PDA

View Full Version : Sun's Intrinsic Color



George
2005-Mar-29, 05:55 PM
Yep. I done did it. :) [Edit....not really, but maybe]

The true color of the sun is as plain as the nose on his face (but not the tip of the nose) :o :) ...

http://img33.exs.cx/img33/6964/mattsnose5bu.th.jpg (http://img33.exs.cx/my.php?loc=img33&image=mattsnose5bu.jpg)

That is my assistant - Matt. I call him son, too. :)

I'm reluctant to say it. But, the Sun is green in color. This is the color of our sun if we could observe it from above our atmosphere, and, with a reduced intesity to allow our eyes normal color determining vision. [The sun is too bright to allow color determination in space without some attenuation.]

This is the first result of an experiment taking direct sunlight, dispersing the colors, adjusting the levels of each color to match the Sun's visible spectrum in space, then recombining the light into one spot to observe the net color.

This may be the first ever experiment using the Sun's light for color determination. [Does this remind anyone of Spitzer's announcement? :) ]

The following is a rough look at what was done.
http://img50.exs.cx/img50/3146/SPACCSunsColor.jpg

Of course, we have had several threads on this... here's one (http://www.badastronomy.com/phpBB/viewtopic.php?t=9583&)

[Edit. To explain the above (without going to the above link), I am adding the below as an improved explanation of the illustration above. (also as per request)

A - Inlet slit to allow sunlight (ground level light) into SPACC.
[ This actually became a roll of toilet paper with two wooden plugs. The inlet slit was about 10 mm and the final slit about 1 or 2 mm. (I hope Evan will speak to me again as I know he would do a high calibre job on all this. My efforts are almost comical - but, "better-faster-cheaper" (2 out of 3 ain't bad)]

B - White light (for simplicity, assume flat spectrum)
[Of course, the spectral irradiance (light intensity as a function of wavelength) at the surface is not flat]

C - Prism 1
[F2 (flint) 25mm equilateral prism with 5 place accuracy (according to prominent supplier) on the index of refraction for many wavelengths. Interpolation was used to align all the wavelengths as needed (thanks to A Hundred Pardons for the math assistance (real name still withheld till he approves the use of his real false name (ok Candy, it was funnier the first time :) )

D - Typical dispersion
[The exit angle is shown wrong, but the illustration was easier to draw this way. I used an entrance angle of 45 deg. to the plane of prism 1. I used a compas attached to the horizontal 3/4" rod (stake) and rotated it 15 deg to allow a 30 deg. sun angle (15 + 30 = 45). A 45 deg. incident angle produces a 5.84 degree dispersion (red to blue). A 60 deg. angle yields only 3.73 deg. dispersion angle.]

E - Mask (formerly -Template): This is a pre-atmosphere irradiance curve cut into a dark plate. The outlet light is now of the same proportion as sunlight above our atmosphere. [The mask profile is not accurate due to the direct sunlight not having a flat spectral irradiance, and the dispersion angle of the blue end is much more than at the red end. However, the principle is the same.]

F - Prism 2 (to converge the light)
[I tried it without this prism in preliminary tests. However, the convex lens could not reconverge the light as the F2 prism disperses more than the convex glass]

G - Converging light. Note that the blue light, for instance, is wider than the other colors. This presents a minor problem as all the light must be hongenized to correctly represent the Sun's "true" color.

H - Converging lens to homogenize the light. Something a little fanzier may be required. A simple convex lens will help but complete homogenization would be only at a tiny point, I think.
[Surprisingly, I was quite pleased with how the double convex lens handled the recombining of the colors. By adjusting it along two axis, I was able to obtain a white light (mask excluded). Further movement would reveal blue at one end, reverse movement would shift blue to the other side. Therefore, I concluded there is a sweet spot which puts the colors all together (of course, in between the two movements just stated)]

I - "Gray" card or something to reveal the - Suns True Color. I wonder what it is but I am hoping it is cyan or blue just to stir things up.
[I probably should get a graycard but I only used white paper in the shadow of a black plastic wall]

X - "marks the spot"!

1 - This is the above and below (sea level) irradicance plot representation. (The upper curve is the intensity of light for each color radiated by the Sun as observed in space, the lower is what we see down below after the atmosphere has "bleached" it.)

2 - If the upper curve can be used to "cut-out" light from a pure flat spectrum by the amount the sun does not emit, then the light seen will be a true representation of the color of the sun since each color will be in proportion to the suns actual color spectrum.
[Of course, adjustments had to be made in the mask to deal with the our atmospheric effects]

3- Mask (Template). The net light that passes through from a pure flat spectrum source, that is aligned in wavelength with this template, should be a true representation of the light radiated from the Sun. This light, when recombined equally into one area, will reveal the true color of our star, hopefully.
[Again, this is the main principle but adjustments had to be made to deal with the real variations in intensity at different wavelengths]
[end of edit]


The green color is independent of any color seen while viewing the sun from below the atmosphere. If green is correct, green is the color the sun has been for billions of years and will be green for billions more.

The exerpiment was done at the Windmill Observatory (just learned there already is a George Observatory (drat :wink: )

http://img33.exs.cx/img33/5323/windmillobservatory9by.th.jpg (http://img33.exs.cx/my.php?loc=img33&image=windmillobservatory9by.jpg)

Close-up of the preliminary prism arrangement...
http://img33.exs.cx/img33/5061/prismsetup2fg.th.jpg (http://img33.exs.cx/my.php?loc=img33&image=prismsetup2fg.jpg)

Unfortunately, my confidence is not high. The wind was about 25 knots and Matt and I struggled trying to get all lined-up properly. He held the inlet slit steady and I held the template and adjusted the double convex lens to produce the combined homogenized color.

We have not been able to do numerous tests to confirm this color. This is needed as the template was angled more than it should have been.

In spite of this limited confidence, I wanted ya'll to hear the latest on this as ya'll have been the fire behind my rekindled astronomy interests. In this regard, special thanks to the BA. [yes, that is a sucking sound you hear as I hate being debunked by the best :) ]

John Dlugosz
2005-Mar-29, 07:03 PM
How do you view the resulting spot of color? If it's projected onto a white sheet of paper in ambient light, you can say that it's green-ER then the ambient light. But I could argue that the ambient light is blue, since I'm currently adapted to indoor lighting. Or the ambient light is yellow, since I'm standing in the shade.

I mean, just because you created a spot of light doesn't mean we know unambiguiously what color we will perceve it as. It depends on context.

--John

George
2005-Mar-29, 07:53 PM
How do you view the resulting spot of color? If it's projected onto a white sheet of paper in ambient light, you can say that it's green-ER then the ambient light. But I could argue that the ambient light is blue, since I'm currently adapted to indoor lighting. Or the ambient light is yellow, since I'm standing in the shade.

I mean, just because you created a spot of light doesn't mean we know unambiguiously what color we will perceve it as. It depends on context.

--John

Surprisngly, the color was outstandingly green. It was not sorta greenish but very definetly green. However, your point is one of our considerations, as it has been mentioned before. I will be using different papers to reflect the light to help minimize any ambiguity. The reason for the cover and black plastic was to minimize the blue sky as a false reference.

As it turns out, direct sunlight spectral irradiance is quite flat. This not only allows for easier experiment work but it also provides a good reference light itself as it is quite "white".

The toughest question you can hit me with is likely.....If the Sun is green, how come white objects in space look white and not green? :roll:

Ed White (in his first American walk in space) looks white (his suit) not green. White rocket stages look white not green. Even if the intensity of the light produces white, the edges of white objects should be dim enough to reveal green.

Maybe some NASA buff can help me with this one. Is there much color correction by NASA??? Would they adjust to make white objects white as their reference?

George
2005-Mar-29, 09:50 PM
For those interested in the nitty gritty....

I used the irradiance data from hyperphysics web site. I later decided I better compare it with a more recognized standard. The following shows both, along with a blackbody plot.

http://img96.exs.cx/img96/5834/irradiancecomparisons6ow.th.jpg (http://img96.exs.cx/my.php?loc=img96&image=irradiancecomparisons6ow.jp g)

They are in agreement nicely but with small variation in the light green, yellow and orange portion of the spectrum.

Again, the experiment was to replicate this spectral irradiance profile. This was done simply by taking the known spectral irradiance as seen down here and adjust it to the levels shown on the above plot. Admittedly, I used the hyperphysics surface irradiance plot for my reference. It was taken with a solar altitude of 30deg. We matched this same altitude in hopes it would yield the best result. Skies were clear and we had had recent rains to minimize atmospheric issues (hopefully).

Notice how the peak is in the blue (near green). Also notice that the peaks (or humps) exceed blackbody power levels. I am guessing this is due to corona emissions. Does anyone know or want to guess also?

Jorge
2005-Mar-29, 10:33 PM
maybe a dump question... can someone translate this in a:
17y old kid who has never ahd phisics/biologie/chemstry/... understandable language?

I'm interested in this but well itw a bit to much complicated data to take in for me

George
2005-Mar-29, 11:02 PM
maybe a dump question... can someone translate this in a:
17y old kid who has never ahd phisics/biologie/chemstry/... understandable language?

I'm interested in this but well itw a bit to much complicated data to take in for me

Sure, and I am not near as smart as you might think I am. I am definetly only an amateur.

Everything that radiates light, emitts photons of different wavelengths.

Take something simple that is extremely limited in different wavelength emissions - a laser. If you were to plot it's intensity (y-axis) against it's wavelength, you would see a vertical line rather than the bell shaped curve (as in the illustration in my prior post).

If you could add other color lasers together and shine them all onto one spot, you would see a net color. This color would be based on the different color lasers you chose and it would also be determined by the intensity of each laser.

The sun, like most stars, tend to emit visible light at all colors. However, each color has it's own intensity level. The Sun happens to emit the strongest amount of light (intensity) in the blue and green portions of the spectrum. This can be seen on the graphs I gave. The middle portioin of blue is around a light wavelength of 470 nanometers, green is about 520 nm. The peak intensity of sunlight (in space) is under 500 nm (blue).

Howevr, the peak intensity does not determine the color of an object. You must combine all the colors and at the proper level for each color. This is replicating the spectral irradiance. In other words, we reproduced the sun's visible light output, matching color and intensity properly. This was accomplished by using direct sunlight as we have it on the surface, blocking some of the colors proportionately to get each intensity correct, then recombining all of it into one spot. Ta-duh....green was the result of round one. :)

[edit, BTW....Matt just turned 17 this month, so you two have a little in common. What color is your nose? :wink: ]

Jorge
2005-Mar-29, 11:13 PM
thank you,
that explained it nicely for me :)

I'm very interested in Astronomy the pitty is i'm not that smart(well am but more in the computer kind of stuff, no real use in astronomy)

also a problem i have is that finding book on the mather is rahter hard.
most books i find are well basic and cover the same things, there are no more advance(read books that go indept) i can find.

most interesting books i have are english(not that that is a problem i prefer english anyway) but they where hard to come by.

hmmz i seem to be getting off topic.

well thanks for the explanation

George
2005-Mar-29, 11:22 PM
thank you,
that explained it nicely for me :)
Thanks. I needed to do a better job at explaining it anyway.


also a problem i have is that finding book on the mather is rahter hard. most books i find are well basic and cover the same things, there are no more advance(read books that go indept) i can find.
I tend to buy the bargain books at the fancy book stores.

However, I have bought a few "The Complete Idiot's Guide to....(whatever)". These keep things simple, yet are quite informative. The BA's book is fun stuff, too, and even easier to read.

The last Idiot's Guide was on "The Sun" (Dr. Pasachoff). He, no surprise, did discuss color but no mention of it's true color.

TravisM
2005-Mar-30, 01:13 PM
Did anyone mention that the light from the sun had to pass through our atmosphere?
How then can you claim that's the color of the sun above the atmosphere.

George
2005-Mar-30, 02:43 PM
Did anyone mention that the light from the sun had to pass through our atmosphere?
How then can you claim that's the color of the sun above the atmosphere.
If you had a bowl of rain water and salt, then you removed all the salt from the water, you would have pure rain water. [There's bound to be a much better example, but chem lab was 30 yrs. ago.] The experiment takes existing sunlight then adjusts for all the atmospheric effects. Since astronomers have determined the intensity of sunlight for each wavelength as observed in space, and we know the values for direct sunlight as observed down here, we can make the adjustments necessary. Newton could easily have done it, but he had no spectral irradiance curves to work with.

Nevertheless, it is still a great question and it is critical to the experiment.

Take a look at the following plot. This is the spectral irradiance of the sun as observed at the surface when the Sun is 30 deg. above the horizon (AM2).

http://img102.exs.cx/img102/1282/surfirradiance6nz.th.jpg (http://img102.exs.cx/my.php?loc=img102&image=surfirradiance6nz.jpg)

If the plot was flat (one power level only), then I could simply disperse this perfect reference light through a calibrated prism, pass it through a cut-out plate (or gate - I need a good name for it) where the shape of the cut-out is exactly the shape of the spectral irradiance plot of the Sun as it radiates each wavelength in space. This would produce the exact amount of blue, green, yellow, etc. that would qualify it do be a true replication of the Sun's radiance. Combining this light would produce a net color - the intrinsic color of the sun.

However, the plot is not perfectly flat. Therefore, I had to simply adjust for the differences. For instance, lets use a reference intensity of 1200 w/m^2. All levels would simply have to adjust to 1200. Notice the intensity of light at 405nm is about 1/2 of the needed intensity. In other words, we are receiving 1/2 of the amount of blue light at 405nm that we need to conduct the experiment properly. To acquire 2x this amount, I simply double the slot opening at 405nm on the spectral irradiance plot. This gives me a 1200 level of 405nm blue light.

This might be a little confussing for some. This idea might help. I could have made two gates, one to adjust the direct surface sunlight to make all levels the same intensity, then another gate to "cut-out" the light levels to the levels matching the sun's profile in space, AMO (Air Mass 0). However, I chose to combine the two into one cookie cutting light gate. It does not look at all like the almost bell-shaped curve, but rather more like a mountain slope sloping from red up to the top at violet. This is because blue light is so diminished from the sun as observed here on the surface, I had to compensate by allowing more blue light (and others) in through the gate.

The irradiance plot above is the plot I used. I have had trouble trying to locate other surface irradiance plots. I found one from Finland (I think) which seemed to be almost identical. It also showed, essentially, the same profile but at different intensities for different solar altitudes (Air masses). I wish I had an actual surface spectral irradiance profile for San Antonio for any day I choose, this would further the integrity of this work. I suspect, however, the variation will not alter the net color.

George
2005-Mar-30, 03:07 PM
I might as well post the cookie cutting light gate. (A better term for it would be nice - any ideas?)

http://img198.exs.cx/img198/3544/template3ht.th.jpg (http://img198.exs.cx/my.php?loc=img198&image=template3ht.jpg)

The width of this matches the width of the light through the prism.

Notice the two vertical yellow tick marks. This is the location of yellow light in the spectrum. Yellow light, ironicially, is extremely narrow in the spectrum (@580nm). This makes for a nice reference location for the gate in the light stream. (The real light gate uses white lines to help see the yellow light.)

You might also notice just how far to the right yellow is. This is because the higher frequency lights (green, blue, violet) disperse from a prism at a greater angle than the low freq. light.

The top portion of the gate is the result of compensating for atmospheric effects plus the actual profile of light (spectral irradiance) of the sun as seen in space.

[Edit: Let me add that you can take a quick look at the last two images. Our eyes work from about 400nm to 700nm. Notice just how flat this range is in the first figure (prior post). As a result, the Sun is white as we see it down here (but not at sunrise and sunset, of course). Now notice how much more blue and green have to be added to make up for the losses due to our atmosphere. No surprise, it will not look white. However, I still am a little surprised with green. Possibly the strong green plus the addition of blue and yellow (also creating green) is the answer.)

Romanus
2005-Mar-30, 04:19 PM
Even if the peak emission is in the green, there is still the question about why there are almost no lone stars (with the exception of Beta Librae (Zubeneschamali) according to James Kaler, and that only to some people) that look green to all observers. For instance, an B star like Rigel or Regulus has a peak emission well into the violet and UV range, yet they don't appear violet to the eye.

Part of my source:
http://curious.astro.cornell.edu/question.php?number=14

TravisM
2005-Mar-30, 04:27 PM
That hits it.

And I'd call the cookie cutter light thinggy a Mask.

And it's red in the evening/morning because it's travelling through more of the atmosphere. Check the website that's attached to this board... Phil does a great work up on this.

