( firstname.lastname@example.org)
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.