There isn't going to be a formula for extracting spectral distribution from a kelvin color temperature. The color temperature is an expression of what color the light appears to our eyes. Since light is a mix, there are many different ways to mix light in different frequencies to get the same visible color. That's why you have a formula that will give you the one key ingredient without which you cannot make the end result, and that's it. The rest of the mix is somewhat negotiable. Your formula, as best as I can tell, identifies the one-most important frequency without which the color temperature used in the formula would not be possible..,. I'm pretty sure that the spectral distribution curves you sometimes get with a bulb you purchase are measured with a special lab tool called a "radiometer" or spectral radiometer... It uses some sort of sophisticated sensor to produce the graph you get... That device is really expensive....
Dude you are absolutely right and I am completely sorry. I was frustrated with the lack of answers I could find and annoyed because I felt like most of what you were telling me I already knew. I am ashamed of my outburst so again i'm really sorry and hopefully you can forgive me. However, this still does not seem to make sense to me: Bear with me on this again I will break my argument down because I claim that a black body radiator at 4400K does not produce white light. 1. Let us say I have a piece of sodium and something that will heat the metal to very high temperatures. 2. My goal is to produce light with a specific wavelength of 660nm. 3. I must find a relationship between Temp(K) and wavelength(nm) and so I use the formula: Peak Wavelength(m) x Temp(K) = a constant 2.898x10^-3 4. I rearrange my formula to make it easier Temp(K)= 2.898x10^-3 / Peak Wavelength 5. I plug in my desired wavelength of 6.6x10^-7 (660nm in meters for formula) Temp(K)= 2.898x10^-3 / 6.60x10^-7 Temp(K)= 4390K You are saying that light temp mimics the light emitted from a black body radiator at the same temp, right? So why then would a lightbulb advertised at 4400K produce bright white light when a piece of sodium, which is heated to 4390K, emits light with a wavelength of 660nm? Either lightbulb manufacturers know something I don't, their bulbs do not in fact follow the rules of black body radiation, or they go off of a different formula (which would explain the discrepancy). Probably, I am overlooking something, but that is why I am asking for an explanation in the first place. I am going to go back to the formula and triple check to make sure it is applicable and legitimate.
Wow! Thanks for the apology man. That's unusual online I came to CG from another site because of so many flames. Friends now? lol. Anyway, I am jones-ing for a cigarette SO bad, but I'd love to help you get this straightened out a bit as best as I can... Gimme 15 minutes or so and I'll start hacking up a long confusing reply on how I think it works lol... Also- I started a similar thread to try to tackle the same thing you are working on... Perhaps we can all work together? Check the links in my signature, and check back here too- I'll try to come up with a decent answer/response for you
OK so I went over this again and I think I have a more satisfactory answer for you. J There’s an excellent 1page or so article on this here: Learn About Light: Seeing Color: GE Commercial Lighting Products That you may also find very helpful. I’ll try to hit the important parts: You’re right there, and you’re wrong there. In reality- “white” light is a mix of other colors that come together to make light that our human eyes perceive to be “white” in color. You know- just like a rainbow separates the “white light” we see from the sun into a bunch of colors… Same concept really… Remember art class? “when dealing with light- the color white indicates the presence of all color, when painting white is considered to be the absence of all color, a blank canvass.” Take the chart below: See how the colors mix together and make white in the middle. That’s kinda what our eyes do. The rods, cones etc respond differently to different wavelengths, but the concept is the same. Think of it this way: instead of the 3 discs of color coming together, your eyes are sort of “counting” the nanometers, in a way. When you open your eyes and look at a light source and see the “color” of that light, your eyes are basically calculating “this amount of light intensity at 450nm, plus this amount of light intensity at 451nm, plus this amount of light intensity at 452nm. this amount of light intensity at 453nm” etc, and averaging it together in your brain to show you the “color” of light you “see” at that moment with your eyes... That is why it is possible to have a light source that looks white, but still has a peak wavelength somewhere else on the spectrum. Think about it, there’s no “wavelength in nanometers” for white light… Can’t be done. The sodium will catch fire violently early on. Sodium is very combustible. It reacts so strongly with only air that it is typically stored in mineral oil of some sort. That’s where you and I are disagreeing a bit… You see- your formula is a good sound scientific formula, but it only applies to blackbody radiation type experiments using a “near ideal blackbody radiator (a piece of carbon or something).” Check out this link: Blackbody Radiation It is a java applet that simulates an “ideal blackbody radiator” experiment. It’s basically pretending you are heating a block of carbon (not sodium) until it gives off light, and then the applet is telling you the temperature (like on a thermometer) of the blackbody, and the screen shows what a spectral output chart (same as for