W.F. Tomba
2005-Mar-30, 07:04 PM
Very interesting!

Is the use of actual sunlight in this experiment just for convenience? If you had an artificial pure white light source, it seems you could do the same thing and actually save a step.

I think I understand pretty well what you've done now that you've explained it, but the top half of the diagram at the beginning of the thread still confuses me. I don't know what step 2 is supposed to show, and the diagram seems to indicate that you have simply assumed a perfectly flat surface irradiance curve, when your explanation indicates that you did not. (The curve in step 3 looks identical to what I assume is the above-the-atmosphere spectral irradiance curve in step 1.)

If you plan to use that diagram to explain this project to more people, I would suggest reworking it, perhaps labeling the parts of the figures.

And in addition to white spacesuits and rocket stages, why isn't the Moon green? (Or is it, as seen from space?)

George
2005-Mar-30, 07:08 PM
Even if the peak emission is in the green, there is still the question about why there are almost no lone stars (with the exception of Beta Librae (Zubeneschamali) according to James Kaler, and that only to some people) that look green to all observers. For instance, an B star like Rigel or Regulus has a peak emission well into the violet and UV range, yet they don't appear violet to the eye.

Part of my source:
http://curious.astro.cornell.edu/question.php?number=14
You bring up a great point.

Ironic is it not? :) Since there are "no green stars" in the minds of most, this certainly adds color to this topic. (sorry, I can only go so long without puns. :) )

Actually, I am pleased to hear of Beta Librae. I did not know of any potential green ones.

For argument sake, there are reasons for not seeing green. If you consider images taken from land based scopes, the atmosphere does a major number on color. However, since hotter bluer stars produce so much more blue light, they should look more blue regardless. These stars are brighter which improves their number in images, I presume.

If we could get a "true color" image of a sister star to the sun (same temp. and composition) it should look green, assuming green proves to be the sun's intrinsic color. It would be nice to see astronomers take an interest in this idea and have Hubble take a color calibrated look at several solar twins.

This brings up another point, just how color calibrated can they make the Hubble? Can they be confident each exposure time is near perfect for each color in order to feel confident enough to claim a true color rendering? I have assumed the idea was to be reasonably close only, since color to the eye is not all that critical to science. Hopefully, the BA will jump in here and comment on some of this stuff as he seems well versed in Hubble's capabilities.

Another bit of information would be to get solar twin star's spectral irradiance. If some scattering ocurrs in open space, we would see it in the spectral break-down.

Hopefully, I can get some more tests done soon to be more confident in a color determination. If it is green, it will be all the more fun as, indeed, it would add fuel to the fire, color to the argument, invoke Irish sentiment (btw, the first announcement went out just prior to St. Patricks day :) - what about it, Eroica? Do you think it's green! :) )

[edit: I failed to try your Cornell reference. Nice. It makes some sense, but....time will tell]

George
2005-Mar-30, 08:06 PM
Is the use of actual sunlight in this experiment just for convenience? If you had an artificial pure white light source, it seems you could do the same thing and actually save a step.
If it's spectral irradiance was flat for all colors, it would be great! However, I think it does not exist. Bulb mfgs. are more interested in "natural" light which is surface irradiance (not that flat), or in efficency where they have three sharp peaks of light radiating (red, green and blue) to make white. Please don't assume I speak for them as I only have read a little on lamp outputs.



I think I understand pretty well what you've done now that you've explained it, but the top half of the diagram at the beginning of the thread still confuses me. I don't know what step 2 is supposed to show, and the diagram seems to indicate that you have simply assumed a perfectly flat surface irradiance curve, when your explanation indicates that you did not.
My hope was to simplify the example to make it more understandable. Obviously this did not eschew obfuscation. :) Sorry. A pure light source (as I drew in 2) would make all this easier to grasp.


(The curve in step 3 looks identical to what I assume is the above-the-atmosphere spectral irradiance curve in step 1.)
Yes. That is the idea as it represents the full color spread & intensity of the sun. Therefore, duplicating this will provide a net color. However, this step 3 is also an over-simplification. Since blue light disperses more than the longer wavelengths, the curve is streched as noted in the last illustration above (the "Mask").



If you plan to use that diagram to explain this project to more people, I would suggest reworking it, perhaps labeling the parts of the figures.
I did provide a link which provides those details. Hmmmm, maybe I should provide on second thought.


And in addition to white spacesuits and rocket stages, why isn't the Moon green? (Or is it, as seen from space?)
The moon material is rather dark. Astronauts on the moon reportedly did not use their visors as it was too dark for them. I do not know much of lunar material, but I feel fairly sure it would not be considered a good reflector of light where each visible wavelength is reflected evenly. If it was a good reflector, and the Sun is green, we wouldn't be doing this experiment since it a green sun would be obvious. So, the quest has begun in earnest. :)

George
2005-Mar-30, 08:36 PM
That hits it.

And I'd call the cookie cutter light thinggy a Mask.
Mask. Yes. Thanks, I like that much better. Template or light gate are not great terms, and "star gate" is already taken. :)


And it's red in the evening/morning because it's travelling through more of the atmosphere. Check the website that's attached to this board... Phil does a great work up on this.
Is this it?.... blue skies (http://www.badastronomy.com/bad/misc/blue_sky.html)

Very nicely done (no surprise).

Maybe two things I could possibly improve....

When a stream of photons of all different colors comes into the atmosphere, the red, orange, yellow, and even green ones tend to get through unimpeded.
Green will scatter about 3x more than red. Blue, however, is at least equal in strength to green before entering our atmosphere, but blue scatters even more than green. Essentially, his point is still correct.

The other interesting thing about Rayleigh scattering is it's dependence on particle sizes. Most know the scattering varies as the 4th power of light's frequency, but the particle sizes can influence scattering strength by as much as the 6th power! The upper atmosphere may have an advantage due to the smaller particle sizes even though density is less. Unfortunately, I must admit to be not that knowledgeable on it, but it would make for an interesting discussion.

The key to this experiment is knowing the two spectral irradiances - in space and at the observation site. I am confident with the space plot but less confident in the surface plot. These two curves determine the mask shape.

W.F. Tomba
2005-Mar-30, 09:49 PM
(The curve in step 3 looks identical to what I assume is the above-the-atmosphere spectral irradiance curve in step 1.)
Yes. That is the idea as it represents the full color spread & intensity of the sun. Therefore, duplicating this will provide a net color. However, this step 3 is also an over-simplification. Since blue light disperses more than the longer wavelengths, the curve is streched as noted in the last illustration above (the "Mask").
OK, then I'm confused again. You just wrote

The key to this experiment is knowing the two spectral irradiances - in space and at the observation site. I am confident with the space plot but less confident in the surface plot. These two curves determine the mask shape.
If the two curves determine the mask shape, how can the mask be identical to one of them? That would mean that the other one is irrelevant to the mask shape.

Jorge
2005-Mar-30, 09:50 PM
i isn't it "stargate" in one word? if so that would leave "star gate" open ;)

btw: i haven't red BA's book(i know shame on me :oops: ) but i haven't seen it yet in any shop here, only online shops but they don't delever where i live.

George
2005-Mar-30, 10:53 PM
i isn't it "stargate" in one word? if so that would leave "star gate" open ;)

btw: i haven't red BA's book(i know shame on me :oops: ) but i haven't seen it yet in any shop here, only online shops but they don't delever where i live.

You might want to PM the BA on this. He likely has a solution. Surely, you do get mail, right? A book should be no problem.

Jorge
2005-Mar-30, 11:00 PM
i isn't it "stargate" in one word? if so that would leave "star gate" open ;)

btw: i haven't red BA's book(i know shame on me :oops: ) but i haven't seen it yet in any shop here, only online shops but they don't delever where i live.

You might want to PM the BA on this. He likely has a solution. Surely, you do get mail, right? A book should be no problem.

haven't though of that :) might do that once i got some money again.
maybe i'll ask my parent for my end of shool year report card :)

George
2005-Mar-30, 11:24 PM
OK, then I'm confused again. You just wrote

The key to this experiment is knowing the two spectral irradiances - in space and at the observation site. I am confident with the space plot but less confident in the surface plot. These two curves determine the mask shape.
If the two curves determine the mask shape, how can the mask be identical to one of them? That would mean that the other one is irrelevant to the mask shape.
It is a little confussing.

The goal is to reproduce the exact spectral irradiance of the sun. Spectral irradiance is the intensity of sunlight at each individual wavelength. The spectral irradiance of the sun is shown in a prior illustration above. It looks sorta bell shaped but with a few jags here and there.

If I had a perfect linear spectrum of light (wavelengths equally spaced) that was also perfectly flat (same intensity for each wavelength), then I could simply use a mask that was shaped the same as the profile found in the spectral irradiance of the sun as seen in space. This was easy to illustrate and it was suppose to get the idea across quickly so as not to loose too much interest on the reader. Of course, the real world requires more tweaking.

The confusing part stems from the experimental method to get to that point. Since I am using prisms, the dispersion is not linear. This means the mask will be stretched to allow for the greater dispersion of the higher frequency light. Also, the mask height for each wavelength must be adjusted for the lack of intensity as seen, especially, in the blue and violet sections.

Let me know if it is still confussing. I think I've gotten a little to comfortable with it. :-?

Ironically, it was because the irradiance plot was originally confussing to me, that I came up with this idea. Several sites will show the y-axis as a derivative of wavelength, yet others do not with the same graph. I hate it when they make a simple graph complicated. Having been around here a while, I've had to stir-up a little calculus now and then. So, I was bothered by these graphs with their derivative plots. I suddenly realized that if you integrated the curve, you would get the total irradiance (power) of the sun. This helped explain the deriviative in the y-axis. This further opened my eyes to see that a simple copy of the curve into a matching wavelength stream of light would reproduce not only the power (if the intensities were also the same) but also the net color. The rest, hopefully, will be respectable history later. :)

W.F. Tomba
2005-Mar-31, 01:26 AM
The goal is to reproduce the exact spectral irradiance of the sun. Spectral irradiance is the intensity of sunlight at each individual wavelength. The spectral irradiance of the sun is show in a prior illustration above. It looks sorta bell shaped but with a few jags here and there.

If I had a perfect linear spectrum of light (wavelengths equally spaced) that was also perfectly flat (same intensity for each wavelength), then I could simply use a mask that was shaped the same as the profile found in the spectral irradiance of the sun as seen in space. This was easy to illustrate and it was suppose to get the idea across quickly so as not to loose too much interest on the reader. Of course, the real world requires more tweaking.

The confusing part stems from the experimental method to get to that point. Since I am using prsims, the dispersion is not linear. This means the mask will be stretched to allow for the greater dispersion of the higher frequency light. Also, the mask height for each wavelength must be adjusted for the lack of intensity as seen, especially, in the blue and violet sections.
(bolding mine)

OK, I think I see what was confusing me. I was trying to find a representation of that last part in the diagram, but it's not there. The diagram represents how you would do it if you had a perfectly linear, perfectly flat spectrum to begin with, rather than how the experiment was actually done. The use of natural sunlight as your light source adds an extra step that's not represented in the diagram. I was looking for that extra step in the diagram and not finding it.

George
2005-Mar-31, 01:35 AM
The confusing part stems from the experimental method to get to that point. Since I am using prsims, the dispersion is not linear. This means the mask will be stretched to allow for the greater dispersion of the higher frequency light. Also, the mask height for each wavelength must be adjusted for the lack of intensity as seen, especially, in the blue and violet sections.
(bolding mine)

OK, I think I see what was confusing me. I was trying to find a representation of that last part in the diagram, but it's not there. The diagram represents how you would do it if you had a perfectly linear, perfectly flat spectrum to begin with, rather than how the experiment was actually done.
I think ya got it. :) The math to do the corrections, fortunately for me, is fairly easy and straight forward once you see it in a spread sheet.


The use of natural sunlight as your light source adds an extra step that's not represented in the diagram. I was looking for that extra step in the diagram and not finding it.
It does add more work in the math and making of the mask, however, another step is not necessary. The bell shaped mask shown in the original diagram is, obviously, not representative of the real mask needed to correct for the non-linear prism dispersion and the non-flat sunlight we have down here.

W.F. Tomba
2005-Mar-31, 01:38 AM
It does add more work in the math and making of the mask, however, another step is not necessary. The bell shaped mask shown in the original diagram is, obviously, not representative of the real mask needed to correct for the non-linear prism dispersion and the non-flat sunlight we have down here.
I meant an extra conceptual step. The mask actually combines three different adjustments into one physical step.

George
2005-Mar-31, 01:49 AM
It does add more work in the math and making of the mask, however, another step is not necessary. The bell shaped mask shown in the original diagram is, obviously, not representative of the real mask needed to correct for the non-linear prism dispersion and the non-flat sunlight we have down here.
I meant an extra conceptual step. The mask actually combines three different adjustments into one physical step.
I suppose you're right. The mask y-axis represents the spectral irradiance in space adjusted by the non-flatness of direct sunlight received at the surface. The x-axis is adjusted for wavelength to match the non-linear dispersion of the prsim from violet to red.

The x-axis problem is why I invested in a couple of quality prisms of known index of refraction for each wavelength.

There is still a chance I can keep all this under $300. :)

John Dlugosz
2005-Mar-31, 05:24 PM
Ed White (in his first American walk in space) looks white (his suit) not green.

Whoever made the print (or tuned-in the video broadcast) would "fix" it without thinking too much about it.

Maybe he'd contemplate: well, using indoor film outdoors looks blue, using outdoor film indoors looks orange, so now I know that using ourdoor film in space looks green. Maybe they'll come up with space film.

--John

John Dlugosz
2005-Mar-31, 05:32 PM
You must combine all the colors and at the proper level for each color. This is replicating the spectral irradiance.

I understand that.

So what did the other guys figure differently to get "peachy-pink" (http://www.badastronomy.com/phpBB/viewtopic.php?t=9583&highlight=peachy+pink&)?

George
2005-Mar-31, 05:41 PM
You must combine all the colors and at the proper level for each color. This is replicating the spectral irradiance.

I understand that.

So what did the other guys figure differently to get "peachy-pink" (http://www.badastronomy.com/phpBB/viewtopic.php?t=9583&highlight=peachy+pink&)?

Chromaticity. (I understand it. :-? ) (edit- oops, I thought I had said..."I don't understand it"). They plotted blackbody temperature points along their color chromaticiy chart. They used 5780K for the Sun. I have no idea how accurate the chromaticity chart is to real vision which is what makes this experiment using real light-in-the-eye important. It (this experiment) is more of the real McCoy.

A Thousand Pardons
2005-Mar-31, 06:17 PM
(thanks to A Hundred Pardons for the math assistance (real name still withheld till he approves the use of his real false name (ok Candy, it was funnier the first time :) )
O that was what that was for :)

Approved

John Dlugosz
2005-Mar-31, 07:24 PM
Chromaticity. (I understand it. :-? ) They plotted blackbody temperature points along their color chromaticiy chart. They used 5780K for the Sun. I have no idea how accurate the chromaticity chart is to real vision which is what makes this experiment using real light-in-the-eye important. It (this experiment) is more of the real McCoy.

I agree that black-body radiation never looks green. So, the difference between the BB curve and the two sources of real sun spectral data you showed on your graph may be the key: it departs from BB with a blue/green peak. The blue cancels out the pink, so we perceive green.

Do you plan on publishing in a magazine or something? At the very least you should put that on its own web page.

--John

John Dlugosz
2005-Mar-31, 07:37 PM
I just thought of something.

You should measure the spectrum of the final spot and make sure it really agrees with the N-L spectrum you were trying for. Hopefully you can get someone interested enough to loan you a really good one.

--John

George
2005-Mar-31, 08:41 PM
(thanks to A Hundred Pardons for the math assistance (real name still withheld till he approves the use of his real false name (ok Candy, it was funnier the first time :) )
O that was what that was for :)

Approved

Ok. Thanks. :)

George
2005-Mar-31, 09:11 PM
Chromaticity. (I understand it. :-? ) They plotted blackbody temperature points along their color chromaticiy chart. They used 5780K for the Sun. I have no idea how accurate the chromaticity chart is to real vision which is what makes this experiment using real light-in-the-eye important. It (this experiment) is more of the real McCoy.

I agree that black-body radiation never looks green.
I really don't know if this is so. I meant to say "I don't understand it".


So, the difference between the BB curve and the two sources of real sun spectral data you showed on your graph may be the key: it departs from BB with a blue/green peak. The blue cancels out the pink, so we perceive green.
Maybe, maybe not. It seems like prognostication only. I do not know how they support their understanding of their color determination. Have they actually observed near blackbody objects at those temperatures match their prediction?

