uaerc

Hi

I dont have much choice to get specific bulbs in this part of the world.........so after a lot of research and findings the best i could pick from my market was the Osram 70w MH Powerstar HQI - TS DAYLIGHT...... specs below.

I want to use this for the Glossostigma project i am planning to start soon......tank size will be small - med. mostly in the range of 60x40x40..... Do you gurus out there think that this bulb will do the job? co2 injecting is available (DIY pressurised. FEx. setup), nutrients will be acquired from ADA.

CRI is 85 on the bulb with a 5200k ...... more specs in the attachment.......cheers.

Attached Thumbnails

 

Vasudeva

Hi, this is a great post i have learned a lot, but i am just bit confused on kelvins and nanometers. can i have a bulb with 10,000k and still have a spike in 650 nm? where do Kelvins and nanometers come together? are kelvens just measuring what color i see? thanks for the great informational post!

defdac

Yes, that is correct. To have cooler, bluer, colors (higher color temperature) the blue part of the spectrum needs to be relatively higher than that red spike though.

When you heat up an idealized piece of matieral known as a "black body" it will emit different colors depending upon it's temperature: http://en.wikipedia.org/wiki/Black_b...ody_simulators

The relation between kelvin and the spectral distribution colors (nm) is a bit complicated. You can calculate the kelvin from a spectrum, but not the other way around since many spectrums will generate the same color:
http://www.fourmilab.ch/documents/specrend/

Shad0w

Totally lost
I was always thought that color temperature (kelvin) based on the combination of R, G, B spectrum. Depending on intensity of each spectrum you have different kelvin.

Now If I understand you correctly, you can get the same kelvin with different R, G, B intensity combination. How could that be? By the way your link is way to heavy for me.

defdac

Just to be clear: A light source can only have *one* spectrum, not several.

But you're still somewhat correct - the spectrum in most fluorescents consists of three or more spikes in their spectrum. Most often one spike in the red part of the spectrum (R), one green (G) and one blue (B).

Each spike comes from one phosphor inside the tube and there are often three main phosphors in aquarium fluorescents, and that is why they often are referred to as "Triphosphor"-tubes.

The amount of each phosphor will as you have figured out tune the exact color of the lamp and from this color we can then deduce it's kelvin.

You can compare it with TV-screens: They also have three colors and out of these three colors they produce a range of human visible colors (the space of all the producable colors in the monitor/TV is called it's gamut, http://en.wikipedia.org/wiki/Gamut )

Good question. If all different lightsources/lamps in the world were made up of the exact same three RGB-spikes in their spectrum you're *almost* correct - then you would have a more direct mapping between the RGB-values and kelvin.

But lamps are not made up of the same RGB-spikes in the same part of the spectrum. There are many different blue phosphors for example: ZnS:Ag,Cl (460 nm spike) or BaMgAl10O17:Eu,Mn (450 nm spike).

But that's not all. One kelvin measure, for example 10000K, can have a "tint"! Everyone with cameras that produce RAW-files knows what I'm talking about. When you set the white balance of a RAW picture you have one kelvin-slider and one tint-slider (Adobe Camera RAW and Lightoom for instance). When you drag around the tint slider the picture will change it's hue to more green or more purple.

This shows that there are more than one RGB-setting for every kelvin measure.

defdac

(This also makes it painfully clear that one kelvin-measure says absolutely nothing about what color a lamp has.. All you know is in what area in the gamut the lamp will be, but not what exact color it will have: It can be greenish, white or purple for every single kelvin degree..)

Vasudeva

so if i have a fixture with four bulbs is it better to have 4 bulbs that spike equelly in in ech color or to have each bulb specialize in a diferent color?

davemonkey

Go with bulbs that have spikes in the Photosynthetic range, but you can get four very distinct bulbs that fit that to give you the balance you want. Chosing bulb "color" is a matter of trial and error (within reason) trying to find the look you want. The main thing is to get good quality bulbs with red and blue spikes at the proper wavelengths.

-Dave

Vasudeva

I have noticed that bulbs do not vary much in spikes, but in kevlins. So is the challenge to find that right "look" and geting the lights for your plants is not that hard?

defdac

Yes that is correct. Some bulbs are also better at pushing down foreground plants (Aquarelle/Aquastar comes to mind), and some bulbs produce leggy growth (GroLux comes to mind)

That said, you can calculate how much photosynthesis a bulb will produce and that varies a bit from bulb to bulb. It does not only depend on how the light are spread within the spectral distribution, but also how good the phosphors are converting electricity to light. For example GroLux have a spectrum that fits very will with the photosynthesis action spectrum, but is not able to push very hard at it (low blue and red intensity).

Most non-yellow/green triphosphor tubes will hit fairly hard at the chlorophyll and push down the plants, so then it's mostly matter of taste.

And since it is taste it would be good if the manufacturers had a better color definition than kelvin..