Fire and dispersion techy help please?

[beryl]

Hi,Serg: Thank you for the compliment. I will search for more about 'cone beam analysis', or is this a term you created?

To all:
I want to explain my use of the term 'dispersion angle'; I use it to describe the angle between colors of the spectrum in a dispersed beam, which changes with refraction.
'Dispersion' itself is a property of the material and does not change.
Does anyone know a specific technical term for 'dispersion angle' or 'spectral fan'? Of course, in 3D, this becomes a cone. Is that what you are talking about with 'cone beam analysis', Sergey?

 

[Serg]

----------------
On 9/17/2004 5:50:42 AM beryl wrote:

Hi,Serg: Thank you for the compliment. I will search for more about 'cone beam analysis', or is this a term you created?

To all:
I want to explain my use of the term 'dispersion angle'; I use it to describe the angle between colors of the spectrum in a dispersed beam, which changes with refraction.
'Dispersion' itself is a property of the material and does not change.
Does anyone know a specific technical term for 'dispersion angle' or 'spectral fan'? Of course, in 3D, this becomes a cone. Is that what you are talking about with 'cone beam analysis', Sergey?
----------------

re: is this a term you created?
May be. try see beam tracing, I think it is old term

re: Does anyone know a specific technical term for 'dispersion angle' or 'spectral fan'?

I do not know, Have same problems
re: Is that what you are talking about with 'cone beam analysis', Sergey?

Not only dispersion angle produce cone in 3D. Angles of light( virtual facet) source( or purple in reverse tracing) is very important too.

interaction dispersion angle and angle of light produce real cone beam( and distribution color ) . Some phenomena's distort cone.

 

[Garry H (Cut Nut)]

Beryl I mentioned what Sergey is discussing early on - first page of this thread.

Sergey slow down

- your english is very bad on the last thread - I know what you are saying and I can not understand you!!!

You might rembeber I said that GIA's fire paper was fataly flawed because they used paralel rays of light - these only occur with sunlight.

Normal light COMES in beams - or Sergey's word is CONES.

I still wanted to continue with your project, and we now can see that GIA's method is also flawed because even using a ray of parallel light - you have shown that dispersion is not reversable! Amazing!

So now that Sergey is back from vacation - perhaps we might discuss Beams or cones of light passing through virtual facets and the size of pupils etc etc.

I do not beleive anyone else has considered this methodology, and I know that Sergey has spent a lot of effort in this new domain - the results I have seen are stunningly wierd. But they seem to back up one of my old lines - Brilliance is the enemy of fire.

I want to change the quote to this:
Brilliance can be the enemy of fire.
OR
Brilliance is often the enemy of fire.

 

[Serg]

----------------
On 9/18/2004 9:12:22 AM Garry H (Cut Nut) wrote:

Beryl



I still wanted to continue with your project, and we now can see that GIA's method is also flawed because even using a ray of parallel light - you have shown that dispersion is not reversable! Amazing!


----------------

Hi, Garry

This your statement is not correct completely. Of course GIA DCLR method is not correct and angles of dispersion depend from direction. But I think color( or dispersion) is reversible. We can use back tracing for calculate Fire. Different angles of dispersions ( for different directions) are not problem.

 

[Garry H (Cut Nut)]

Please explain Sergey?

 

[Garry H (Cut Nut)]

Sergey has asked me to answer my question with this question:

How we should calculate(grade) Fire by back raytracing?
How we should calculate(grade) Fire by direct raytracing?

I do not think either solution can work, and here is part of the reason why Sergey is right about beams or cones:

I have spent 1 hour looking at my CZ’s in my usual café (many of you know I always carry the little black box with a good and bad cut CZ in my pocket).

(All the other coffee addicts already know I am crazy)



The left side main facet – from the left side of facet only –first came pale red, orange then the most intensity (because the halogen globe 2.5M above the table is making the entire facet brightly illuminated) then (greenish) blue through to blue and finally fainter violet (because now the dispersion is only coming from the right side of the facet).



But if I move my head vertically, starting from the brightest central position, then as I move my head up and down, the same thing happens – red at top, bright in middle, paler blue at lowest head position.