Their view is in contrast to an article I read about 2 years ago. The author's understanding of how the eye functions and his knowledge of colors caused him to claim the sun is green. Astronomy article but my search showed it must not have been. If anyone can log into Sky & Telescope and search for "green and sun" (or something similar, I would appreciate it. I would like to email him. ]

[edit: I did not properly address your point. The variation between bb and the actual radiance of the sun is noteworthy but I can not say for sure if it would make a peachy pink object turn green. My point questions the validity of a peachy pink sun prediction. Maybe some perfect blackbody at 5850K object would still look green if viewed in space.]


Do you plan on publishing in a magazine or something? At the very least you should put that on its own web page.
--John

Well, once I have conducted enough tests and confidence is high, I might do something besides BABB. My wish would be to get some professional interested enough to take this and run with it. Otherwise, I will probably consider all the advice ya'll give me. I don't mind doing something elaborate, formal (possibly even grandiloquent, albeit, highly out of character for me :) ), but (with the exceptions of several patents), I have never done any science publication.

George
2005-Mar-31, 09:25 PM
I just thought of something.

You should measure the spectrum of the final spot and make sure it really agrees with the N-L spectrum you were trying for. Hopefully you can get someone interested enough to loan you a really good one.

--John
Cool idea. If my work is done properly, the result should present an irradiance curve almost identical to that which is seen in space (which is the whole point of the experiment). That would be the icing on the cake and a great way to validate further. I wonder what you would get with plotting the spectral irradiance curve of a peachy pink object? :) That may be a bit unfair as the same color can be seen using more than one spectral radiance profile.

I may end up investing in some unit that will do it. I've looked at several during these last few months. This is another reason I want a professional involved. I would think most atmospheric physics departments have several units.

If I did decide to splurge for one, anyone want to give a recommendation on which one? It must give an accurate power reading for each wavelength from 400nm to 700nm.

Jorge
2005-Mar-31, 09:45 PM
why don't you start by writing to a local astronomy magazine, if your lucky they might publish something or place a metion it somewhere.

and then maybe some other ppl will come in and help?

George
2005-Apr-01, 04:31 AM
why don't you start by writing to a local astronomy magazine, if your lucky they might publish something or place a metion it somewhere.

and then maybe some other ppl will come in and help?

Publishing should get some attention. I don't know just how that works, as I've only done one magazine article in my life, but it was just a history article on curbing.

If no pro wants to run with it, I will have to do much better at this so my confidence is high before advancing from here. Green is not a final result at this point. Of course, I might just go to the media and force them to come here. :)

Late this afternoon I made more measurements to refine the set-up. I also recomibined the spectrum without the mask and was able to get a yellow-white dot of light. This is a good sign the concept works. Many more masks need to be calculated to allow for minor variations. If they all produce a net of one color, I will be happy indeed.

Jorge
2005-Apr-01, 10:05 AM
well i'll be away fro the computer for a week, i'll look forard to reading in ths thread when i get back. best of luck

John Dlugosz
2005-Apr-01, 04:16 PM
Another link to look at:
http://www.vendian.org/mncharity/dir3/starcolor/.

In correction, it looks like the first site I listed did indeed use the solar spectrum, not the BB curve. His graph has different endpoints, but eyeballing it it appears to match your data.

As for spectrometers, my information on them is ten years old, and I'm sure they are much better now. They have small hand-held units for the printing industry.

Meanwhile, as a control, prepare another mask that is supposed to generate a black-body curve. You can do so for various temperatures. Make sure the result is the well-known observed color seen by metal workers etc. That is, red-hot-white-blue but never any sign of green or peach.

That will help verify that your mask works properly. Also, have someone else calculate the mask from first principles and see if it turns out the same as yours.

--John

George
2005-Apr-01, 06:12 PM
Another link to look at:
http://www.vendian.org/mncharity/dir3/starcolor/.

In correction, it looks like the first site I listed did indeed use the solar spectrum, not the BB curve. His graph has different endpoints, but eyeballing it it appears to match your data.

It is hard to tell if there are slight differences because I use a 5850K bb reference. They are using a 5780K comparison. Their plot (orange one) seems to have more orange and red than the Neckel Lab or hyperphysics data, but the difference may be due to the aforementioned reason.


Meanwhile, as a control, prepare another mask that is supposed to generate a black-body curve. You can do so for various temperatures. Make sure the result is the well-known observed color seen by metal workers etc. That is, red-hot-white-blue but never any sign of green or peach.
I have not found anything useable regarding actual observed color from blackbody objects. Not much gets heated near these temperatures. Lamp mfgs. have some sites showing colors in this range. IIRC, they favor a blue sun. :)

One problem stems from the lack of the existance of true blackbody objects. A hole in a furnace became the first real blackbody object. [I still can't understand why a furnace hole would have all proper intensities at the specific wavelengths when the light still comes from a non-blackbody emitter (the furnace). :-? ]


That will help verify that your mask works properly. What color should it be? If they have not compared their computer results with actual observations for verification, how confident can we be.


Also, have someone else calculate the mask from first principles and see if it turns out the same as yours.
Ahhh, I like this idea, too. Are you volunteering? :) Would you like the data and the procedure I used?

Grey
2005-Apr-01, 06:50 PM
Also, have someone else calculate the mask from first principles and see if it turns out the same as yours.

Ahhh, I like this idea, too. Are you volunteering? :) Would you like the data and the procedure I used?
I'd be willing to take a shot at it. Send me the details, if you have a chance.

George
2005-Apr-01, 08:16 PM
Excellent. Multiple heads are better than one. :)

Here is the data set I have used so far..... here (http://hyperphysics.phy-astr.gsu.edu/hbase/vision/solirrad.html)

There are two sets, one for above our atmosphere (AM0) and one for on our surface with a solar altitude of 30 deg. (AM2).

The purpose is to try and reproduce the AMO graph by adjusting the AM2 data.

(All data is w/m^2 except wavelengths)

For instance, look at 500nm. AM0=1942 and AM2=1165. I used a datum of 800. The peak AMO point is 2074. 38.57% of 2074 reduces to the 800 level I desire. I want all the AM0 values to be reduced proportionately to maintain the same profile. Therefore, to reduce the 1942 level proportionately, I reduced 1942 by the 38.57% and obtained 749.1.

If all the light from the sun on the surface was exactly at the 800 level, we would be done, almost. Since it is not, we must adjust accordingly. To obtain 800 from the AM2 level of 1165, multiply by 68.67%. This means the actual light level at 500nm is a bit too high and must be reduced by that amount.

Multiplying the prior 749.1 x 68.67% produces 514.4. This would be our data point to represent the mask at 500nm, but we need one more step. The transmittance of the prsim at 500nm is 87.38% which adjusts the 514.4 level up to 588.72.


Lambda Trans.
400 87.00%
410 87.04%
420 87.08%
430 87.11%
440 87.15%
450 87.19%
460 87.23%
470 87.26%
480 87.30%
490 87.34%
500 87.38%
510 87.41%
520 87.45%
530 87.49%
540 87.53%
550 87.56%
560 87.60%
570 87.64%
580 87.68%
590 87.71%
600 87.75%
610 87.79%
620 87.83%
630 87.86%
640 87.90%
650 87.94%
660 87.98%
670 88.01%
680 88.05%
690 88.09%
700 88.13%


This will get us the y-axis mask point.

To obtain the matching x-axis point, we must use the index of refraction amount to know where 500nm will end up. The following the the n value for the prism I have.

F2 Index of Refraction.

Lambda n
400 1.65388
410 1.64789
420 1.645379
430 1.64334
440 1.64104
450 1.63882
460 1.63676
470 1.63486
480 1.63310
490 1.631464
500 1.62994
510 1.62851
520 1.62716
530 1.62591
540 1.62476
550 1.62367
560 1.62271
570 1.62176
580 1.62077
590 1.61983
600 1.61901
610 1.61823
620 1.61748
630 1.61676
640 1.61607
650 1.61542
660 1.6148
670 1.61418
680 1.6136
690 1.61309
700 1.61259


[I am rushing this due to work issues, hope this is understandable]

I appreciate any who are interested in this project. A collective BABBer effort resulting in future recognition would be very cool.

[The transmittance is actually based on the use of two F2 prisms which is why it may seem less than you would expect]

[Edit: Nope. After digging up the data, transmittance is based on only one F2 prism. ]

George
2005-Apr-01, 09:30 PM
Oh, one more thing. Use 45 deg. for your incident angle into the equilateral prism. This should yield the same mask shapes for all of us (don't I hope :) ). Snells Law must be used to get your x-axis point with the n values above.

George
2005-Apr-04, 01:47 PM
I'm back and not in a rush mode.

The procedure above may have been confusing. Shall I clarify?

George
2005-Apr-04, 04:33 PM
I've discovered that the larger secondary prism was too small to accomodate the full width of the light passing through the mask. Therefore, "green" may represent my lack of experimental prowess more than it does the sun's intrinsic color. :-? Further tests to follow.

A Thousand Pardons
2005-Apr-04, 07:22 PM
So what did the other guys figure differently to get "peachy-pink" (http://www.badastronomy.com/phpBB/viewtopic.php?t=9583&highlight=peachy+pink&)?


I've discovered that the larger secondary prism was too small to accomodate the full width of the light passing through the mask. Therefore, "green" may represent my lack of experimental prowess more than it does the sun's intrinsic color. :-? Further tests to follow.

Don't forget the last time, when someone thought the color of the universe was green (http://www.badastronomy.com/phpBB/viewtopic.php?t=404), and it turned out to be beige (http://www.badastronomy.com/phpBB/viewtopic.php?p=8847&#8847) :)

George
2005-Apr-04, 07:35 PM
Don't forget the last time, when someone thought the color of the universe was green (http://www.badastronomy.com/phpBB/viewtopic.php?t=404), and it turned out to be beige (http://www.badastronomy.com/phpBB/viewtopic.php?p=8847&#8847) :)

:) . I remember. That one was kinda odd. They wanted to tell us the average color of the universe but failed to consider they knew not the color of our own star. :)

That's another reason why determining the color of the sun is so important. If we don't know it, how can we be confident we know the color of the objects out there which are 25 to 35 magnitudes dimmer (not counting the tough ones). :)

[Edit: you refrenced to a ToSeek post and a secret command.....let me practice it...... <strike>yellow</strike>. Ok, I give up. How did he do it? #-o ]

A Thousand Pardons
2005-Apr-04, 08:23 PM
[Edit: you refrenced to a ToSeek post and a secret command.....let me practice it...... <strike>yellow</strike>. Ok, I give up. How did he do it? #-o ]
Back when html was allowed, I think

George
2005-Apr-04, 08:54 PM
[Edit: you refrenced to a ToSeek post and a secret command.....let me practice it...... <strike>yellow</strike>. Ok, I give up. How did he do it? #-o ]
Back when html was allowed, I think

Ok, but it still is working for ToSeek. Oh well, you still get the picture of my opinion of a yellow sun. :)

A Thousand Pardons
2005-Apr-05, 09:44 AM
[Edit: you refrenced to a ToSeek post and a secret command.....let me practice it...... <strike>yellow</strike>. Ok, I give up. How did he do it? #-o ]
Back when html was allowed, I think

Ok, but it still is working for ToSeek.
Yes, I think the software allows html for the older posts, which were posted when it was allowed.

Grey
2005-Apr-05, 04:18 PM
I'm back and not in a rush mode.

The procedure above may have been confusing. Shall I clarify?
I think I understood it, I just haven't had a chance to take a crack at the numbers. I'm afraid I'm not likely to have a break until the weekend at this point, but I will get to it.

George
2005-Apr-05, 06:54 PM
I think I understood it, I just haven't had a chance to take a crack at the numbers. I'm afraid I'm not likely to have a break until the weekend at this point, but I will get to it.
Great and no hurry. I have ordered some BK7 convex lenses to allow greater mask sizing. (I won't use the secondary F2 lens for de-dispersion.)


And in addition to white spacesuits and rocket stages, why isn't the Moon green? (Or is it, as seen from space?)
I've played with this a little.

What color moon would you like? :)
Maroonish?... Earth rise Apollo 10 (http://www.hq.nasa.gov/office/pao/History/alsj/a410/as10-27-3890HR.jpg)
Light Green?.... Apollo 11 shot (http://www.hq.nasa.gov/office/pao/History/alsj/a11/as11-44-6626HR.jpg)
Heavier green?.... Apollo 14 shot (http://www.hq.nasa.gov/office/pao/History/alsj/a14/as14-67-9379HR.jpg)
Green and blue combo colors?... combo (http://www.hq.nasa.gov/office/pao/History/alsj/a12/as12-46-6854HR.jpg)
Pinkish peach?....Apollo 14 Alan Shepard with Flag (http://www.hq.nasa.gov/office/pao/History/alsj/a14/as14-66-9232HR.jpg) (Thanks Ed Mitchell for that one :roll: :wink: )
Bluish moon?.... Apollo 14 Ed Mitchell with same flag (http://www.hq.nasa.gov/office/pao/History/alsj/a14/as14-66-9233HR.jpg)

These last two say a lot since it was, likely, the same camera with the same flag taken in sequence but with different final colors.

Color changes due to scanning are also likely.

Here's one for you ladies....Pinkish?... Apollo 14 (http://www.hq.nasa.gov/office/pao/History/alsj/a14/AS14-67-9389HR.jpg)

There are more bluish images of lunar soil and reflections, if I had to pick one, than the other colors. However, gray is the most predominant still.

I wish I knew more about space photography so I could pin down any color bias from the sun. It seems pretty tricky at this point.

Jorge
2005-Apr-09, 10:12 PM
lol, no baby blue moon :)

George
2005-Apr-11, 02:04 AM
Don't count baby blue out, Jorge. :)

These illustrations might also prove helpful in understanding the method of the madness.

http://img190.echo.cx/img190/9245/suncolorexp15iz.th.jpg (http://img190.echo.cx/my.php?image=suncolorexp15iz.jpg)
1) Sunlight from a narrow slit. (Surface sunlight)
2) Prism with known index of refraction for each wavelength.
3) Dispersion of the colors after refraction.
4) Mask before entering light stream
5) Mask in light stream showing how each color might align. (Not accurate in this illustration.)
6) Yellow alignment markings. The yellow light is quite narrow in the spectrum. These marks indicate where the yellow light portion of the spectrum must pass through the mask, thereby, aligning the mask to all the colors properly.


http://img190.echo.cx/img190/7969/suncolorexp28fo.th.jpg (http://img190.echo.cx/my.php?image=suncolorexp28fo.jpg)
A) Mask in light stream
B) Those yellow alignment marks again
C) The resulting light allowed to pass through the mask.
D) One of two convex lenses to converge the light into one spot. (2nd not shown).
E) Converging light with proper amounts of each color.
F) I don't really like Fs.
X) X marks the spot of the revealing true color of the sun.

This method is an attempt to restore the sunlight received at the surface to it's original levels of intensity for each color. This is accomplished by allowing proportionately more of each color, attenuated by our atmosphere, through a mask. The net result should be a restored spectrum in accordance to the suns spectral irradiance as observed in space (but at a much lower intensity).

The net result....the intrinsic color of your host star - Sol.

The Suns at bottom of illustration serve as possible true colors. Which sun do you like? :)

[I wonder what color the sunspots are....really? :wink: Maybe later. ]

George
2005-May-17, 11:34 PM
:)

Thought I'd keep this active for any interested. A great deal of credit goes to you.

To further the solar quest, out of the blue, I called A&M's physics dep't. and was happy to learn the name of their astronomer - Don Carona. He was super nice and gave me (plus my wife and aggie daughter) a tour of A&M's domes. He encouraged me and he was helpful.

The net result is I am, likely, giving a seminar at Stephen F. Austin early this fall on this subject. I offered a live demonstration along with a Power Point presentation.

Robert Reeves has just published his digital astrophotography book and he gave me, after I requested some of his pictures, his presentation disk which has beautiful images. He spoke last friday at the club.

[btw, he started his program by stating he graduated High School a few blocks away the year of the pilot for Star Trek, and as he was giving his speech (last Friday], he said his fellow HS grad (Peter Weller, aka Robocop) was staring in the showing of the last Star Trek episode.]

I am still open to ya'll's help. For instance....

