The 3500k tubes summed up
Sarah Austin
For those not familiar:
http://www.patmullins.com/img/colortempchartorig.jpg
an article about color temperature is at
http://www.mediacollege.com/lighting/colour/colour-temperature.html
I've been a big fan of "full spectrum" lighting indoors, since reading
"Health and Light" in the 70's.
Supposedly the best "full spectrum" lights are the Vita-Lites made by
Duro-Test. I've been using them indoor since the 70's too.
Since then I've been having some discussions in one of the usenet
lighting engineering groups, with some of the experts there.
It seems that the term "full spectrum" is frequently misused for sales
pitches.
And as far as the color they give out, it's very bluish, even though the
CRI ( Color Rendering Index ) can be pretty high and they illuminate
very nicely for indoors.
I was going along with "conventional wisdom" in this regard, with the
birds here too. That if "full spectrum" is supposed to be so good for
humans, it must be best for the birds too.
Then one of the myths was burst recently when I read the article at
http://www.birdsnways.com/wisdom/ww42ev.htm
Seems birds don't need "full spectrum" light for the UV, they just need
some decent quality light and enough hours of it per day.
At the same time, after discussions with the lighting engineers, they
pointed out that the color spectrum we should use, depends on what we
want to have in the home or office.
They also pointed out that if we wanted to get away from the bluish
lighting in the house, that the 3500k color temp is what they call
"bright white" and is whiter instead of bluish, yet not yellowish like
the lamps tha put out 2700k ( the most common in compact fluorescent )
offer.
So I started out by picking up a couple of compact fluorescents at Home
Depot, that are bright white. Ecosmart 159 452 which are 30 watts and
put out the equivalent of 125 watts each at a color temp of 3500k. (
about $9 each but very worth it! )
I put up a couple of them in the living room ceiling fixture and they
lit the place up in a very bright white. Very nice.
Soon Jennifer wanted a pair of them for her office, so I gave her those
two.
Then I did some shopping for some 48" T8 tubes with the same color temp,
and as high a CRI rating as I could find. I found these:
http://www.businesslights.com/tcp-xhl-32w-3500k-85-cri-t8-fluorescent-
lamp-case25-p-2060.html
TCP F32T8/835 3500K 85 CRI 3400 Lumens Fluorescent Lamp case/25
* On sale why quantities last
* 31032835XHL F32T8/835 Lamp Specifications
* 32 watts
* Color Temperature: 3500� Kelvin
* Color Rendering: 85 CRI
* MOL: 48 inches
* Initial Lumens: 3400
* Mean Lumens: 3200
* Average Rated Life: 24,000 hrs
* Minimum starting temperature: 10�C
* Low Mercury - TCLP Compliant
* Designed to operate on Instant Start ballasts
I ordered a case, which was the minimum order qty. Got them yesterday
and put them in a fixture over my work desk, replacing the bluish
colored Vita-Lites.
Amazing!
It lights the whole office up, with just 64 watts of lighting, like a
bright sunny day, and instead of the bluish cast of the Vita-Lites, with
a bright white color, a slight bit on the warm side, but not yellowish
like 2700k looks.
So I did a little experiment with the birds. Peko and Pearlie, our
tiels, have been hanging out near the open window in the daytime, which
now has the southern light of the fall sun coming in. It's actually like
these lamps, a bright white with a slight bit of yellow in it.
So I swapped out the Vita-Lites over Peko and Pearlie's cages with these
new 3500k ones. Immediately they moved over under those, instead of near
the sunny window. They stayed under them all afternoon.
So tonight I swapped them all for the 3500k ones. What a difference.
Now admittedly this is unscientific and it depends on what you like, but
for my office, this is a great color spectrum, lighting it up brightly
in bright white for only 64 watts of power. It's so bright that I almost
feel like I need sunglasses sometimes, it's amazing.