This means there is a cone of fire emenating from this facet!
Not a strip of fire, but a cone of fire, with a bright middle section.

Ray tracing can not explain this. Beam or Cone lighting does.

 

[Garry H (Cut Nut)]

This is a schematic of the experiment.

 

[beryl]

There has been correspondence offline; some of which must be shared with the readers; it depends on the definition of a ray ...
If a 'ray' is a thread of only one color, then its transmittance path will be the same in both directions, since it has no dispersion (the partial reflections will be different, however).
In this case there is no such thing as a 'ray' of white light; it is a bundle of rays traveling together which separate upon refraction because of their different refractive indices.
'Beams' are bundles of rays and thus a 'ray' of white light is really a 'beam'. In 3D the dispersion I have illustrated here becomes a cone. If the source is not round, such as a polygon formed by facet edges, then the cross-section of the 'cone' will not be round, as is usually implied by the word; rather it will be a slender pyramid.

 

[Garry H (Cut Nut)]

I hear what you are saying Beryl.
Thanks for sticking with it.
Me; I have reached my outer limits.

But I do understand what you have said.

But for me, this is a quest to understand enough to know how and why various approaches to quantifying diamond beauty work or do not work.

This image shows a 'ray' or bundle of different frequencies of light travelling parallel, (which is not likely to happen in nature) being dispersed as they leave the crown facet of a diamond.

1. The ray has been dispersed a tiny bit as it entered the table (at a high angle of incedence) and has been dispersed agian - but this time - at an angle approaching 90 degrees to the original ray path plane of entry. If I have this right - it means the ray will be broken into different dispersed rays after entry to diamond, and then each or those dispersed rays will be then dispersed again.

Now this gets too complicated to model in 2D - and certainly this DiamCalc image is not going to make it possible for us to see the cone of dispersion that I could see in the cafe yesterday.

I think in our example with the prism - we would not see a comne of colors - it would be more or less a 2D rainbow. The reason I saw a 3D cone was because the 'ray' path had been 'twisted around' inside the stone - and because the intial entry was at a very steep angle (about 72 degrees incidence from normal into the table - 29% reflectance).

Many other dispersions are more 2D as you move your head around looking at them.

2. If a bundle of rays of white emmissions from a small halogen light globe 2.5m away enter the table, they are not really parallel. At that distance if i look at the stones table and line it up with a halogen light that is that far away - the reflection is almost the same size as the table. Therefore the lamp size at that distance was big enough to swamp the entire crown facet with intensly bright light had the light been refracted back with out being dispersed.

I saw a bright spot only in the middle of the facet - and red on the left and top and blue / violet on the right and lower areas of that bezeel facet.

Was I seeing white because of the interference like this with slightly divergent and convergent rays getting all mixed up?


Anyway, at this time, I think the simple ray tracing techniques are insufficient to make a computer based grading system.

 

[Garry H (Cut Nut)]

This is relevant

 

[strmrdr]

Has anyone addressed the function of facet junctions and near facet junctions play in creating fire?

 

[strmrdr]

.

 

[strmrdr]

The only difference is that one trace just barely misses the junction going in and the other hits it.

 

[Serg]

----------------
On 9/21/2004 6:45:41 AM beryl wrote:

If the source is not round, such as a polygon formed by facet edges, then the cross-section of the 'cone' will not be round, as is usually implied by the word; rather it will be a slender pyramid.

----------------

Hi Bruce,

See some pictures.

Dist between purple and diamond = 250 mm
Purple = 4mm
Dist between diamond and sphere = 2000 mm
Diameter of source light = 0 mm
Diameter of diamond = 6 mm
Number of reflections = 4
Is it a good scale?

 

[Serg]

Dist between purple and diamond = 2500 mm
Purple = 4mm
Dist between diamond and sphere = 1000 mm
Diameter of source light = 0 mm
Diameter of diamond = 6 mm
Number of reflections = 4

 

[Serg]

Dist between purple and diamond = 250 mm
Purple = 4mm
Dist between diamond and sphere = 1000 mm
Diameter of source light = 30 mm
Diameter of diamond = 6 mm
Number of reflections = 4

 

[Serg]

Dist between purple and diamond = 2500 mm
Purple = 4mm
Dist between diamond and sphere = 1000 mm
Diameter of source light = 0 mm
Diameter of diamond = 6 mm
Number of reflections = 4

Bigger Scale

Enough?