A catchy slogan to inspire the nation would be nice. :)


[b][i]Quest for the Color of the Sun!
Catch the color of the Sun
The Color of the Sun
[just wait till we get the color and then make the slogan. ]
Emblazon the blazzing Sun
Sun - The Movie [somewhat optimistic future slogan]

You can see why I ask your help. :-?

umop ap!sdn
2005-May-18, 05:11 PM
Continuing the discussion from another thread:

If you don't mind, I would like to move this discussion to the neighboring thread ( "The Sun's Intrinsic Color") so others might comment. "The Sun's Intrinsic Color".
Okay, but I haven't had time to read this whole thread so if I say something silly that's already been covered, that's why. :lol:


If the color is as intense as I first saw it, background color won't be an issue. A green laser, for instance, looks green all the time, I would assume.
Oh, okay... well that's worth a try I guess. :D



Whereas if you had a mask that had the Sun's observed pre-atmosphere spectrum curve, then that spectral curve would get multiplied by the light coming into your instrument, so your eyes would be comparing (post atmospheric sunlight * pre atmospheric sunlight) with (post atmospheric sunlight) thus giving you the true color within our limited Earthbound color space. :)
This is interesting but new to me. Are you suggesting a pre-instrument (SPACC) mask to adjust the input into the SPACC?
I'm not sure what a SPACC is. If you don't have a flat spectrum background reference, then making a dot of light that differs from the background in the same way that the intended spectrum would differ from a perfectly white light source is the next best thing. :)


If you make a mask in the shape of a pre-atm curve, you will get a lot of imbalance in the ratios of long to short wavelengths since the longer wave light is not reduced significantly by our atmosphere.
But if your pre-atm curve has so many percent of peak intensity at 450nm, then your spot will have that same percentage of daylight intensity at 450nm. You'd have the true color plotted within the spectral distribution of daylight. ;)


I would assume if we made a mask that would produce a flat spectrum, we would have the ultimate background lighting. This could be done but I hope it won't be necessary.

Are you suggesting this for background lighting?
That'd be the ideal, but the experiment can still be done without having to do that.

We've discussed 3 possible ways of designing the mask(s) - one might try doing all 3 and comparing results. :-k

George
2005-May-18, 07:22 PM
Whereas if you had a mask that had the Sun's observed pre-atmosphere spectrum curve, then that spectral curve would get multiplied by the light coming into your instrument, so your eyes would be comparing (post atmospheric sunlight * pre atmospheric sunlight) with (post atmospheric sunlight) thus giving you the true color within our limited Earthbound color space. :)
This is interesting but new to me. Are you suggesting a pre-instrument (SPACC) mask to adjust the input into the SPACC?
I'm not sure what a SPACC is. If you don't have a flat spectrum background reference, then making a dot of light that differs from the background in the same way that the intended spectrum would differ from a perfectly white light source is the next best thing. :)
Are we talking of adjusting two light sources; one for the spot and one for background?

SPACC (Stellar Post-Atmospheric Color Corrector) is the acronym (until someone suggests, hopefully soon, a better one. :) [TravisM gave me the term "mask" in lieu of my inaccurate term of template.]

The first and my prior post on this thread may serve to explain what I am trying to accomplish.



If you make a mask in the shape of a pre-atm curve, you will get a lot of imbalance in the ratios of long to short wavelengths since the longer wave light is not reduced significantly by our atmosphere.
But if your pre-atm curve has so many percent of peak intensity at 450nm, then your spot will have that same percentage of daylight intensity at 450nm. You'd have the true color plotted within the spectral distribution of daylight. ;)
I am better with numerical examples, I'm afraid. Consider the 450nm level. Pre-atm is 2006 (w/m^2) vs. the post-atm. level of 1022. (These are derivative plots, of course.) At 680 nm, the pre level is 1511 vs. 1140 for the post level. How will multiplying the spectrums produce the result you desire?

The original idea I had for the mask was to use it to take post atm. spectrum and chisel the colors down to cause the spectrum to be flat. Then, I would use another mask to chisel out the colors into the shape of the pre-atm. spectral irradiance. I quickly realized I could do both with one mask, fortunately. Yet, my explanations tend not to eschew obfuscation. :)


We've discussed 3 possible ways of designing the mask(s) - one might try doing all 3 and comparing results. :-k
I'd like your thoughts on the current plan as per this thread.

umop ap!sdn
2005-May-18, 09:50 PM
Are we talking of adjusting two light sources; one for the spot and one for background?
Not necessarily.


SPACC (Stellar Post-Atmospheric Color Corrector) is the acronym (until someone suggests, hopefully soon, a better one. :)
Oh that's the name of your instrument? Cool! :)


I am better with numerical examples, I'm afraid. Consider the 450nm level. Pre-atm is 2006 (w/m^2) vs. the post-atm. level of 1022. (These are derivative plots, of course.) At 680 nm, the pre level is 1511 vs. 1140 for the post level. How will multiplying the spectrums produce the result you desire?
Wow, there's that much difference at 450nm? :o

In order to use numerical examples, I'd need the numbers to be percentages of maximum. :-? Okay, for the sake of example, let's say the pre-atm levels for 450, 550, and 680nm were 90%, 80%, and 67% of peak and the post-atm levels were 46%, 55%, and 51%. To get a curve that shows the right color within the ambient color space, you'd need the SPACC to output light with a (90% of 46 is 41, etc) 41%, 44%, 34% curve.


The original idea I had for the mask was to use it to take post atm. spectrum and chisel the colors down to cause the spectrum to be flat. Then, I would use another mask to chisel out the colors into the shape of the pre-atm. spectral irradiance. I quickly realized I could do both with one mask, fortunately. Yet, my explanations tend not to eschew obfuscation. :)
I understand what you're planning to do; I'm just not sure it'd give the intended result.


I'd like your thoughts on the current plan as per this thread.

If, using my numbers above, your SPACC reconstructs a pre-atm 90%, 80%, 67% curve, your eyes will compare that to the ambient 46%, 55%, 51% daylight and perceive the color as though the curve were (90 is 196% of 46, etc) 196%, 145%, 131% which would represent the difference scattered by the atmosphere and look quite a bit bluer than the Sun's true pre-atm color.

George
2005-May-18, 11:06 PM
SPACC (Stellar Post-Atmospheric Color Corrector) is the acronym (until someone suggests, hopefully soon, a better one. :)
Oh that's the name of your instrument? Cool! :)
The original instrument was a S.A.D. (Solar Attenuation Device) operated by a S.A.P. (SAD Activation Personell). It was a simple strobe made from a used pencil and glued to a paper plate. Possibly the world's cheapest scientific instrument (about 5 to 20 cents). :) It showed the Sun is white except at lower altitudes.

Here is the spectral irradiance for both pre and post atmospheric levels. [Post atm. is at a solar altitude of 30deg]....... here (http://hyperphysics.phy-astr.gsu.edu/hbase/vision/imgvis/solarirrad.gif)


If, using my numbers above, your SPACC reconstructs a pre-atm 90%, 80%, 67% curve, your eyes will compare that to the ambient 46%, 55%, 51% daylight and perceive the color as though the curve were (90 is 196% of 46, etc) 196%, 145%, 131% which would represent the difference scattered by the atmosphere and look quite a bit bluer than the Sun's true pre-atm color.
Now I understand, sorta. Are you saying, since the ambient light (post atm) is much stronger from red to yellow than the blues and greens, this will cause contrast to the spot and cause one to see more blue. Is it really just a matter of multiplication? I would think not, but I am weak in color pedantry. :)

Also, your saying I can adjust the mask to offset the contrast effect due to ambient background lighting. This would be easy to do.

Are there other ways around this? What if there is no ambient light? This was part of my original plan if abiguity arose. I wanted overhead cover as the sky is strong in blue, of course. If the spot is strong (hue,saturation), would ambient light be an issue? One nice thing is the flatness of ambient light (except for blue and violet). [The link does a nice job of showing this.] This fact should help.

umop ap!sdn
2005-May-18, 11:27 PM
Here is the spectral irradiance for both pre and post atmospheric levels. [Post atm. is at a solar altitude of 30deg]....... here (http://hyperphysics.phy-astr.gsu.edu/hbase/vision/imgvis/solarirrad.gif)
Thanks! :)


Now I understand, sorta. Are you saying, since the ambient light (post atm) is much stronger from red to yellow than the blues and greens, this will cause contrast to the spot and cause one to see more blue.
Basically, yeah.


Is it really just a matter of multiplication? I would think not, but I am weak in color pedantry. :)
I would imagine multiplication would at least be a close enough approximation, but I could be wrong. :)


What if there is no ambient light?
Then if the spot wasn't saturated enough it might look white.


If the spot is strong (hue,saturation), would ambient light be an issue? One nice thing is the flatness of ambient light (except for blue and violet).
The more saturated the spot, the less effect ambient light has.

For some reason I was thinking the Sun's post-atm curve was a lot less flat than that. But the curve you liked to is flat enough for our purposes from about 480nm to almost 700nm, and it doesn't even reach 50% until hitting 400nm. I think that'll work nicely then; a reconstructed pre-atm curve would look only slightly too blue to an eye adapted to daylight.

Mendel
2005-May-18, 11:38 PM
Oh bravo =D>

I had heard about the true color of the sun being something very different to white or yellow but I didn't think it was feasible to demonstrate it with a nice picture like that! Congrats on your achievement!

George
2005-May-19, 12:43 PM
Oh bravo =D>

I had heard about the true color of the sun being something very different to white or yellow but I didn't think it was feasible to demonstrate it with a nice picture like that! Congrats on your achievement!
Thanks. But, a proper experiment has not be done. Of course, we can still have fun with it. :)

SPACC II is in design/build.

George
2005-May-19, 01:10 PM
If the spot is strong (hue,saturation), would ambient light be an issue? One nice thing is the flatness of ambient light (except for blue and violet).
The more saturated the spot, the less effect ambient light has.

For some reason I was thinking the Sun's post-atm curve was a lot less flat than that. But the curve you liked to is flat enough for our purposes from about 480nm to almost 700nm, and it doesn't even reach 50% until hitting 400nm. I think that'll work nicely then; a reconstructed pre-atm curve would look only slightly too blue to an eye adapted to daylight.
I like this news! Adding the ambient spectrum as a variable to the mask profile could be a little trickier than I thought, but it appears it may not be necessary. Hopefully, the spot (we need a better term with more pazzaz than "spot") will have high saturation. Of course, it will be harder for others to swallow a strong blue sun (or the one I'd like to see, a green sun! :) )

I'll try to drag in here a blue sky spectral irradiance to see how it compares to the pre-atm solar irradiance. If it is close, we may have more evidence favoring a blue sun. I'm guessing it might be a cyan sun. [If it is, I wonder if I can do a parody from the theme song of "Cheyenne"? :) But, only the old folks, like me, might catch on to that one. :( ]

This experiment is probably something like 70% colorful fun and 30% science, so shouldn't posters put some color in each post? 8) :)

umop ap!sdn
2005-May-19, 03:17 PM
(we need a better term with more pazzaz than "spot")
Localized Reconstructed Sunlight? :D
Entity at Position X? :lol:
Final Viewable Output?
Grand Unified Sunlight? :)


Of course, it will be harder for others to swallow a strong blue sun (or the one I'd like to see, a green sun! :) )
It should be at least a little bluish and definitely quite greenish, based on the curve.


I'll try to drag in here a blue sky spectral irradiance to see how it compares to the pre-atm solar irradiance. If it is close, we may have more evidence favoring a blue sun. I'm guessing it might be a cyan sun.
Just going by how flat the post-atm curve is in the red through green regions, I'd say the blue sky curve would more or less follow the Sun's pre-atm curve within the range of 480-700nm. :) But because of the Sun's peak in the green-blue and subsequent tapering off towards shorter wavelengths, a cyan Sun is likely.


This experiment is probably something like 70% colorful fun and 30% science, so shouldn't posters put some color in each post? 8) :)
http://www.eknent.com/etc/yes1.gif http://www.eknent.com/etc/thumb.gif

George
2005-May-19, 05:28 PM
(we need a better term with more pazzaz than "spot")
Localized Reconstructed Sunlight? :D
Entity at Position X? :lol:
Final Viewable Output?
Grand Unified Sunlight? :)

1) LRS - Much better than "spot". Articulate and not overly scientific
2) EPX - I like incorporating the "X" as everyone knows - "x marks the spot". But entity and position are not quite right. "X" invokes some mystery yet simplicity, and it implies a spot. [It also matches my illustration. :) ] Let's take this road further, somehow.
3) FVO - Hmmm. Too little drama. Maybe - Solar X, or not.
4) GUS - Nice acronym with character. The general public might not fully comprehend your suggestion, however. :)

Still, a great start in embellishing our quest. =D> :)

Maybe we should get Candy to help us with a contest for these new terms, obviously needed to articulate this scientific endeavor.



Of course, it will be harder for others to swallow a strong blue sun (or the one I'd like to see, a green sun! :) )
It should be at least a little bluish and definitely quite greenish, based on the curve.
Odd that other sites don't seem to think this way.

BTW, I notice your color code use above. Is there a place that lists these color codes?



I'll try to drag in here a blue sky spectral irradiance to see how it compares to the pre-atm solar irradiance. If it is close, we may have more evidence favoring a blue sun. I'm guessing it might be a cyan sun.
Just going by how flat the post-atm curve is in the red through green regions, I'd say the blue sky curve would more or less follow the Sun's pre-atm curve within the range of 480-700nm. :) But because of the Sun's peak in the green-blue and subsequent tapering off towards shorter wavelengths, a cyan Sun is likely.
I hereby declare that you have now obtained the itch for the - [deepvoice="on"] [i]Quest for the Colorof theSun! [voice=off]

That now makes two, at least, on the planet that think cyan is probable. :)

Take a look at this blue sky plot.... here (http://www.laser2000.se/datablad/ocean_optics/hr2000_cg.htm) (bottom graph). If this makes a blue sky, then maybe............ .



This experiment is probably something like 70% colorful fun and 30% science, so shouldn't posters put some color in each post? 8) :)
http://www.eknent.com/etc/yes1.gif http://www.eknent.com/etc/thumb.gif

We may be a bit to colorful I fear. So, maybe "discretion is the better part of color" (or something like that) :)

umop ap!sdn
2005-May-19, 06:52 PM
BTW, I notice your color code use above. Is there a place that lists these color codes?
Not that I know of. I just mostly use web-safe colors (3 pairs of identical hexadecimal digits ranging from 00-33-66-99-CC-FF, in RGB order).


I hereby declare that you have now obtained the itch for the - [deepvoice="on"] Quest for the Colorof theSun! [voice=?off?]
Certainly. http://www.eknent.com/etc/yes1.gif


Take a look at this blue sky plot.... here (http://www.laser2000.se/datablad/ocean_optics/hr2000_cg.htm) (bottom graph).
Cool! I had no idea it had all those peaks in the IR and UV. I notice that the peaks are around 450, 410, and 360nm as opposed to the Sun's 475nm or so.


We may be a bit to colorful I fear. So, maybe "discretion is the better part of color" (or something like that) :)
OK but I gotta still color my wavelengths. :lol: Been into spectroscopy so long that the numbers remind me of the wavelengths' perceived color. It's a lot like synaesthesia, but acquired instead of inborn. :D

John Dlugosz
2005-May-19, 08:50 PM
Whatever you do, just measure the resulting spot and confirm that the color profile matches the above-atmosphere spectrum for the sun. Then you know you have your mask right, and have also corrected for anything else in the overall system.

After all, the question isn't "what is the actual spectrum of the sun?". That is the given! The question is, "what does the human visual system make of this spectrum, under various intensities and ambient illuminations?". So all that matters is to produce light of that spectrum, and using the sun to do so is just a coincedence!

My money's on peachy-pink. The visual model works right for making pigments and gells and phosphors, so they must know what they are talking about. But I'd love to see it demonstrated!

--John

George
2005-May-19, 10:52 PM
Not that I know of. I just mostly use web-safe colors (3 pairs of identical hexadecimal digits ranging from 00-33-66-99-CC-FF, in RGB order).
Thanks. Those should be easy to find. I might even make time to list them with a little Color. :)



Take a look at this blue sky plot.... here (http://www.laser2000.se/datablad/ocean_optics/hr2000_cg.htm) (bottom graph).
Cool! I had no idea it had all those peaks in the IR and UV. I notice that the peaks are around 450, 410, and 360nm as opposed to the Sun's 475nm or so.
I did a quick comparitive graph.....

http://img264.echo.cx/img264/3808/irradiancecompwithsky9ig.th.jpg (http://img264.echo.cx/my.php?image=irradiancecompwithsky9ig.jpg)

I stretched the blue sky considerably (vertically) in order to compare it better with exoatmospheric irradiance. This doesn't prove anything but it does tend to reinforce a possible colorful sun.