From what the guys in the lighting group said, these XHL tubes cost a
little more each, because they added special phosphors that make them
put out 10% more light ( the XHL apparently means extra high lumens? )
and it raised the CRI to 85 too, which is considered very good. ( a lot
of cheapo cool whites that are commonly used in offices, have a CRI of
around 65 )
And the concept of "full spectrum" at 5500k may be a flawed one. That's
the color of the blue sky on a clear day, BUT, the actual sunlight that
comes from the sun, is more of a bright white, with slight yellow
warmth. So do we really want bluish sky colored light cast into our
rooms, or the bright white of actual sunlight?
It's fascinating.
But I'd really recommend this for office settings and if your birds like
it better, for them too!
If any of you whom I know from RMSA, would like to stop by and see this,
just let me know. ( you too Mary! )
Don Klipstein
<To make the story shorter, happy experiences with F32T8 fluorescents
having 3500 K color temp. and claimed CRI typical or maybe slightly
highish for the upper common CRI type,
and had a better experience so far than with "full spectrum" lamps>
I do have just a couple minor comments on a bit here:
>From what the guys in the lighting group said, these XHL tubes cost a
>little more each, because they added special phosphors that make them
>put out 10% more light ( the XHL apparently means extra high lumens? )
>and it raised the CRI to 85 too, which is considered very good. ( a lot
>of cheapo cool whites that are commonly used in offices, have a CRI of
>around 65 )
CRI of 85 is usual to maybe slightly highish for the upper or "8" grade
of triphosphor T8 fluorescents. I consider 84, maybe 85 to be usual.
The lower or "7" grade of triphosphor T8 fluorescents has CRI around
76-77, maybe 78.
"Old tech cool white" (my words) does indeed have CRI of usually 62,
sometimes as high as 66.
>And the concept of "full spectrum" at 5500k may be a flawed one. That's
>the color of the blue sky on a clear day,
5500 K is typical of sunlight and skylight combined, if none of the sky
is obstructed, the sun is fairly high above the horizon and air quality is
good.
Sunlight alone varies widely. My experience in/near Philadelphia is
that in midday direct sunlight without light from blue sky has color temp.
usually 4100-4600 K, but near "high noon" close to summer solstice when
air quality is unusually good for summer in/near Philadelphia, I have seen
direct sunlight get into the 5000-5100 K range (estimated by eye in
comparison to reasonably known light sources).
Supposedly with ideal air quality at high noon at summer solstice in
Washington DC, direct sunlight achieves 5200 K. And supposedly direct
sunlight excluding light from blue sky with sun at zenith with ideal air
quality near sea level, the color temp. is 5400 K.
Light from blue sky varies in color temp. but is mostly in the range of
6500 to 20,000-plus K, with some significant bit of blue sky achieving
infinite color temp. or a hair more blue than infinite color temp.
One figure for "blue sky" is 7500 K, which is supposedly fairly typical
for light from blue sky (and excluding light reflected by the ground) on a
north-facing vertical surface near "high noon" on a cloudless day with
good air quality. Since this favors the brighter and less-blue light of
sky closer to the horizon, I consider "average blue sky with good air
quality" to have color temp. higher than 7500 K, possibly the 9300 K that
the P4 phosphor in black-and-white CRTs is normally supposed to achieve.
(And in the past several years, I have noticed cheap B&W TV sets to vary
in color temp. by "my eyeball estimate" from 7500 to 12,000 K.)
>BUT, the actual sunlight that comes from the sun, is more of a bright
>white, with slight yellow warmth. So do we really want bluish sky colored
>light cast into our rooms, or the bright white of actual sunlight?
Direct sunlight often appears slightly warmer than is usually perceived
from its actual color temp.
One reason is that human vision has a fairly well known desire for lower
color temp. when illumination level is lower and higher color temp. when
illumination level is higher, although this trend breaks down somewhat at
illumination levels below "candlelight" levels of a few lux or something
like that.
At illumination level of maybe 1,000 lux or more, fairly typical of
offices, classrooms and more-brightly-illuminated retail spaces, 4100 K
tends to "look good" to me.