 

[michaelgem]

Hi Garry,

I checked out the pricescope threads you sent me. I had not seen any of them. Thanks for the ‘heads up’.

I have several ideas for input when I find the time to do a proper job of responding.

I believe there are solutions to these issues that can greatly simplify things and suffice for all intents and purposes. One of these is ‘reverse raytracing’ or ‘back tracing’

Back tracing is the best way to analyze light performance in gemstones.

If you want to handle the 'irreversibility' issue use various wavelength monochromatic rays, which are reversible.

Meanwhile, I agree with Sergey's following words:

"But I think color( or dispersion) is reversible. We can use back tracing to calculate Fire. Different angles of dispersion (for different directions) are not a problem."

Cheers,

Michael Cowing

www.acagemlab.com

 

[beryl]

Nice pics, Serg. I assume that the polygons are the shapes of the beam 'sources' in the gem.

 

[beryl]

I received the following by e-mail. It supports Sergey's statement, with which I agree if we redefine a ray as having only one color.
"My understanding ... there is (no) such a thing as a ray of white light. I know each ray (has) its own wavelength and that at least 3 rays ... are needed to make white-appearing light. Finally, since those ... rays, of ... differing wavelengths, have ... different RIs, they can enter together, but they won't travel together. Thus we don't have rays of white light entering,
etc. This all means that each ray's path (power is different) is completely reversible for a particular wavelength: the path forward is identical to the path backward."

 

[Serg]

Sorry for misprint.

I printed Purple instead Pupil.

Thanks, Garry

 

[beryl]

Responding to post by Garry:

”But for me, this is a quest to understand enough to know how and why various approaches to quantifying diamond beauty work or do not work.” - Garry
--------------------------------
I think that we cannot quantify diamond beauty. We can describe what a diamond is but must then let the viewer decide what he/she prefers. - BLH

"This image shows a 'ray' or bundle of different frequencies of light travelling parallel, (which is not likely to happen in nature) being dispersed as they leave the crown facet of a diamond.
”1. The ray has been dispersed a tiny bit as it entered the table (at a high angle of incidence) ...” Garry
---------------------------------
Check my calcs again – dispersion ANGLE is greater if it LEAVES at greater angle. - BLH

"... and has been dispersed again - but this time - at an angle approaching 90 degrees to the original ray path plane of entry. If I have this right - it means the ray will be broken into different dispersed rays after entry to diamond, and then each or those dispersed rays will be then dispersed again. - Garry
---------------------------------
The beam is not dispersed ‘again’; it was dispersed (= dissociated into its component colors) when it entered, as each color ray was refracted differently. At exit, EACH of these colors is REFRACTED again, increasing the ANGLE between them .
Be careful with word ‘disperse’. ‘Dispersion’ is a property of the material. ‘Disperse’ is to separate a beam (by refraction difference) into its component colors, ‘dispersion angle’ is a term I used to describe the amount the colors are separated due to the geometry of the gem and the incidence of light to it. - BLH

" 2. If a bundle of rays of white emissions from a small halogen light globe 2.5m away enter the table, they are not really parallel. At that distance if i look at the stones table and line it up with a halogen light that is that far away - the reflection is almost the same size as the table. Therefore the lamp size at that distance was big enough to swamp the entire crown facet with intensly bright light had the light been refracted back with out being dispersed.
I saw a bright spot only in the middle of the facet - and red on the left and top and blue / violet on the right and lower areas of that bezel facet.
Was I seeing white because of the interference like this with slightly divergent and convergent rays getting all mixed up?". - Garry
---------------------------------
As you have already illustrated twice with the prism illustration by Octonus, you were seeing a beam which fit entirely within your pupil (I also mentioned it here). Adjacent smaller bundles within this beam are all dispersed similarly so that the reds of one cross the violets of the other, etc, and the common area in the center becomes a combination = white light again. The colors are seen separately only at the edges where they are less mixed-up – red at one side and violet at the other. Since green has the same wavelength as the average of white light it cannot be seen in a beam of any width. This is why we must use a slit in a spectroscope – to make the width of the beam as narrow as possible so that there is no crossover-mingling of colors (and we also collimate the light so that there is no crossing-over before it enters the prism). - BLH