We may be a bit to colorful I fear. So, maybe "discretion is the better part of color" (or something like that) :)
OK but I gotta still color my wavelengths. :lol:
That should be a new BABB requirement! =D> Classic stuff!


Been into spectroscopy so long that the numbers remind me of the wavelengths' perceived color. It's a lot like synaesthesia, but acquired instead of inborn. :D
Your experience is well appreciated.


After all, the question isn't "what is the actual spectrum of the sun?". That is the given! The question is, "what does the human visual system make of this spectrum, under various intensities and ambient illuminations?". So all that matters is to produce light of that spectrum, and using the sun to do so is just a coincedence!
The concern we had was the ambient light tricking the eye/brain. If the color saturation is weak, a background color source could alter what is really there. However, since the ambient spectrum is flat (except in blue), we are assuming all is well. [note: we are attempting to add color to every post to reinforce the colorful spirit at work :) ]

I plan to cover the project and could, eventually, fully enclose it to simulate the sun in space (eliminating the ambient issue).

Of course, at the rate I'm going, an astronaut might hear of this and take a simple strobe (a S.A.D.) up with him/her and see it for themselves. :(

George
2005-Jun-01, 04:11 AM
dah dit dah dit ..... dah dah dit dah

Here Comes the Sun...and I say...

For those curious in this suspensful saga for complete mastery of the Sun's spectrum....

Using two prisms and two double convex lenses brought us closer today to homogenizing the adjusted spectrum into one spot. It is looking white at this point. There was some fringe blue that did not quite get in the mix so I suspect there is some hope for a bluish white or a cyanish white stellar host. It is likely going to be subtle and, thus, much more effort will be required. Shucks.

Of course, this result, would help explain why white objects in space look white....duh. :)

I have several ideas on how to improve the result. Hopefully, this summer will bring a significant sigma result. [Of course, ultimately, John Ds suggestion to verify with matching spectral irradiances will nail it. ]

Question....
Im piggy-backing the SPACC using a Celestron 8"SCT with go to. Do I have to do the star align procedure every darn time to start tracking? There should be some way to skip it for daytime use as how is anyone going to slew to a star for calibration? What am I missing?

A Thousand Pardons
2005-Jun-01, 04:22 AM
Question....
Im piggy-backing the SPACC using a Celestron 8"SCT with go to. Do I have to do the star align procedure every darn time to start tracking? There should be some way to skip it for daytime use as how is anyone going to slew to a star for calibration? What am I missing?
I've never used that, but you have to tell it something, don't you?

In the daylight, tracking the Sun (or moon!) uses a different rate than tracking stars at night.
Is there something in the manual sections on that?

PS: I guess I should have asked, do you have a manual?

PSS: you can align to terrestrial objects that you define I believe.

George
2005-Jun-01, 04:52 AM
In the daylight, tracking the Sun (or moon!) uses a different rate than tracking stars at night, right? Is there something in the manual section on that?
Once alignment procedure is complete, I can choose from about 4 trakcing rates (siderial, solar, king, lunar).

However, it seems silly to do the alignment for solar use. I did read the manual but found nothing to allow the skipping of the alignment procedure. I have a choice of either auto align or two star align. Both sends the scope to more than one star which is pointless (pun intended of course) in the blue daylight sky .

The terrestrial mode is simply turning off tracking. This is a good suggestion, however, as surely this mode would allow skiping the alignment procedure. I thought I tried going to "Menu" first but it would not let me, I think. I'll try again. Thanks.

[FWIW, I am encouraging the use of color in each post to add to the spirit of the quest. :) ]


[Added: Your post is the most Color full edit ever on BABB!]

George
2005-Jun-01, 05:54 PM
For those curious about what the SPACC looks like.....

http://img85.echo.cx/img85/8042/may31layout6ne.th.jpg (http://img85.echo.cx/my.php?image=may31layout6ne.jpg)

Here is a photon's view (sorta) of how the mask works.

http://img85.echo.cx/img85/7063/spaccpathview2ed.th.jpg (http://img85.echo.cx/my.php?image=spaccpathview2ed.jpg)

I thought you might enjoy seeing that even a narrow slit can produce a strong spectrum with only a little shade. The ink used by inket printers seem to have some reflective property which helps. The horizontal lines on the mask are due to my old lousy printer (now completely shot).

As you might notice, the mask is not quite properly set as blue is being cheated a bit. I'm awaiting a new printer to make a better mask.

After the convex lens is another prism which sends the spectrum to another double convex lens. This combination homogenizes the light fairly nicely. This is projected onto white paper.

Each element is easily adjusted for the many trial & error positions by using strong magnets. Thin metal plates were glued to the prisms. The main SPACC table has a thin sheet metal section on top.

Surprisingly, the weight has not overcome the Celestron drive works as it tracks fairly well.

That is a horrible shot of my son Matt. I got him off guard as he really doesn't normally look that way. :)

[Added: I will make a mask for the blue sky irradiance and see if we get a blue sky color, too.]

Maksutov
2005-Jun-13, 06:53 PM
I visited Yerkes Observatory a few years ago and since then have visited via the net. I noticed this bit about the 40" refractor in their scope information page:


40-inch refractor
The focal ratio is f/19 with a conventional doublet lens designed and fabricated by the Clark brothers. The doublet is corrected for visible (yellow) light.
Now that's interesting re the type of light the telescope's objective for corrected for! :wink:

Here's the site. (http://astro.uchicago.edu/yerkes/)

A Thousand Pardons
2005-Jun-13, 07:24 PM
That may be because yellow(green) is in the center of the visible spectrum, and allows the least totol out of focus light.

George
2005-Jun-20, 02:05 PM
That may be because yellow(green) is in the center of the visible spectrum, and allows the least totol out of focus light.
Yes and no, I think. Take a close look at the last image posted showing the dispersion onto the mask. The mask is cut from 400 to 700 nm. Vision is, supposedly, 420 to 680 nm. Note the middle point of the refraction is be between green and blue. Refraction is not linear. So the "middle" is not yellow.

However, most of the colors are off-centered from this point. Therefore, better color addition might be around 580nm (yellow). Just a guess as I am not educated in optics (just subborn in this particular quest :) ).

[edit: Sorry I am slow on this post (out of country). Also, you are not as Color full as before. :-? :) ]

George
2005-Jul-09, 03:09 PM
dah....di dah....dit, dah....dah....di dah:

For all those who have requested updates...

I made my mask too small and part of the spectrum went over the mask into the subsequent lenses. This produced a second "spot". After saying "ug", I realized this spot is the normal color we see. What we see is almost a flat spectrum (except in blue)? This produces a white spot which now gives me a reference color to compare my off-white "true color" spot.

I observed through sunset to see if the SPACC would produce matching colors at this new spot. It seemed to but I am located between two hills so I missed the real color changes. I also learned that it will be hard to determine the subtle color difference as the two spots both varied equally, apparently, in color changes. This discussion was addressed earlier in this thread and is now become another issue for us. This is due to the apparent fact the Sun is not a strong, solid color but an off-white or, possibly, white star.

This project was suppose to be done in about two months. However, of even greater complexity and peculiararity, I've been asked to do a seminar on this. I recently learned the title of the seminar - "solar spectroscopy". [Did I spell that correctly?] I refuse to tell you which university as I would never want to impugn their intellect. Of course, this is all ya'll's fault! I could have played much more golf than join this bazar board. :roll: :)

If you are blessed with spectroscopical wisdom, plus submit your contributions to the following new thread. (http://www.badastronomy.com/phpBB/viewtopic.php?p=500558#500558)

grant hutchison
2005-Jul-20, 06:53 PM
OK - I take your hint and move over to this thread to discuss colour constancy.
Here's the way I think about the question "What colour is the Sun?" or indeed "What colour is [star of your choice]?"
With the naked eye, you have three interesting regimes in which you can register star colour (and two trivial boring ones):

1) Close up, with a visible disc. In this setting, most stars are too bright to look at directly. So they are OUCH-coloured. If any impression is left it's going to be blanket overstimulation of all cones, which equals white. Because of the phenomenon of colour constancy, you're also going to judge the ambient light shed by this star to be white - because that's the way your brain handles any continuous spectrum.
(The exception will occur when the surface brightness of the star falls below the threshold of comfortable gaze, which is somewhere in the low M class - at which point your cones will be able to start distinguishing colour as you look directly at the stellar disc.)

Now zoom outwards ... As you move away from the Sun/star, it will decrease in brightness by the square of your distance, and decrease in apparent angular area by the square of your distance. The result is that the surface brightness stays the same. So the Sun is still OUCH-coloured all the way out to Neptune - it's just a smaller OUCH-coloured spot in the sky. But eventually you hit the diffraction limit of the human eye - around 1 arcminute. When the Sun is only an arcminute across, it will stop getting smaller on your retina - instead, its light will be spread more and more thinly across a spot an arcminute across in your visual field - its surface brightness will begin to fall. That brings us to the second regime for the colour of stars:

2) Distant, no visible disc, and possible to regard directly. At this stage, you're actually going to see some colour, because your cones are being differentially stimulated in proportion to the spectral flux. The colour you see will depend on your ambient light, however, which will set your mental white point. So if you are viewing the Sun from a room beyond Pluto which is lit by incandescent light, I think it'll look a bit blue; but if you have arc-lights turned on in your vicinity, the distant Sun might look yellow or even pink. If the Sun is the only source of illumination you have, though, colour constancy will tune your brain to make it appear white.

Zoom back some more - the Sun/star now gets dimmer and dimmer as you move away. Eventually, when the distant star gets somewhere between apparent magnitude 1 and 2, its light is so dim it can barely register on the cones. Next regime:

3. Bright stellar magnitude, just able to register colour.. You start to lose colour detail. Mainly, this affects "red" stars. In regime 2, if you were observing an M-type star from somewhere with solar-type illumination, it would appear orange. But if you get far enough away, the minimal blue and green wavelengths emitted by these stars just don't register on your cones - so the star turns as red as Antares. For blue stars, the visible spectral slope is much less, so I'm not aware of glaringly blue stars in the same way that red stars are convincingly red.

Next. Somewhere around apparent magnitude 2, and depending on the observer and location, you get:

4. Complete loss of colour perception.You're running on rod vision, and the star appears white.

And then the final, really dull regime:

5. Complete loss of contrast perception. Eventually the star is too dim to stand out against the ambient sky brightness. Conventionally, this is apparent magnitude 6, but in very dark locations with a good accommodated eye you'll see considerably dimmer stars.

Grant Hutchison

George
2005-Jul-20, 07:31 PM
OK - I take your hint and move over to this thread to discuss colour constancy.
Welcome to the quest! =D> Hopefully it is not just an itch I have and a few others, too. :)


1) Close up, with a visible disc. In this setting, most stars are too bright to look at directly. So they are OUCH-coloured. If any impression is left it's going to be blanket overstimulation of all cones, which equals white. Because of the phenomenon of colour constancy, you're also going to judge the ambient light shed by this star to be white - because that's the way your brain handles any continuous spectrum.
The sun's intrinsic color is being determined on the basis of equal attenuation across the spectrum to get below the "Ouch level" (special Sun glasses, if you're pun tolerant). :) So, if the surface brightness is reduced to some optimal level for the cones to discrimanate color easily, would one see the sun as having color if only a black background exists? [This assumes it has some color, of course.] Is this viable?

Comparing the sun "spot" obtained by SPACC to a spot made from a true flat spectrum might help overcome the color constancy issue. Currently, I am using the almost flat post-atm (AM2) level as a comparitive spot. What do you think of this idea?

I have also thought of using different color backgrounds to wiggle the color out of it by various comparisons. However, this would require some study on my part. :-?

grant hutchison
2005-Jul-20, 07:48 PM
So, if the surface brightness is reduced to some optimal level for the cones to discrimanate color easily, would one see the sun as having color if only a black background exists?I think you're into my Regime 2, in which you're getting sensible data from your cones, but are at the mercy of whatever your brain thinks of as ambient light. If you are looking at just your "corrected" solar image in an otherwise dark environment, then I think it'll be seen as ambient and therefore white. If you're also aware of bright Earth-surface sunlight around you while you examine your "corrected" solar image it might appear bluish by comparison.


Comparing the sun "spot" obtained by SPACC to a spot made from a true flat spectrum might help overcome the color constancy issue. Currently, I am using the almost flat post-atm (AM2) level as a comparitive spot. What do you think of this idea?And eliminating other brighter sources of light? Thinking about black body curves makes me anticipate that your "corrected" sun will be slightly greener than a completely flat spectrum, and your post-atm light will be slightly yellower than a flat spectrum. So post-AM2 just marginally warmer than "corrected". If these are the only two, equally bright, light sources in your visual universe, then post-AM2 will look a little yellow, and "corrected" a little blue-green.

Grant Hutchison

George
2005-Jul-21, 12:17 AM
So, if the surface brightness is reduced to some optimal level for the cones to discrimanate color easily, would one see the sun as having color if only a black background exists?I think you're into my Regime 2, in which you're getting sensible data from your cones, but are at the mercy of whatever your brain thinks of as ambient light. If you are looking at just your "corrected" solar image in an otherwise dark environment, then I think it'll be seen as ambient and therefore white. If you're also aware of bright Earth-surface sunlight around you while you examine your "corrected" solar image it might appear bluish by comparison.
Current SPACC results are favoring a white sun. This is, FWIW, different than a traditional yellow sun.

You might be interested in reviewing another analysis of the Sun's color (peachy pink) (http://www.vendian.org/mncharity/dir3/starcolor/sun.html) :o

If you find yourself in a strange corn-ball, antic-pedantic mind set, you could venture to the initial tests of the suns color using S.A.D. (http://www.badastronomy.com/phpBB/viewtopic.php?p=186347#186347) It demonstrated the sun appears white to Earth's inhabitants in an extremely economical fashion. :)



Comparing the sun "spot" obtained by SPACC to a spot made from a true flat spectrum might help overcome the color constancy issue. Currently, I am using the almost flat post-atm (AM2) level as a comparitive spot. What do you think of this idea?And eliminating other brighter sources of light? Well...not really. I have a beige covering which blocks blue skylight and I may end up making a dark room with black plastic now that I have at least the AM2 reference light.


Thinking about black body curves makes me anticipate that your "corrected" sun will be slightly greener than a completely flat spectrum, and your post-atm light will be slightly yellower than a flat spectrum. So post-AM2 just marginally warmer than "corrected". If these are the only two, equally bright, light sources in your visual universe, then post-AM2 will look a little yellow, and "corrected" a little blue-green. That is what I hoped for. I got a slightly yellow white AM2 spot adjacent to a white SPACC spot (though I expected a bluish, or greenish, white color). However, ambient lighting existed on the card, too (as mentioned above).

I have yet to combine all the blue into the spot, however. There is some hope for a non-white sun when compared to AM2 lighting.

pghnative
2005-Jul-22, 12:37 PM
I've been lurking in thread a while, and I think that the last few posts (plus the "magenta dwarf" thread) begs a philosophical question:

Since color is just the brain's interpretation of wavelength (frequency) data, and since the brain treats any reasonably continuous spectrum as white, then isn't the color of the sun "white" by definition?

George
2005-Jul-22, 02:31 PM
Since color is just the brain's interpretation of wavelength (frequency) data, and since the brain treats any reasonably continuous spectrum as white, then isn't the color of the sun "white" by definition?
It is obviously now a big issue now that the SPACC results are indicating a white, possibly near white, result. This issue came up earlier but I was hopeful it would not be a big factor.

Nevertheless, it is still a bit surprising white may indeed be the result because the difference between the pre-atm. irradiance and post-atmosphere irradiance is not insignificant. For instance, at 400nm (violet) the difference is almost 3x more irradiance above our atm., and at 460nm (blue) it is still about 2x. Green is also more intense above so I was hoping the strong blue added to yellow would augment the green and produce a green star. When my first SPACC-1 result came in solid green, I nearly shouted "Eureka!". :) Typical novice experimenter. :-?

My plan now is to make a few masks which should produce a known color result. A blue sky irradiance mask is next on my list.

