best way to get 6500K with screw-in

[Gints Klimanis]

Hi,

I'm looking for screw-in lightbulbs for ~5000K as well as 6500K. The 5000K are to match the lighting in the rest of my scene (well, which is my house) all over. Also, I'd like to setup some 6500K for viewing my computer monitors that are set to 6500K whitepoint.

For the computers, should I use a good compact fluorescent light and reduce the green spike with a gel ? Do I really need to bother reducing the green spike ?

[Marcus Marchesseault]

It seems to me that most compact fluorescents don't have the green spike as much as the old 4ft. industrial lights of yesteryear. For a crude test, I use the pinkish parts of my hands to judge if magenta is lacking. Some fluorescent worklights wipe out the pink/magenta in my skintone and I avoid those except for when I want lots of bluish light. Bulbs like that are great for moonlight because they provide some of the colorless appearance things lit only by the moon acquire. I believe some of the economy/environmentally friendly 4100K 4ft. bulbs probably have a big green spike. I think they get the watt to lumen ratio up high by using a lot of green. Green is apparently the most efficient color for converting the arc of the bulb into the visible spectrum.

If you don't find 6500K bulbs easily, your situation seems perfect for a bit of CTB gel. Bringing up the blue with maybe a 1/2 CTB should be enough. You don't need a vast amount of blue to cast that TV look over a darkened room.

[Gints Klimanis]

Thanks, Marcus. I'll inspect the offerings at naturallighting.com . So far, most of the spiral CFL's are comparable to the offerings from TCP .

There is some surprises with a CRI of 96 :

http://www.naturallighting.com/web/s...on=show_detail

I have some nice T8 32W 5000K fluorescent tubes with a CRI of 98. I wonder why there are so few products with 95+ CRI .

[Emre Safak]

You should not worry about viewing 6500K monitors in 5000K lighting. In fact, this is standard practice. Don't ask me why; see for yourself.

[Marcus Marchesseault]

I am not a lighting engineer, but I think the really high CRI needs more types of phosphor elements per bulb. Inexpensive bulbs use 3, but the better ones may use 5 phosphor colors. Also, the higher the CRI, the lower the lumens per watt in many circumstances. There is probably a diminishing return beyond a certain CRI.

[Richard Andrewski]

Originally Posted by Gints Klimanis

Thanks, Marcus. I'll inspect the offerings at naturallighting.com . So far, most of the spiral CFL's are comparable to the offerings from TCP .

There is some surprises with a CRI of 96 :

http://www.naturallighting.com/web/s...on=show_detail

I have some nice T8 32W 5000K fluorescent tubes with a CRI of 98. I wonder why there are so few products with 95+ CRI .

Most lamps sold are in the 80 CRI range for household use and use a normal, inexpensive and commonly available phosphor powder inside the tube. The powder, and the way it is mixed determines the color temperature. The higher CRI tubes use a "tri-phosphor" powder that is apparently only available from Japan (now anyway) and comes in three colors: red, green and blue. The proportion these high quality powders are mixed determines the color temperature. In general the daylight tubes have a much easier time reaching the higher CRI values of 90 and above. It is quite a trick to get a tungsten color tube (3200K for instance) to do 90 CRI though. By its very definition the color temperature of tungsten is not close to noon daylight sun (5600K) so this is much more tricky to get 90CRI or above in those color temperatures. And remember what CRI means--its the comparison of a light source to daylight and the ability to properly render colors is considered highest in the area of noon daylight or 5600K. The warmer color temperatures below this will have more red in them so are biased away from true rendering anyway. Tricky--but not impossible. But I am wary of manufacturers that claim they have more than 90CRI on a 3200K tube though.

Another problem with these higher quality tubes is that the lumen value drops when using the tri-phosphor powder (over the regular, cheaper powder). So some other tricks have to be used to keep the lumen value of the tube up in the high CRI type.

[Gints Klimanis]

Richard, Thanks for the details on fluorescent tubes. I haven't seen any claims of 90 CRI for tubes in the sub-3500K range. Philips is selling T12 48" ColorTone 5000K tubes at Home Depot in the $3-6 range, depending on whether you get a low price on a case or buy them individually. As we all would want, a higher CRI for CFLs makes a more compelling case for them to displace incandescent bulbs.

[Richard Andrewski]

Originally Posted by Gints Klimanis

Richard, Thanks for the details on fluorescent tubes. I haven't seen any claims of 90 CRI for tubes in the sub-3500K range. Philips is selling T12 48" ColorTone 5000K tubes at Home Depot in the $3-6 range, depending on whether you get a low price on a case or buy them individually. As we all would want, a higher CRI for CFLs makes a more compelling case for them to displace incandescent bulbs.

I see claims on that all the time. One guy said in a thread recently that a tungsten lamp was inherently 100CRI. Not sure how that would be possible at 3200K. Bias toward red can't be more like 5600K daylight which is the benchmark for color rendering. Kino Flo claims 95CRI on their True Match lamps for all color temperatures.

I am having 55w lamps manufactured right now for my company and I can tell you it is very challenging to get over a CRI of 88 in that color range. I think many make claims they can't really back up. Its really hard for end users to verify the claims too. The test equipment (known as an integrating sphere) to check it is really expensive. This kind of equipment is normally only found in a manufacturer of light bulbs. So unless you have a friend at a bulb factory (like me--haha) its really hard to tell if something is what it claims.

[Gints Klimanis]

>tungsten lamp was inherently 100CRI

I see 100 CRI on spec sheets for incandescent lights. Unlike you, I don't have the expertise or an expert friend to verify that. Thank you for all of your help.

[Emre Safak]

Tungstens are incandescent too. Incandescent simply means they emit light when they get hot (which hot lights certainly do). CRI is measured relative to the correlated color temperature (CCT) of the source. A perfect CRI means that for a given CCT the source renders color identically to a blackbody.

[Richard Andrewski]

Originally Posted by Emre Safak

Tungstens are incandescent too. Incandescent simply means they emit light when they get hot (which hot lights certainly do). CRI is measured relative to the correlated color temperature (CCT) of the source. A perfect CRI means that for a given CCT the source renders color identically to a blackbody.

That's true. If you think about it as that your looking for a "warm white" and not a true white then it would be 100 (or pretty close) CRI.