a light-bulb) would look like for the light being given off at that temperature. The thing like I said is that your formula is for blackbody radiation, so your answers will only be within a relevant range if the material you are using is a near ideal blackbody. Sodium is not one of these by definition… It will still produce light when stimulated correctly, but that’s why your numbers “won’t jive” is because although sodium “works” at producing light, it’s not a blackbody so it won’t “play by the rules” of your equation. Notice also I said “stimulated”. In a blackbody experiment, the photons are released through the excitement of the atoms (and electrons thereof) using heat. Heat doesn’t always need to be the factor that causes this… For example in a fluorescent tube the “excitement of the atoms and electrons” is achieved through the electrical current passed through the vaporized mercury inside.. Again one more way that things stray a bit- once again if the material used to produce light is not a blackbody and the method of excitement is not heat, then the equation won’t work L Anyway, back to the applet- I ran your numbers, and technically you are right, just not the way you were hoping. See pic: You can see your numbers do jive, but it has to be under those circumstances: blackbody radiator material used, and heat as an excitement method. At a real (like on a thermometer) temperature of 4,382 Degrees KELVIN, a piece of graphite will produce a mix of light, that appears white, and has a peak wavelength of 661.787nm… Go online and play with the app, and you can see the spectral output of the carbon change shape as you “turn up the heat.” Make sense now? If they had an applet like that for sodium it would work like this: 1) the screen would say “nothing” where the color curve should be. 2) when you drag the mouse to increase the heat, at some point the screen would say “the metal combusted/exploded rapidly” L What I’m saying is 2 things 1) when the material used to produce the light is a near ideal blackbody radiator, and the energy source is heat, then the Kelvin temperature of the material and the Kelvin “color” temperature of the lamp are the SAME. 2) If the 2 conditions mentioned above are not met, the “the numbers won’t jive” as you have already discovered… This is the problem I am trying to illustrate that you’re noticing too: the industry kinda does lie to us… You see, since the “numbers won’t jive” with your formula once you start using other materials like sodium, mercury, metal halide, plasma etc- so what they do is guess. They basically look at the light with their eyes, and “remember” that color. Then they basically heat a blackbody radiator until it “looks” the same to the eye and slap a label on it. Obviously they don’t heat a piece of carbon up in a lab every time they make a new light bulb but you get the pint. This whole business of color temperature came about when “Lord Kelvin” discovered that an iron rod always seemed to have similar hues of color at similar temperatures. The concept was realized before people even understood the concept of light in nanometers was fully hashed out by Einstein. Our little “spat” on GC is nothing compared to the huge pissing contest scientists had “back in the day” over the issue, so lets not take each other too personally lol. J Anyway- hopefully the question is answered in your mind by now, but here’s one more way of explaining it- Back in the day there was really only one kind of light for the home- incandescent. It worked fine because nobody gave a damn as long as they could see what they were doing. Nobody gave a damn about light in nanometers etc. Those lights are simple: use electrical resistance to generate heat, and use the heat to give off light. Tungsten is a near ideal black body radiator, so your formula does work in that case… Since an incandescent uses heat to produce light from tungsten, the temperature on the bulb is the temperature that the filament inside is heating up to. A 2700K incandescent light bulb is basically a 2700 degree Kelvin blackbody radiator experiment! Then came along the fluorescent lighting for homes etc. These bulbs did not have “color temperature” on the packaging they had “correlated color temperature” or “CCT” written on the packaging right before the temperature in degrees Kelvin. This is their way of saying: because this bulb doesn’t use true blackbody radiation to produce light, the equation won’t jive etc, but we did pick the closest match from figures you’re used to so you can make a decent visual comparison… Hope this helps J CHEERS!
Thank you weasel that does pretty much clear it up for me. I am curious what lux your own grow is at? For everyone who is still trying to learn more I would go check out weasel's thread on lights^^^.
i just got one thing to say, if hps are becoming obsolete and there are better light sources availible, how come large grow ops and greenhouses all still use the 1k hps bulbs? (this is refering to weed growing and non-weed growing greenhouses)
Without any research my best guess would be that our technology has not gotten to the point where a bulb better than hps is more economical. And actually how do you know that all greenhouses are in fact not switching to other bulbs like plasma for instance? For example, calculators were far more efficient at doing mathematical problems than the alternative (the human brain) when they came out, but they were just too expensive that they were not worth the investment for many. But just as some companies did buy calcs at insanely high prices, I bet that so to are big greenhouses using plasma systems because they can see their benefits... but idk maybe not, thats just my guess.