The more I fiddle with SPACC-2 the better we both get. :)

I am still curious if the SPACC instrument is unique. #-o

grant hutchison
2005-Jul-22, 04:09 PM
Since color is just the brain's interpretation of wavelength (frequency) data, and since the brain treats any reasonably continuous spectrum as white, then isn't the color of the sun "white" by definition?Well, not actually by definition - although there are many defined "white points" for various purposes, I think the solar spectrum is too messy for anyone to use in an official definition. (Might be wrong on that, though.) But I think it's certainly white by default - it's overload-white when you're close, it's rod-vision-white when you're far away, and in between its peceived colour will be influenced by your ambient lighting, but will be white if the Sun is the only significant source of light you've got. There's a big "grey area" in the desaturated middle of the colour triangle, and your brain is content to latch on to any part of that and call it "white" if it turns up as the colour of your predominant light-source.
So I guess my take on it is that though there is much fun and interest to be had from the question "What colour is the Sun?", in logical terms it is ill-posed - it has a hidden erroneous assumption that we are actually able to perceive a unique colour for the solar disc. We can certainly assign the solar spectrum to a unique point in "colour space" coordinates, but what physiological colour we associate with those coordinates depends on what we're currently calling "white".

(I'm specifically interested in brown dwarfs, over on the other thread, because of their discontinuous spectra, which should give rise to real colours that'll register reproducibly on the human eye. The L-dwarfs, for instance, are heavily weighted to the red wavelengths, in much the same way as the spectrum radiated by an old electric-element heater or hob. Turn off the lights, making your electric element the only light source in the room, and it still looks red - the colour produced by its visible spectrum is just too saturated for your brain to "buy" it as a white light.)

Grant Hutchison

George
2005-Jul-22, 07:41 PM
So I guess my take on it is that though there is much fun and interest to be had from the question "What colour is the Sun?", in logical terms it is ill-posed - it has a hidden erroneous assumption that we are actually able to perceive a unique colour for the solar disc. We can certainly assign the solar spectrum to a unique point in "colour space" coordinates, but what physiological colour we associate with those coordinates depends on what we're currently calling "white".
If nothing else, we should be able to demontrate the Sun should be considered white and not yellow, thereby debunking the original color assesment by the famous discoverer of the Sun - Ug (around 49,000 BC). [Documentation on his stunning discovery is rather sketchy, however ] :wink:

There is still some slight hope the Sun does posesses a tinge of color, assuming an observer uses a device to evenly attenuate the flux.

grant hutchison
2005-Jul-22, 08:00 PM
If nothing else, we should be able to demontrate the Sun should be considered white and not yellow...Yes, I noticed some discussion of the "yellow sun" idea on this thread as I scanned through it. I have to admit that this concept came as a complete oddity to me when I read Phil Plait's book, and I remember thinking it must be some strange perceptual quirk of his that he walked around seeing the sun as yellow. The idea that this is a widespread perception that requires explanation and/or debunking is slightly unhinging for me: like suddenly encountering, in middle life, a whole area of research aimed at explaining why clouds appear green to most people when their reflective properties suggest that they ought to appear white.

Grant Hutchison

George
2005-Jul-22, 09:35 PM
If nothing else, we should be able to demontrate the Sun should be considered white and not yellow...Yes, I noticed some discussion of the "yellow sun" idea on this thread as I scanned through it. I have to admit that this concept came as a complete oddity to me when I read Phil Plait's book, and I remember thinking it must be some strange perceptual quirk of his that he walked around seeing the sun as yellow.
Actually, I think he sees it as white (ground observation at a fair altitude), but admits he does not know if it might have some color. Most people assume it is really yellow. Few consider it white, including scientific web sites, with and without H-R diagrams.

There are several threads on this subject. However, the ultimate formulae to explain this perceptual phenomena is prescribed.... here (http://www.badastronomy.com/phpBB/viewtopic.php?p=174986#174986). [It's 100% serious and 100% corn :) ]


The idea that this is a widespread perception that requires explanation and/or debunking is slightly unhinging for me: like suddenly encountering, in middle life, a whole area of research aimed at explaining why clouds appear green to most people when their reflective properties suggest that they ought to appear white.
I would guess atmospheric physicists have this one nailed down. It should be easy to comprehend considering most bodies of water look green or blue (due to water's red absorption properties).

The Sun's color is not so easy. Consider the ramifications for a humble, hard-working astronomer/solar physicist when all his physicist's friends learn he doesn't know the Sun's color. :cry: :wink: Fortunately, there is BABB to the rescue! We're just doing our job here (even if we don't have one). :)

grant hutchison
2005-Jul-22, 10:59 PM
I have to admit that this concept came as a complete oddity to me when I read Phil Plait's book, and I remember thinking it must be some strange perceptual quirk of his that he walked around seeing the sun as yellow.
Actually, I think he sees it as white (ground observation at a fair altitude), but admits he does not know if it might have some color.
Ahem:

If asked, I would say that the Sun is yellow. (Bad Astronomy John Wiley & Sons, New York, 2002. Chapter 4.)



... like suddenly encountering, in middle life, a whole area of research aimed at explaining why clouds appear green to most people when their reflective properties suggest that they ought to appear white.
I would guess atmospheric physicists have this one nailed down. It should be easy to comprehend considering most bodies of water look green or blue (due to water's red absorption properties).
You're kidding me, right? That fictitious example was intended to show how bizarre I find the report that "most people" see the real, outdoor, noonday Sun as yellow - it's as if I suddenly found out that most people see clouds as green.
But everyone sees white clouds.

Right?

Tell me that's right.


Grant Hutchison

Van Rijn
2005-Jul-23, 12:42 AM
You're kidding me, right? That fictitious example was intended to show how bizarre I find the report that "most people" see the real, outdoor, noonday Sun as yellow - it's as if I suddenly found out that most people see clouds as green.
But everyone sees white clouds.

Right?

Tell me that's right.


Grant Hutchison

Well, sometimes they are red, sometimes gray to black. Sometimes other colors - but if they are green, I'm probably having a really bad day.

8)



Seriously, green clouds are possible - due to green dust and so on. As for the sun - perception of color has at least as much to do with the mechanics of the eye and brain as the actual color. Since I can't look at the noonday sun directly without filters, it is a bit hard to say exactly what color it is. And given that light intensity affects color perception, it becomes an almost impossible question to answer.

George
2005-Jul-23, 01:06 AM
I have to admit that this concept came as a complete oddity to me when I read Phil Plait's book, and I remember thinking it must be some strange perceptual quirk of his that he walked around seeing the sun as yellow.
Actually, I think he sees it as white (ground observation at a fair altitude), but admits he does not know if it might have some color.
Ahem:

If asked, I would say that the Sun is yellow. (Bad Astronomy John Wiley & Sons, New York, 2002. Chapter 4.)
I did not remember that one. :-? However, he has stated he is unsure but [now] thinks of it as more white than yellow. This I heard while enjoying his guest spot on Slacker Astronomy (http://www.somethingpositive.net/050517-slackextra.mp3). [About 1/3 the way on the track - I'm still laughing at this one. :) ]




... like suddenly encountering, in middle life, a whole area of research aimed at explaining why clouds appear green to most people when their reflective properties suggest that they ought to appear white.
I would guess atmospheric physicists have this one nailed down. It should be easy to comprehend considering most bodies of water look green or blue (due to water's red absorption properties).
You're kidding me, right? That fictitious example was intended to show how bizarre I find the report that "most people" see the real, outdoor, noonday Sun as yellow - it's as if I suddenly found out that most people see clouds as green.
But everyone sees white clouds.

Right?

Tell me that's right.
Yes. You are correct. Clouds look white (except when they look green, red, orange, .... :) )

I thought you were serious about a study on green clouds (as in a serious thunderstorm). As Van Rign has stated, clouds can be green. Green clouds meant hail storms to me, in the past.

Of course, now I understand your point, though I am surprised you are surprised with people's color views of the sun. :)

George
2005-Jul-28, 12:54 AM
Grant....

Just thought you might like to see your mask....

http://img41.imageshack.us/img41/8830/tbrwondwarfmaskprofile4qr.th.jpg (http://img41.imageshack.us/my.php?image=tbrwondwarfmaskprofile4qr.jpg)

The y-axis is compensated for the atmospheric depletions for each wavelength. [Since the violet and blue levels are weaker at our surface, the greater y-level will make up the difference and restore it to an AM0 level.]

The x-axis is adjusted to match the non-linear dispersion.

[edit. This, of course, is before the not-so-exact Exacto-Knife cuts out the color regions.]

The letter coloring reflects optimistic irradiance. :)

A second mask will be placed in tandem to produce a true flat spectrum to serve, hopefully, as a better reference light.

Jorge
2005-Jul-28, 06:36 PM
and there is life again 8)

George
2005-Jul-28, 08:31 PM
and there is life again 8)

Oh, we were never quite dead. :) I still have hope the Sun can be demonstrated to have a shade of color (possibly blue). What little I have read on the color constancy issue that Grant (and others) have raised has not convinced me, yet, that a black background setting (space) would cause the retinex to produce white vs. [added, a white solar color as we may be seeing in] a pure white (flat spectrum) background.

For instance, I read how Edwin Land (inventor/founder of Polaroid) once demonstrated how an orange region in a painting would still appear orange even when red light was used as an illumination source. However, using a tube (e.g. paper roll tube) to restrict one's view of only the same orange region, will cause the person to see red in lieu of orange. The key is the surronding area and, apparently, the issue of color constancy. [Again, this is not a new issue but one that is not resolved for the SPACC program (which, unfortunately, means educating me :-?) .] This is definetly a BABB team effort.

For the color pedantics, some color terms. (http://www.color-tec.com/1gloss.htm) Yikes. :o

Jorge
2005-Jul-28, 10:03 PM
Aren't blue stars hot, bright and short lived?

and i though it was hard to calculate a gradient in Delphi #-o

I have an idea to help a bit,
Mail the nice poeple of slacker astronomy and ask if they'll do a show about this. then you can join the via skype. That might attract some poeple to help you!

George
2005-Jul-29, 03:11 AM
I have an idea to help a bit,
Mail the nice poeple of slacker astronomy and ask if they'll do a show about this. then you can join the via skype. That might attract some poeple to help you!
It was enjoyable hearing the Slacker folks and BA discuss the issue of the Sun's color. =D> Keep in mind this is just a fun project that might, maybe, cause the general public to stop, look up and think about what is up there. Wether the Sun is white or cyan or light green will not, likely, change anything as it will not reveal any new information regarding the properties of an observed body.

It (SPACC-n) might be helpful in being a reference for improved calibration times for the color filters to produce "true" color images. With larger scopes which could someday see exoplanets, the demand may be higher than today for improved color imaging. I haven't really considered it much as I am not qualified to say for sure. Maybe the BA or some real astronomers here have some thoughts on this.

Regardless, the fact that it is stil an unknown keeps us itchin', don't it? :) So, the quest continues.

Jorge
2005-Jul-29, 03:17 PM
i guess that just part of being human,... to try and understand the unknown, it attracts us (and might eventlually kill us too)

George
2005-Jul-29, 10:44 PM
For the fun-loving quest adventurers....

Thanks to Grant, a side road has been taken.

http://img184.imageshack.us/img184/4430/questnameep79ok.th.jpg (http://img184.imageshack.us/my.php?image=questnameep79ok.jpg)

It would be a stormy day, though it started with no wind as if to alure me into the open....

http://img184.imageshack.us/img184/8586/stormtosw5oi.th.jpg (http://img184.imageshack.us/my.php?image=stormtosw5oi.jpg)

I wondered if a break would come as the Sun danced behind forming clouds....
http://img115.imageshack.us/img115/903/picture0108ru.th.jpg (http://img115.imageshack.us/my.php?image=picture0108ru.jpg)

A break in the overhead clouds meant no rain.....

http://img102.imageshack.us/img102/1015/cloudsabove6ki.th.jpg (http://img102.imageshack.us/my.php?image=cloudsabove6ki.jpg)


I quickly set-up, aligned, mounted SPACC, set for Solar tracking, mounted solar filter and lens, attached SPACC elements, installed mask, etc. Once set, it started - rain, though the clouds were quite distant. :-? There was something about a magenta, or maroon, star out there that touched my spirit as if to beckon me onward, deeper into the quest. I implented atypical gear to deal with the weather.

http://img289.imageshack.us/img289/6420/meatscope8hv.th.jpg (http://img289.imageshack.us/my.php?image=meatscope8hv.jpg)


I spoke to the audience with authority and confidence. Gratification came when I observed the gallerie's facial expression of excitement....
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.



http://img74.imageshack.us/img74/2833/shadowface8yg.th.jpg (http://img74.imageshack.us/my.php?image=shadowface8yg.jpg)


The mask was aligned and the adjusted spectrum can be seen in the large subsequent lens. This color was, eventually, recomgined into a spot. [My new free printer made the mask orange in color, though it was suppose to print it as dark gray. I have replaced it.]

http://img74.imageshack.us/img74/655/masktclassbig6qk.th.jpg (http://img74.imageshack.us/my.php?image=masktclassbig6qk.jpg)


Time ran out quickly as the sun sank and the clouds rose.....
http://img115.imageshack.us/img115/903/picture0108ru.th.jpg (http://img115.imageshack.us/my.php?image=picture0108ru.jpg)

There was a brief moment when the spot varied in color due to my quick lens adjustments trying to obtain a full color recombination. There is indeed a chance it is a magenta, purplish-red or maroon star. Further runs should bring resolution.

George out.

Jorge
2005-Jul-29, 11:11 PM
darn it, i was hoping for a comic book styled report :lol:

First: working with images is great!
Second: nice telescope
Third: ... i forgot what iwas writing got distracter by a girl sry

George
2005-Jul-30, 01:34 AM
darn it, i was hoping for a comic book styled report :lol:
I'll try to be a little less formal next time. :)


Third: ... i forgot what iwas writing got distracter by a girl sryHey. No time for that, the quest requires dedication beyond mere mortal weaknesses. [-X

Jorge
2005-Jul-30, 10:48 AM
darn it, i was hoping for a comic book styled report :lol:
I'll try to be a little less formal next time. :)


Third: ... i forgot what iwas writing got distracter by a girl sryHey. No time for that, the quest requires dedication beyond mere mortal weaknesses. [-X

8) Yes, master, I have a weak mind

hhEb09'1
2005-Jul-31, 11:43 AM
There is indeed a chance it is a magenta, purplish-red or maroon star. The sun, right?

George
2005-Aug-01, 12:28 PM
There is indeed a chance it is a magenta, purplish-red or maroon star. The sun, right?
No. :) I guess it is a little odd to look at the Sun to "see" another star. :) Yet, that is what we are doing, as far as "true" color determination.

The sun is our reference light. The Solar spectrum is aligned and passed through a mask made in a modified shape of, in this case, a T-class Brown Dwarf's spectral irradiance. Grant gave me his data and I adjusted it to compensate for the Sun/atmosphere variations in intensity.

The objective is to obtain the correct intensity of color for each respective wavelength. Combining these colors will produce the "true" color of the star. [In the case of the Sun' color determination, the color is based on an attenuation of the flux for optimal viewing.]

My problem has been in recombination. I should have it working in the next 380,000 years, hopefully. :wink:

Launch window
2005-Aug-13, 08:37 AM
darn it, i was hoping for a comic book styled report :lol:
I'll try to be a little less formal next time. :)


Third: ... i forgot what iwas writing got distracter by a girl sryHey. No time for that, the quest requires dedication beyond mere mortal weaknesses. [-X

8) Yes, master, I have a weak mind

so can we all agree on the Sun's color ?

hhEb09'1
2005-Aug-13, 05:54 PM
so can we all agree on the Sun's color ?The "intrinsic color" would be that without the effects of the atmosphere, in the context of this thread. None of us has actually experienced that, though, so we have to depend upon indirect results.

George
2005-Dec-17, 02:16 AM
And the Quest proceeds....

I thought I would take the Thuilier 2002 data on solar irradiance [added: AMO plot, i.e. from space] and compare it with this same data converted to photon flux. If the eye's cones register by the number of photons, and not so much the energy, then this may be a better indicator of the color we see.

http://img235.imageshack.us/img235/1031/thuilierfluxcompar1kf.th.jpg (http://img235.imageshack.us/my.php?image=thuilierfluxcompar1kf.jpg)

This would explain why the SPACC results are so white. The photon flux plot is nearly flat. This even distribuiton to the cones might cause only a white image. Anyone know if this is correct?

I simply multiplied by wavelength and normalized to far red. The color background is fairly close to being correct in relation to the x-axis wavelengths.

Jorge
2005-Dec-18, 09:27 PM
I have no idea what you just said...
But it's good to see that your still working on this :)

George
2005-Dec-18, 10:46 PM
Thanks, Jorge.