"Anyway, at this time, I think the simple ray tracing techniques are insufficient to make a computer based grading system." - Garry
-------------------------
I think you are trying to include too much in a computer-based grading system. Again, as I said 2-3 years ago, and in Moscow recently, I think we are already there if we assign grades based on simple proportions - like AGS but including the ‘appeal continuum’ you defined as ‘BIC-TIC-FIC’. Use this for grading/appraising purposes; then let the buyer select the characteristics within this that he/she prefers .
The problem is futher complicated if you are trying to include fancies. For grading, stick to SRB's. Don't try to fit the ocean in a bucket. - BLH

 

[Garry H (Cut Nut)]

Beryl I will not try to carry the quote forward - it will be too messy. But my answers are:-

1. Sergey and I are hell bent on quantifying beauty. There will be ranges and tastes and styles and diversity of beauty. We will start with 'good' and 'bad'. We will manage to narrow down to not so good and not so bad. And then very good and so on.

2. The inbound ray was at a high angle of incidence (72 degrees) and the out bound was at a low angle (about 35).

3. The ray enters the diamond (and is dispersed) and what I tried to explain is that the exit dispersion was in a plane at about 90 degrees to the entry dispersion. That is why I could see a cone of dispersion - not a flat plane

. Does that make sense?

4. The light was not dispersed within my pupil - the dispersion at 60cm away had a very very wide 8cm or +3 inches. My pupil, even after a couple of glasses of wine, is not that wide

. I had to move my head from side to side, and in this case, because of the 90 odd degree 'twist', up and down.

This is my entire point to recounting this experiment - using your calculations the amount of angular dispersion of the rays.
For 72 degrees inbound - the angular dispersion is 0.438 degrees (from your chart) and that would be what causes the up down verticle dispersion. For the horizontal dispersion - left right - the exit angle is about 36 degrees (14 internal). Now at 60cm view distance can you please calculate how far I would be able to move my head and still see the dispersed light? I bet it is not near 3 inches?

So if that is the case then Sergey is correct - the diameter of the globe causes much more dispersion. i.e. white light comes in beams or cones and does not behave as rays in creation of dispersion. (I wish I was smarter and could explain this better).
You can see that really clearly in Sergey 3rd image above with the 30mm light globe vs the pin point globe.

5. And finally - re why have a computer based grading system?
Why not just use an HCA type approach?
Because HCA can not be used to design a beautiful desireable cut from any shaped piece of rough diamond.
Gabi has the ability and the brand value to be able to take a rough diamond and fashion a one off 'beauty'. We want to make an expert computer system that can do that too.

Imagine - free form, asymmetric, art cuts, that are truly beautiful to more than 1/2 the people on earth!

 

[beryl]

That's enough for me. Have fun.

 

[Serg]

----------------
On 9/21/2004 5:25:26 PM beryl wrote:

Nice pics, Serg. I assume that the polygons are the shapes of the beam 'sources' in the gem.----------------

Bruce,
It is good interpretation for 1st, 2nd and 4th images, if you use 4 ìì light source(instead pupil).
The interpretation of 3d image is more complex . I do not know good interpretation in the term " beam 'sources' in the gem:

 

[beryl]

Hi, folks
I just bought a TI-83Plus hand-held calculator ($130 on sale); it does a thousand things that don't interest me at the moment but it is programmable (like the TI-59 I had in 1980).
I have programmed it to solve the problem I have illustrated: INPUT = WEDGE (prism) ANGLE, ANGLE OF BEAM ENTRY
OUTPUT = EXIT ANGLE OF VIOLET & RED RAYS, DIFFERENCE BETW. THEM.
If anyone is interested I will send the steps to enter = very short and simple program. I can program another program for refraction and power, and other related problems if you wish.

 

[Serg]

----------------
On 9/17/2004 4:19:03 AM beryl wrote:

I do not understand 'cone beam analysis' as a method, and related things mentioned by Serg. Where can I read more about it? I think the work I have done here tells us some things.----------------

Hi Bruce,

I had remembered about DTI.