Past maybe 5,000 or 6,000 lux, the 6000 K that I consider typical of
overcast sky appears to me to "look good".
6500 K - I still wonder where that came from, maybe skylight from closer
to horizon or average north window light with average partly cloudy sky
conditions or blue side of overcast sky -
appears to me to "look good" at 20,000-plus lux.
If illumination level only gets into the hundreds of lux, I like 3500 K
best. At 10's of lux, I find everything on the dim side, but in my
experience 2700 K is "usually least-bad" and good when illumination that
dim is desired. At around a lux to a few lux, the 1900 K typuical of
candle flames looks good to me when I want lighting that dim.
At fractional lux levels typical of brighter moonlight, I like mid-upper
3,000's K typical of moonlight.
At "dimmer moonlight" level of illumination in the .01 lux range or
less, I find photopic vision to not work well, and see mainly by scotopic
vision. At some point in that ballpark and dimmer, vision has little
sensation of color.
- Don Klipstein (d...@misty.com)
TKM
"Don Klipstein" <d...@manx.misty.com> wrote in message
news:slrnhfk2p...@manx.misty.com...
Don makes some excellent points. As lighting people, we tend to pay too
much attention to the spectral characteristics of a light source and then
debate whether or not it looks natural or "good" to our eyes and pay little
attention to the illumination level under which we make the evaluations.
Our evaluations and opinions change dramatically with illumination level.
So, as Don indicates, illumination level counts!
Indeed, it is just as important as the spectrum itself in my view. How many
have seen clear mercury lighting, HPS or even LPS at thousands of lux?
Color acceptability is substantially improved. Sunlight and skylight at
5500 Kelvins is visually "warm" coming from the sky at 100,000 lux, but
"cool" coming from a fluorescent tube at 400 lux.
A similar thing happens when we expect non-visual benefits from light.
Using a specially-colored fluorescent tube for birds, fish, amphibians or
other critters is only dealing with a small part of the job. It's the
"dose" of radiation or total energy that matters. The dose includes
providing the needed visible light spectrum, of course, plus infrared and UV
and it also includes duration of exposure, when that exposure is provided
(to handle circadian effects) and intensity (illumination level).
That all became apparent when the reports about "full-spectrum" lighting on
people appeared a few years ago. The only thing you could be sure of with
full-spectrum lamps is that they looked different and cost more. Unless you
handled the dose questions, they couldn't have anything but a minor effect
on health or well being unless, of course, you consider them some kind of
psychological treatment.
Terry McGowan
Andrew Gabriel
d...@manx.misty.com (Don Klipstein) writes:
>
> Direct sunlight often appears slightly warmer than is usually perceived
> from its actual color temp.
>
> One reason is that human vision has a fairly well known desire for lower
> color temp. when illumination level is lower and higher color temp. when
> illumination level is higher, although this trend breaks down somewhat at
> illumination levels below "candlelight" levels of a few lux or something
> like that.
>
> At illumination level of maybe 1,000 lux or more, fairly typical of
> offices, classrooms and more-brightly-illuminated retail spaces, 4100 K
> tends to "look good" to me.
>
> Past maybe 5,000 or 6,000 lux, the 6000 K that I consider typical of
> overcast sky appears to me to "look good".
>
> 6500 K - I still wonder where that came from, maybe skylight from closer
> to horizon or average north window light with average partly cloudy sky
> conditions or blue side of overcast sky -
> appears to me to "look good" at 20,000-plus lux.
>
> If illumination level only gets into the hundreds of lux, I like 3500 K
> best. At 10's of lux, I find everything on the dim side, but in my
> experience 2700 K is "usually least-bad" and good when illumination that
> dim is desired. At around a lux to a few lux, the 1900 K typuical of
> candle flames looks good to me when I want lighting that dim.
Fits Kruithof curve well.
http://en.wikipedia.org/wiki/Kruithof_curve
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]