I've been reading a little on how the eye works, and I was thinking the cones produce color based mainly on photon count, not energy. A "violet" photon has about twice the energy of a "red" photon. A "blue" photon is not quite as energetic as a violet one. The energy equation for a photon is simple....e = hf (h is Planck's constant and f is the photon's frequency).

The solar spectral irradiance (as seen in space) is very strong in the blue in energy (not no. of photons). The plot I added (above) is, hopefully, a conversion to a relative no. of photons our eye would encounter if they were up yonder. This might be a better representation to use for determining the Sun's color.

Since my early results with the SPACC indicate the sun's color is primarily white, I was wondering why. I thought I would reduce it from an energy graph to a photon flow rate instead. Notice the flatness of the photon flux [no. of photons/wavelength]. If all three cones are getting an equal no. of photons hitting them, will they not produce a white image?

Maybe someone out there will comment on this approach.

Added is in [ ]

Jorge
2005-Dec-19, 09:48 PM
ooh, that wel very wel dommed down :)
I actually understand it now.

This sound to me that it seem quite posible.

George
2007-Aug-03, 10:23 PM
*Bump*

Here is more news from your local in-house heliochromology dep't. The Sun, as in numerous previous episodes, is not a yellow dwarf star. However, it seems that it is also not a white star with a bluish-white center, like I'd hoped. It's very likely just a pure white star, though not necessarily a white dwarf.

Here is why I think this is the case. Our eye is better described as responding to photon flux rather than energy flux. As from a prior episode, converting the spectral irradiance of the Sun, as seen from space, from a spectral energy flux to a photon flux reveals a reasonably flat distribution (see prior post #104). A flat distribution means no one color or specific set of colors have dominance over the others. It is a little more complicated because the bandwidth for each color is not the same -- it is noteworthy that yellow has the most narrow band width. Nevertheless, it is most likely that a flat distribution of photon flux at each wavelength will produce a white appearance for our Sun.

However, that is for a total (ie integrated) Solar flux and does not represent colors that might exist regionally. Across the disk is a large temperature variation as described by the center to limb variation (CLV). This temperature difference is from about 5000K at the limb to about 6390K at the center.

My hope was that a 5000K limb temperature might give us a yellow tinge and a 6390K a bluish appearance. [The lighting industry results I have found sure encouraged a blue color for such a high temp.]

Unfortunately, since solar telescopes that do not use filters (eg McMath Pierce at Kitt Peak) project only a solid white image, including the limb region, then white is the true limb color, as seen from above our atmosphere. This conclusion is obvious because our atmosphere attenuates the blue end of the spectrum; therefore, adding the "blues" back into this solar projection would never produce a yellow result. The question remained, however, whether or not the color of the central region would appear blue if those "blues", taken away by our atmosphere, were added back into the light beam from the central region only.

One way to tackle that problem is to use Planck blackbody models. Assuming we will obtain an even better Planck distribution whenever we observe small regions of the disk, I converted Planck energy distribution for the 6390K central region into a photon flux as our eye might see this central zone if we were seeing it from above our atmosphere. Here is the result.

http://img167.imageshack.us/img167/4057/6390kphotonfluxqy1.th.jpg (http://img167.imageshack.us/my.php?image=6390kphotonfluxqy1.jpg)

Shucks, it looks far to flat to give us hope for a blue center. Worse, the blasted peak is on the edge of yellow. [Actual AMO (above our atmosphere) spectral irradiance places the photon flux peak in yellow.] But, of course, the peak is still a pimple and does not emerge from the whitewash.

Using the same Planck model, it seems to me a star would have to be close to a surface temperature of around 7500K minimum to allow much of a blue look. Any comments as to the lowest stellar temperature you find for a blue star would be helpful.

umop ap!sdn
2007-Sep-17, 05:34 PM
Not long ago, while working on a programming project, I was doing something involving star colors, and I remembered your search for the Sun's true color so I thought I'd share what I found.

Basically I had the program use Planck's Law (http://en.wikipedia.org/wiki/Planck%27s_law_of_black_body_radiation) to calculate, based on temperature, the luminosity of each star in three wavelengths - red, green, and blue - and draw the star on the screen as a point or disc of the resulting color.

Well, here's what it gave me (this is for a view from space with no atmospheric scattering):

http://julie503.nfshost.com/sun.jpg

Enhancing (http://julie503.nfshost.com/sun_enh.jpg) it reveals a green tinted Sun.

I'm not sure if the equation gives its result in photon flux or total energy; I used the function of wavelength if that makes any difference. Also there is no CLV in the image but I will more than likely implement it in the near future.

(For the curious, the exact wavelengths I used were 620, 540, and 460 nm to most closely approximate the colors produced by the average computer monitor. After I changed the code to utilize Planck's Law in this way, views of the constellations suddenly looked a lot more like actual photographs.)

George
2007-Sep-17, 09:35 PM
I think I understand your method. Using Plancks law, you calculated the relative levels of emission from the Sun at 620nm, 540nm, and 460nm (rgb) and used these values for color rendering for your monitor, right?

What color temperature did you use?

Unfortunately, the Sun does vary from the Planck distribution, so a subtle variation in saturation or hue for your result could easily change your result. Also, you will likely need to convert the energy levels to relative photon flux levels. This weakens the blues and greens considerably. Another problem comes from varation in the band widths adjacent to those colors. Red extends to about 750nm, so it is much broader than the other two.

A CLV does appear in your image, which is pretty nice. :)

I once thought the Sun had a chance of being green, but that is because it was me that was green. :)

[Added: nice to see you take an interest.]

umop ap!sdn
2007-Sep-18, 12:23 AM
I think I understand your method. Using Plancks law, you calculated the relative levels of emission from the Sun at 620nm, 540nm, and 460nm (rgb) and used these values for color rendering for your monitor, right?
Yep.


What color temperature did you use?
5785K.


Unfortunately, the Sun does vary from the Planck distribution, so a subtle variation in saturation or hue for your result could easily change your result. Also, you will likely need to convert the energy levels to relative photon flux levels. This weakens the blues and greens considerably. Another problem comes from varation in the band widths adjacent to those colors. Red extends to about 750nm, so it is much broader than the other two.
Drat. Any idea how exactly it varies, and to what extent it applies to other kinds of stars?

I can try the photon flux conversion and see what the resultant star colors look like. (I thought the Pleiades looked a little too blue. :lol: )

It would certainly be possible to take varying bandwidths into consideration; it means more processing time but if the results are considerably different than that would make it worthwhile. I do think the wavelengths should be weighted; the eye's sensitivity at 750nm is only a tiny fraction of its peak.


A CLV does appear in your image, which is pretty nice. :)

...

[Added: nice to see you take an interest.]
Thanks, but it's only limb darkening; there's nothing in the code to specifically adjust the hue. And I've been interested all along. :)

umop ap!sdn
2007-Sep-18, 01:39 AM
I took out the grayscale limb darkening and changed it to use the maximum and minimum temperatures instead.

http://www.umop.net/etc/Sun_enh_lmdark.png

The bottom half is color enhanced to bring out the hues. This image varies from pale cyan at the center to yellowish gray at the edges.

Thing I'm not clear on is how to get photon flux from Planck's equation. It only has 2 choices: luminosity per unit of wavelength and luminosity per unit of frequency. Seems to me that photon flux would be in between these two, no?

George
2007-Sep-18, 01:52 AM
Drat. Any idea how exactly it varies, and to what extent it applies to other kinds of stars? No, but you might want to check out the traditional astronomer's method using standard filters: UBVRI. (http://mcdonaldobservatory.org/research/instruments/instrument.php?i_id=9). This systems gives a much better result as these filters are not narrow banded. [I am not experienced in such things, but it looks pretty straight forward and reasonably effective.]


I can try the photon flux conversion and see what the resultant star colors look like. (I thought the Pleiades looked a little too blue. :lol: ) Yes, use E=hf=hc/wavelength. Our eyes respond to individual photons, though not every single one, of course. [At 1 a.u., the photon flux is about 3x1022/m2.]


It would certainly be possible to take varying bandwidths into consideration; it means more processing time but if the results are considerably different than that would make it worthwhile. This is why I like to take the full spectral irradiance data and, using a well known light source, adjust each wavelength of the source to match that of the object in question. This produces the actual color for all to see.

The prototype sorta works but I haven't built and installed the array of thousands of light fibers to homogenize the color.

I think I will call this device an asterochromometer. Of course, I am not necessarily in charge of naming such devices as I am really only a volunteer heliochromologist; I'm not a pro heliochromologist. :)


I do think the wavelengths should be weighted; the eye's sensitivity at 750nm is only a tiny fraction of its peak. But that is why it must be weighted, a lot of flux at 750nm might still produce a slight color change. Color is actually the product of the spectral radiance of the star and the spectral sensitivity of our eyes.


Thanks, but it's only limb darkening; there's nothing in the code to specifically adjust the hue. And I've been interested all along. :) Yes, I missed that, but it does look nice.

If the Sun has any color, as seen from space at a comfortable intensity level, then it will be in the center of the disk, though there may still be a slight yellowing at the limb, but I doubt it.

George
2007-Sep-18, 02:06 AM
Thing I'm not clear on is how to get photon flux from Planck's equation. It only has 2 choices: luminosity per unit of wavelength and luminosity per unit of frequency. Seems to me that photon flux would be in between these two, no?
No. :)

The energy of a photon can be found in E = h f, where h is Planck's constant and f is frequency. But, frequency is c/wavelength.

Using the Planck BB equation, you can obtain the actual radiation energy levels for each wavelength (though it is in derivative form). [This will give you surface radiance, so you could convert it for 1 a.u. if you want to know actual irradiance arriving at Earth, but the relative strengths are still the same.] Since you know the energy of a photon for each wavelength, due to the prior equation above, you can convert the energy to photon flux.

I mention irradiance because it gives a better picture of the integrated result of all the flux from the entire surface of the Sun. Since the central region has a Planck temperature of around 6390K and the limb about 5000K, you can get an idea why a 5777K (which produces the correct total energy value), or 5785K, or the one I like, 5850K (which seems to best match the actual solar curve), result will not match the actual spectral irradiance of the Sun.

umop ap!sdn
2007-Sep-18, 06:07 PM
I see.... kind of. Well, I tried having it multiply the result of the Planck equation by a value proportional to the wavelength and it gave me a yellowish Sun.

http://julie503.nfshost.com/sun_enh_lmdark_flux.png
(Here again, the bottom half is color enhanced.)

I'm not satisfied with this result because for one thing, the Sun doesn't look yellow to my eyes, and for another, I'm pretty sure its peak is around 475nm. The closest of the 3 wavelengths to that peak is the blue (460nm) and here it's showing up as the weakest. :think:

Makes sense about the integrated flux, thanks. :) I don't know why I didn't think of that....

antoniseb
2007-Sep-18, 06:19 PM
I said this before in this thread, but it seems to me that the light coming from the Sun has to be the very definition of white. Hold a piece of white paper in direct sunlight, and what color does it look? How do we measure white if not as compared to sunlight?

Planet X
2007-Sep-18, 07:05 PM
Not long ago, while working on a programming project, I was doing something involving star colors, and I remembered your search for the Sun's true color so I thought I'd share what I found.

Well, here's what it gave me (this is for a view from space with no atmospheric scattering):

http://julie503.nfshost.com/sun.jpg

(For the curious, the exact wavelengths I used were 620, 540, and 460 nm to most closely approximate the colors produced by the average computer monitor. After I changed the code to utilize Planck's Law in this way, views of the constellations suddenly looked a lot more like actual photographs.)

Hey, that looks kinda like Uranus! Anyway, the sun appears to me as plain white colored sphere, with just a slight pinch of yellow in the halo. Later!

J P

George
2007-Sep-18, 11:15 PM
I see.... kind of. Well, I tried having it multiply the result of the Planck equation by a value proportional to the wavelength and it gave me a yellowish Sun. That doesn't surprise me since the photon flux will shift the distribution redward, which may yield yellow when it was green.

Unfortunately, your result is still coincidental since you only selected three wavelengths to represent all the wavelengths. Assuming you used the same ones for blue, green, and red, you should have calculated a flux at 1 a.u. of:

@ 620nm = 2.58 x 1022 photons.
@ 540nm = 3.04 x 1022 photons.
@ 460nm = 3.20 x 1022 photons.

This gives a ratio of 1/1.18/1.24 for the respective wavelenghts above. However, this comes from the Planck equation. The actual irradiance data will give a ratio of 1/1.03/1.10.

Another consideration is in regard to those color choices as they relate to our vision. 460nm is on the edge between violet and blue, 470nm would be better. 530nm is in the center of green. 620nm is regarded as orange, or orange-red, by some color sites (including one from NASA). 680nm is closer to midrange in red.

Using these wavelengths [460/530/680] gives a ratio of 1/1.07/1.08, respectively. The differences are somewhat significant compared to the Planck values.


I'm not satisfied with this result because for one thing, the Sun doesn't look yellow to my eyes, and for another, I'm pretty sure its peak is around 475nm. The closest of the 3 wavelengths to that peak is the blue (460nm) and here it's showing up as the weakest. :think: The actual peak, using the Thuilier 2002 data set, is 456.63nm. A Planck peak at 5850K, my favorite Planck temp. fit, is 495.38. Oddly enough, the actual peak if converted to photon flux is in the yellow at 585nm, IIRC. [Doing an interpolation on Thuilier's data set, the peak is slightly higher at 605nm (orange).] However, these are not peaks, they are pimples. :) The actual photon flux curve is almost flat up to the blue-green region.


Makes sense about the integrated flux, thanks. :) I don't know why I didn't think of that.... I doubt many people realize just how much the temperature variation really is across the disk.

George
2007-Sep-18, 11:27 PM
I said this before in this thread, but it seems to me that the light coming from the Sun has to be the very definition of white. Hold a piece of white paper in direct sunlight, and what color does it look? How do we measure white if not as compared to sunlight?
A perfectly good example, too. [There is some variance, however, due to the strong color constancy effect that causes objects that are near white to appear white.] White clouds and a white Moon also serve nicely as examples of a white light source illuminating them.

[Any idea, therefore, how the Sun got classified as a yellow star?? :doh:]

Your example only works for terrestrial color determination of the Sun.

Since our atmosphere removes over 25% of the total radiation from the Sun, though a little less in the visible portion, it is not clear just what color the Sun would appear if we could observe it in space at a level accomodating to our eyes. If it appears white down here and the larger extinctions are in the blue, will adding these blues and other colors back into it give us a bluish color? My original hope was a "yes", but it is now a "no". Yet, the central region of the Solar disk just might have a bluish tint since its observed temperature is about 6390K. [Added: the odds seem highly unlikely that this might be true, as per my post #108]

Hornblower
2007-Sep-19, 01:49 AM
My educated belief is that direct sunlight out in space will look white.

If we could look at a swatch of it alongside the normal appearance at sea level, of course it will have a slight blue tint. What we must not forget is that our brain cells have a phenomenal ability to adapt to the hue of the ambient light over a wide range of incandescent blackbody values. My car windows have a light blue tint, but if I am driving with all the windows closed, sunlit snow looks pure white unless I open a window and have the unfiltered light for side by side comparison. That's good enough for me.

George
2007-Sep-19, 02:59 AM
If we could look at a swatch of it alongside the normal appearance at sea level, of course it will have a slight blue tint. What we must not forget is that our brain cells have a phenomenal ability to adapt to the hue of the ambient light over a wide range of incandescent blackbody values.
Yes, our ability to see color correctly under unusual lighting, and incorrectly in certain cases, such as when other compartive colors are present, is quite surprising. Edwin Lamb is famous for his "Mandarin" color experiment (http://www.hhmi.org/senses/b140.html).

However, the Sun is not illuminated by other light sources, nor is it adjacent to other colorful objects. In the blackness of space, if it is a yellow star, it will look yellow. A yellow light bulb will not look white if it "burns brightly in the night" and it is alone in the dark.

Our eyes seem to have evolved to use existing sunlight as the standard for pure white. This is not the same sunlight, however, as seen from space, which is different because of the color loses due to atmospheric extinctions. Nevertheless, the difference does not appear to be enough to alter the net color result if we could see the Sun from space with a properly reduced intensity.


That's good enough for me. :hand: Don't regress to the complacent mean; we need more volunteer heliochromolgists. ;)

Hornblower
2007-Sep-19, 02:19 PM
Yes, our ability to see color correctly under unusual lighting, and incorrectly in certain cases, such as when other compartive colors are present, is quite surprising. Edwin Lamb is famous for his "Mandarin" color experiment (http://www.hhmi.org/senses/b140.html).