. well actually i dont know if all greenhouses are still using hps, but i do know for a fact that all greenhouses in my area do in fact use hps lighting as a supplemental light source
its all about the wattages.....as for the reason why most greenhouses still use HPS, at least. personally, im a CMH (ceramic metal halide) fan. its a full spectrum light, with a hefty about of both blue and red, giving off the appearance of being a white light. IMO, the plants absolutely love it, and have a more "outdoor grown" look than if they had been grown under a HPS. anyhow, the point i was making, is that while other lights may be more efficient at producing usable photons per unit of energy, few of these lights are readily available in wattages large enough to sufficiently illuminate a very large area. example, if youre using a 1000w HPS as supplemental greenhouse lighting, you may only have 1 light for every 36 sq. ft. (6x6), maybe even 64sq. ft. Hell, i know of people who use 1000w as primary lighting on similar footprints. but the CMH is only readily available up to 400w. plasma lights are probably fairly expensive compared to a HPS system, and may not be able to cover the same area currently, due to power restrictions.
I agree with what you say Amoril. Another reason I think HID is so widely used is because it can be vented with vented fixtures so easily. Being able to ventillate a bulb with air other than the surrounding growing area's air is probably a huge plus in such large commercial applications It's a big reason why I think ppl still prefer HID to cfl for small grows... CFLs don't make much heat individually but when they all add up, they can develop some hefty needs to keep the grow area cool enough. When cooltubes aren't an option, that equals lots of air exchange through a cabinet... I know you're talking about greenhouses here and not cabinets, but I'm rambling... As for the CMH- I think they're really cool. I checked out these: 25 Watt PAR38 Metal Halide I know they're small, but I'm looking into some side lighting for my tiny led cabinet. Even though they're only 25W, they are HID nonetheless... I'm just wondering what an outside temp of 175F listed for the bulb, how that would affect cabinet temps.... CHEERS
two things about those bulbs weasel first, i dont think the temps will be a problem. Id wager the surface of the cfls is nearly that hot as well....the lower wattage wont allow it to produce the heat fast enough to heat up much around the bulb though, IMO. for that matter, the 400w runs cooler than a 400w HPS bulb, so i really dont see that being a problem. the one red flag, no pun intended, is that the light is 3000k (the red version, get it?). Its my understanding that the spectral output of the mastercolor 3k is much less desirable than the 4k. I mention it, because the 4k bulb has both more energy in the blue peaks (440nm range), but it also puts out more energy in the 660nm range than the 3k bulb. if youre still in the midst of the upgrade, and havent committed to anything yet, id check these out : Philips MasterColor Ceramic Metal Halide ED-17 they come in 50, 70, 100, and 150w in both the 3k and 4k. theyre cheaper than what you linked, too.....but definitely stick with the Protected ones. its worth the extra $8 or whatever. even one of the 50-70w bulbs hung on the side with a reflector will kick ass in a small space, IMO.
Thank you amoril for that info, it is interesting that you say a 4000K bulb has better performance at 440nm than a 3000k bulb, that was a big part of why I started this thread. I am curious where you got your info, link?
Hey closetgrower. Philips MasterColor Ceramic Metal Halide ED-17 is the link- the charts are at the bottom of the page when you scroll down he's right too- the 4000k is way better looking on paper... CHEERS!
JD I need three questions answered before I can help. 1)How many flouros/flood lights do you have and how many watts does each one use? 2)What kind of bulbs are you using? Be as descriptive as possible 3)And, how many plants do you have/how old are they? As seedlings all plants dont need much light, however (and this is mainly based on information I have gathered and from my very limited experience as a grower) my understanding is that once they are about a week or two old they can handle a lot of light and it is best to supply them with as much of it as possible. Also, if your plants ever looked stretched and their stalks are paleish then you definitely need to give em more light, instinctively they are reaching to get a spot in the sun.
My cfls are 14 watts and someone told me the 65 watt flood lamps are useless even though they are bright and warm,
Hello GC! Every indoor op I have personally seen indoors uses both, 1-1000W HPS and 1-1000W MH. Everyone...Grew The NL#5xHaze, Jack H-Hybrid, White Widow and Kush and all of thme has spactacular results. http://forum.grasscity.com/advanced...6-close-but-how-close-38th-day-flowering.html Check it Frag Out!