You can use http://www.diamondtechnologiesdti.com/raybeam.htm, for few information about beam tracing.

They use term: "Reflection tracing " now.

 

[beryl]

OLD SCIENCE:
I am reviewing the book "Gemstone Faceting' by Dmetro Andrychuk, published in 1977. This was brought to my attention by a few members of the US Faceter's Guild, and one of them kindly sent me a photocopy. I was unaware of this, and also of an article by him in Lapidary Journal in 1977; I subscribed to (and wrote for) LJ at that time but do not recall it.
In Section 4, "Color and Dispersion" (p.41-51) he discusses the comparative dispersion angle of a table-bezel beam according to the direction of its path through the gem. His analysis shows that the dispersion angle for bezel exit is greater than for table exit when the table beam is parallel to the gem axis. That is what I showed here, but only by specific calculations with a 20° prism.
So I did not show anything new. Mike Cowing recently spoke with Mr. Andrychuk's wife - he is now in his 90's and not well.
I hope he recovered the cost of printing the book.

 

[Blueman33]

THE CONCEPT DOESN'T SEEM REALISTIC TO ME BECAUSE LIGHT DOESN'T WORK THE WAY THE MODEL SUGGESTS.

I am not a scientist, but this doesn't make sense to me.

Long thread of which I skimmed. This may be off or alread been said, so I may have nothing to offer so will keep it short with two points.

First point.

The entering ray could not have the same intensity when it exits a diamond and when it enters a diamond because it has traveled and would not maintain the same form. Just as a flash light is more intense 2 feet away than 200 feet away. Not only has the beam traveled further, it is going through something which is denser than air so would dilute it's intensity even more than air.

A flashlight shined onto a reflective wall or window doesn't maintain its form, it dispurses. With all the angles and facets of a diamond it would disperse even more than a simple flat surface.

Like when the sun enters a prism and disperses colored light in several directions. You can't take the colored light and reverse it to put out a sun beam. So to study the 'reverse' is misleading as it seems to assume this thin light beam unaffected by travel or dispursion is what is exiting.

In fact if the sunbeam entered a diamond, when exiting it would not have the intensity to blind you as if you looked at the original sunbeam entering the diamond.

The exiting light beam is different than when it entered because the light is dispursed and diluted as it has traveled through the diamond.

SECOND POINT. The model doesn't seem realistic to me because a tiny toothpick thin beam of light is not what enters a diamond in the real world, it's a massive wide complex set of numersous light beams from the light source(s) and from everything around it reflecting light.

Maybe the study of diamonds and tiny thin light beams is useful. Again, I am not a scientist and know only a little about diamonds. However to me a study of the way diamonds react to light from the real world might be something to look into.

Just my $.02.

 

[beryl]

Blueman33:
Your points are well-taken and we are aware of them. We often pursue an academic approach in order to understand the principles involved; then we apply them to the real-world situation. In this case we wondered if the dispersion of the beam going one way was the same as the other way and found that it is not; this led to the significant post by RankAmateur re devices which measure 'fire' at the table.
1. Be careful with the term 'intensity' One beam may have less 'intensity' than another of the same 'power'. 'Intensity' = 'power' divided by cross-sectional area (recall burning something with a condensing lens). Sergey pointed this out a few years ago and added it to DiamCalc; note the illustration by Garry which shows power in one picture and intensity in the other. Area increases as a beam enters a gem but decreases again as it leaves; it is constant within the gem unless split by facet junctions (StormDr mentioned splitting by junctions).
In addition to absorption, which you mention, you must also consider partial reflections that always accompany refraction and diminish the 'power' of the transmitted beam as it enters and exits the gem. With diamond at least 35% of power is 'lost' this way (not lost, just visible elsewhere).
2. Note that Sergey has also criticized single-ray tracing; his work is now focused on effects with the broad beam, as you suggest. Work by GIA and MSU was based on light sources from all directions at once; only specific narrow beams reach the viewer's eye at any time, as illustrated nicely by Anton Vasiliev's software (available free on Internet or by request).