However, the Sun is not illuminated by other light sources, nor is it adjacent to other colorful objects. In the blackness of space, if it is a yellow star, it will look yellow. A yellow light bulb will not look white if it "burns brightly in the night" and it is alone in the dark.

Our eyes seem to have evolved to use existing sunlight as the standard for pure white. This is not the same sunlight, however, as seen from space, which is different because of the color loses due to atmospheric extinctions. Nevertheless, the difference does not appear to be enough to alter the net color result if we could see the Sun from space with a properly reduced intensity.

:hand: Don't regress to the complacent mean; we need more volunteer heliochromolgists. ;)

Now we are talking about two different visual situations. In my previous post the Sun was the source of the ambient light that filled my peripheral vision. My experience was that my visual system had plenty of adaptive range to adjust to the blue tint of the car windows and see it as white.

If we look at the Sun through a neutral filter out in space from blacked-out surroundings, it will be like looking at defocused stars at night. Under these conditions our visual system seems to go into a default mode in which a G star is slightly yellow, an F more or less white, an A such as Vega "cold white", and a B such as Orion's belt stars slightly blue. This is my experience with a telescope near the zenith in the clearest sky conditions. In smoggy conditions during the summer they shift toward more of a yellowish tint.

The wild card here is the amount of atmospheric yellowing. My attempt at finding some typical ballpark figures for this has been a beastly job. I can find technical writing on the topic, but it is written for fellow professionals under the assumption that they already know all of the nomenclature, and it deals more with fine tuning of the calculations than in finding the initial rough figures, which presumably are already known. My best estimate is that the yellowing at the zenith is comparable to the difference between F and G stars.

If this is reasonably close, the Sun should look about the same out in space as an F star looks from sea level. I stand by my belief that it would look white.

George
2007-Sep-19, 03:46 PM
Now we are talking about two different visual situations. In my previous post the Sun was the source of the ambient light that filled my peripheral vision. Yes, this is important because what we can see here is easily testable. My work is to determine the color of the Sun as seen in space and at a normal photopic intensity level. I have not found any "true color" imaging attempt at this. Of course, someone is bound to eventually go up in space and, with the help of a strobe or neutral filter, simply look at the darn thing and resolve this color conundrum. :)


If we look at the Sun through a neutral filter out in space from blacked-out surroundings, it will be like looking at defocused stars at night. Under these conditions our visual system seems to go into a default mode in which a G star is slightly yellow, an F more or less white, an A such as Vega "cold white", and a B such as Orion's belt stars slightly blue. This is my experience with a telescope near the zenith in the clearest sky conditions. Hmmmm, that is interesting and I wonder why? Had you said G stars, as point sources, appear yellow to you and others, I would argue that the eye's fovea might cause a slight yellowing effect since there are no blue color cones in the fovea. Defocusing, however, should eliminate this possibility. [Surprisingly, the eye's constant movement counteracts this potential blue reduction, or yellowing, but I don't know how fully it handles stellar observations.]

When I looked at 18 Sco, the best known Solar twin, it appears white to me. Do you recall if your yellowish G stars were closer to the G9 class, or G2?

What is odd about your yellowing for the Sun [class of stars, G2] is the fact the solar projections from solar telescopes do not reveal any yellowing, unless some filter is used or, perhaps, smog has rolled in. Even the much lower temperature limb region (5000K) is still white. [On another realted thread, you may have seen the image I obtained from Kitt Peak's McMath-Pierce unfiltered scope.]

In an early attempt to determine the Sun's color, I glued a pencil to a paper plate, painted the thing black, and cut a ~ 0.5mm slot in it. [Appropriately, I coined it SAD (Solar Attenuation Device) and the operator a SAP (SAD Application Personell). No sense getting too serious about this colorful topic. :) It's more fun than hard science.] After spinning it using an electric drill, I was able to reduce the Sun's intensity by 99.9% and the Sun looked white to me, though it was still very bright, if not too bright. [For those who don't know know, once all the color cones are saturated, white is the color that would be seen, even if it wasn't a white object.]


The wild card here is the amount of atmospheric yellowing. My attempt at finding some typical ballpark figures for this has been a beastly job. Yes, I am almost sure there are extinction correction factors for various AM (Air Mass) levels, but I have never sought them. Since I have the spectral irradiance data for a few different air masses, I essentially have all I need. Some Solar data exists that is in increments of only a few thousandth of a nm!


My best estimate is that the yellowing at the zenith is comparable to the difference between F and G stars. Do you find yellowing regardless of telescope and mountain top, assuming good atmospheric conditions? I'm a little puzzled with this view. Perhaps a survey poll might help. [We did one on the Sun's color for midday quick glance views with a result of yellowish-white winning, though white was close.]


If this is reasonably close, the Sun should look about the same out in space as an F star looks from sea level. I stand by my belief that it would look white. I'll bet you're right! :)

umop ap!sdn
2007-Sep-19, 07:05 PM
Unfortunately, your result is still coincidental since you only selected three wavelengths to represent all the wavelengths. Assuming you used the same ones for blue, green, and red, you should have calculated a flux at 1 a.u. of:

@ 620nm = 2.58 x 1022 photons.
@ 540nm = 3.04 x 1022 photons.
@ 460nm = 3.20 x 1022 photons.

This gives a ratio of 1/1.18/1.24 for the respective wavelenghts above. However, this comes from the Planck equation. The actual irradiance data will give a ratio of 1/1.03/1.10.

Another consideration is in regard to those color choices as they relate to our vision. 460nm is on the edge between violet and blue, 470nm would be better. 530nm is in the center of green. 620nm is regarded as orange, or orange-red, by some color sites (including one from NASA). 680nm is closer to midrange in red.

Using these wavelengths [460/530/680] gives a ratio of 1/1.07/1.08, respectively. The differences are somewhat significant compared to the Planck values.
I can certainly recheck my numbers and attempt to use a wider range of wavelengths. :) (Will have to wait until I get off work, and hopefully by then my internet access at home will be functional again.)


However, these are not peaks, they are pimples. :) The actual photon flux curve is almost flat up to the blue-green region.
Well we're talking about an almost white light anyway, so the small variations in the brightness by wavelength make a difference as far as what hue we're talking about at such low saturation. :) Besides, you've shown that changing the selection of wavelengths causes noticeable changes in the final result.

I also suspect Hornblower is correct about F and G stars.


I would argue that the eye's fovea might cause a slight yellowing effect since there are no blue color cones in the fovea.
Naah, the blue-yellow process requires all three kinds of cones. Since blue cones are lacking the only color opponent process the fovea picks out is red-green. I've seen myself how a green object can look blue, a yellow object pinkish gray, or a purple object brown, and it's kind of a strange phenomenon to experience. :D

George
2007-Sep-19, 09:05 PM
Well we're talking about an almost white light anyway, so the small variations in the brightness by wavelength make a difference as far as what hue we're talking about at such low saturation. :) Besides, you've shown that changing the selection of wavelengths causes noticeable changes in the final result. Small changes can make a difference in color, but this is not so much true for objects that appear white. For example, the spectral irradiance of the Sun in the central zone at 6390K is different than that of the limb at 5000K, yet both regions look white, this is after atmospheric extinctions.


Naah, the blue-yellow process requires all three kinds of cones. Since blue cones are lacking the only color opponent process the fovea picks out is red-green. Yes, that was what I was saying, though I know only a little of how the eye works. Of course, since we are observing a relatively large disk in apparent size, the fovea limitation, if there is one, should not be a problem.

Hornblower
2007-Sep-19, 10:11 PM
My G example was Capella, which is a composite of G5 III and G0 III. It should be slightly yellower than a G2 V, and I could barely see a trace of yellow.

If my hunch about the default color vision mode at night is correct, we could be affected by yellowing of our eye lenses with age. I remember seeing an article about this effect in Sky and Telescope some years back, in which older observers reported different perceived colors of stars, but I cannot remember the details. I will search my back issues for it.

If it stays clear tonight I will do some spot checks to follow up on this discussion.

George
2007-Sep-20, 12:22 AM
I wonder if the early lens coatings had a yellowing effect? I had not heard that our eye's lens could do this, but I wouldn't be shocked.

18 Sco is still pretty high early tonight. It is at the northern most edge of the constellation, 5.7 deg WNW from Han. 16 Sco is nearby and is a bluish A3 star. Both are about 5.4 mag.

Hornblower
2007-Sep-20, 03:08 AM
My spot checks tonight, under sparkling clear sky, looking at defocused bright stars in my Celestron 8:

Polaris, type F8 Ib, published color index +0.6
Slightly warm white

Altair, A7 V, +0.2
Neutral white

Vega, A0 V, 0.0
Cold white, perhaps hint of blue

18 Sco lost in trees, unobservable

Back to the eye lens topic. My father had cataract surgery on one eye at age 65, and he reported incredible color contrast. White objects looked blue in that eye and yellow in the other one. My mother reported the same thing after cataract surgery in her late 70's.

George
2007-Sep-20, 06:46 PM
My spot checks tonight, under sparkling clear sky, looking at defocused bright stars in my Celestron 8:

Polaris, type F8 Ib, published color index +0.6
Slightly warm white

Altair, A7 V, +0.2
Neutral white

Vega, A0 V, 0.0
Cold white, perhaps hint of blue I will look at these tomorrow night with my Celestron 8. :)


18 Sco lost in trees, unobservable. I would have said green. ;)

You have probably seen this before, but I like using it since it makes an obvious statement. It is of the Southern Cross.

http://img369.imageshack.us/img369/7200/seipsscrosspu1.th.jpg (http://img369.imageshack.us/my.php?image=seipsscrosspu1.jpg)


Courtesty of Stefan Seip (http://www.photomeeting.de/astromeeting/_index.htm).

They hot ones look white until out of focus. Even the red ones look redish white. [Acrux, B1, is the lower right, and Gacrux, M4, is the top star.]

I don't know if color enhancement was used, admittedly.

George
2007-Sep-22, 11:05 PM
Last night's results using the defocusing technique:

Polaris appeared white, but at certain out-of-focus apparent surface brightnesses, it looked either a little, very little, greenish-yellow (dirty yellow, perhaps) white or, at times, greenish-blue white.

Altair appeared as bluish white, but the blue was very faint and was not noticeable at most of the defocus settings. The air mass value was AM1.2.

I took an image of Altair using a Canon 30D set for 5800K. Then I super saturated it to see what colors might be there.

http://img301.imageshack.us/img301/5601/altairsupersatba1.th.jpg (http://img301.imageshack.us/my.php?image=altairsupersatba1.jpg)

Perhaps the atmospheric extinctions of more of the blues might help explain this odd blue-green result.

Vega revealed a distinctive, though faint, bluish-white color at AM1.1.

The following image of Vega is unenhanced:
http://img301.imageshack.us/img301/5493/vegadu5.th.jpg (http://img301.imageshack.us/my.php?image=vegadu5.jpg)

This is not the result I should have gotten, right? I don't even have blue eyes. :) I don't know what is wrong and I am reluctant to blame the Canon 30D.

You seemed to see a hint of blue, too. Perhaps we should get others to help in this color conundrum by starting with Vega. We should come up with a snappy term for a "color pest", or maybe asterochromologist will do the job more respectfully. :)

Hornblower
2007-Sep-23, 11:47 PM
Last night's results using the defocusing technique:

Polaris appeared white, but at certain out-of-focus apparent surface brightnesses, it looked either a little, very little, greenish-yellow (dirty yellow, perhaps) white or, at times, greenish-blue white.

Altair appeared as bluish white, but the blue was very faint and was not noticeable at most of the defocus settings. The air mass value was AM1.2.

I took an image of Altair using a Canon 30D set for 5800K. Then I super saturated it to see what colors might be there.

http://img301.imageshack.us/img301/5601/altairsupersatba1.th.jpg (http://img301.imageshack.us/my.php?image=altairsupersatba1.jpg)

Perhaps the atmospheric extinctions of more of the blues might help explain this odd blue-green result.

Vega revealed a distinctive, though faint, bluish-white color at AM1.1.

The following image of Vega is unenhanced:
http://img301.imageshack.us/img301/5493/vegadu5.th.jpg (http://img301.imageshack.us/my.php?image=vegadu5.jpg)

This is not the result I should have gotten, right? I don't even have blue eyes. :) I don't know what is wrong and I am reluctant to blame the Canon 30D.

You seemed to see a hint of blue, too. Perhaps we should get others to help in this color conundrum by starting with Vega. We should come up with a snappy term for a "color pest", or maybe asterochromologist will do the job more respectfully. :)
Visually you and I seem to be pretty consistent, with your eyes seeing slightly more blue than mine.

The saturation-boosted image of Altair suggests that the camera's original signal had a slight excess of green, and for all we know it might have been by design. If I were the manufacturer, my inclination would be to tweak its spectral response to get good overall tints on landscapes, cityscapes, human faces, etc. in typical daylight conditions. Possible use of the camera as an astronomical photometric tool would be way down on my list of priorities. An image on the screen that looks like a direct view of an incandescent blackbody does not actually have a continuous spectrum like the original. It consists of three color bands with some blind spots between them, something I can see easily by looking at a white line on the screen through a prism. If a colored object had some critical components in the blind spots, it might be slightly discolored when a white object is perfect. Tweaking for the colored object, if it is considered more important, would throw the white off slightly, and it could become easily visible if exaggerated by artificially boosting the saturation.

Your image of Vega is no surprise to me. I would expect it to have enough more blue to overwhelm any systematic green bias there might be.

George
2007-Sep-24, 03:51 AM
The saturation-boosted image of Altair suggests that the camera's original signal had a slight excess of green, and for all we know it might have been by design. If I were the manufacturer, my inclination would be to tweak its spectral response to get good overall tints on landscapes, cityscapes, human faces, etc. in typical daylight conditions. Yes, this was especially true for films; Fuji film favored green and Kodak favored red, IIRC.


An image on the screen that looks like a direct view of an incandescent blackbody does not actually have a continuous spectrum like the original. It consists of three color bands with some blind spots between them, something I can see easily by looking at a white line on the screen through a prism. Yes, it is hard to get objects to maintain a nice 5800K temperature. :) Lamps have a CRI rating where the best rating matches the Sun's illumination.


If a colored object had some critical components in the blind spots, it might be slightly discolored when a white object is perfect. Tweaking for the colored object, if it is considered more important, would throw the white off slightly, and it could become easily visible if exaggerated by artificially boosting the saturation. Yes. The boosting software could be a problem, too. I am not sure how accurate the camera's software is at color control, but they do a remarkable job today in simulating what we see.

Regardless, I too am doubtful of the green result for Altair.


Your image of Vega is no surprise to me. I would expect it to have enough more blue to overwhelm any systematic green bias there might be. I have started a poll thread (http://www.bautforum.com/astrophotography/65096-your-color-vega.html) in the astrophotography forum hoping some will actually go out and observe it and let us know the color result they see.

Hornblower
2007-Sep-24, 04:50 PM
<a href=http://web-hosting-plan.fyrt.info/budget-hosting-plan-web.html> budget hosting plan web </a>,etc.

What does this have to do with the topic of this thread?

tony873004
2007-Sep-24, 05:45 PM
What does this have to do with the topic of this thread?
That's just spam. I hope the mods delete it. I think there's a filter on this board so people with 1 post can't post urls. I guess this is someone's attempt around this.

In Astrobiology class we discussed why the leaves of plants might be green, the same as the sun's peak output. I think the conclusion was that by reflecting the sun's peak energy, it prevents the leaves from overheating. In hindsight, this makes me wonder why we don't get different color leaves in cold climates.

I forget if this was a serious theory, or just off the top of our heads babbling (a lot of that happened in this class).

There was also something called the "red edge" and how our plants look red in IR photos. I forget if it had anything to do with the sun's output. I'm trying to Google for it now, but with no luck. It's very interesting stuff.

George
2007-Sep-24, 07:30 PM
In Astrobiology class we discussed why the leaves of plants might be green, the same as the sun's peak output. I think the conclusion was that by reflecting the sun's peak energy, it prevents the leaves from overheating. I recall an article from earlier this year addressing this issue, but I can't find it. I think their conclussion was different.

Regardless, if they say the Sun peaks in green, they are wrong. [However, our peak responsiveness of our eye is green.] There essentially is no peak in the solar flux as shown in an earlier post (http://www.bautforum.com/627924-post104.html). The photon flux peak -- since one does exist though it should be called a pimple, IMO -- is actually in the yellow [*gag*] portion of the specturm as seen from space. Terrestrially, the light flux pimple will rise in the orange region, near yellow.