Axial Lights II (Ghostbuster?)

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

In contrast, placing planar glass in a converging or diverging section will result in reflections that map back to a different point on the subject. If that reflection is sufficiently intense and well focused, then it will result in a ghost. An illustration looks like this:
So, finally, having cleared my mind with concept of ghosting -- a dot on a subject gets projected at multiple locations, I have some very, very simple questions:

1. If I shine a laser onto a glass at 45 degree angle (same as shining straight to a 45 degree angled glass), and the glass is 1mm thick, with Rik's analysis, do you see multiple dots on projection plane, spaced at 1.414mm (square root of 2) apart? Is this really true for beamsplitters?

2. If I shine a laser onto a glass at different angle, say, 44.5 degress, do I get the same?

3. If I shine a laser at some extreme angle at the glass, do I even get a dot coming out of the other side of glass?

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

The 3rd question kinda brings out the real thing: in axial setup like mine, do all light rays reflected from a point on the subject pass through the glass, subsequently, contributing exposure, making a ghost? I remember some phenomenon that when light travel in a denser medium (hence I agree with Ray's idea of trying different material), they get trapped and never get out, I do not remember what this is called.

The bottom line is, when you image something through a setup like mine, are you expecting "normal" behavior of lights?

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

Your diagram shows only rays that will not produce ghosts. But that is because of the way they have been selected and drawn. Such a diagram can represent a certain level of conceptual understanding, but it falls rather short of a "reasonable theory".
Hahaha, I think your analysis is the very reasonable. Mine, on the other hand, to a beginner, is reasonable, I said that because I was expecting something at higher level -- good analysis with sounding theory, like Lou's, he got a diagram, a working scheme, and I have actually tried that idea (before this).

So, Rik, with your diagram, which is very intriguing and sound, and I am not disagreeing with you, but does it happen that way in real world? Or maybe those secondary beams are definitely there, just very weak?

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

One of the key observation from real world that is strikingly different from Rik's analysis -- lights reflected by the glass reaching subject are surprisingly parallel and directional, very similar to a laser (hence the analogue of pointing a laser), it seems almost directly coming down from the lens. This is why that if tiny part of the subject is not parallel to the image plane, it will show up as dark spot, if the cut is deep and sharp, it basically is total black, not even any gradient or slopping, producing very high contrast. I had to add some lights from around to lighten up the dark edges of lettering for coins.

So, this is very, very different from Rik's analysis where omnidirectional light source is assumed. In my setup case, lights from subject forming an image on sensor are those reflected lights almost also parallel to optical axis.

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

Sorry about repeat posts, I should have replied AFTER I did experiments :D

Now, come to think of it, I do not think Rik's second diagram is correct because it assumes omnidirectional light hitting the subject. In real world, it is not a cone like, more or less cylindrical[edit] with my setup. With my setup, with all proper flags, blocking out all lights (mostly omnidirectional) except those reflected from the glass, you only have highly directional, and importantly, parallel to optical axis lights hitting the subject.

Having said that, the more I looked at my original diagram, the more I think it represents real world better minus the "secondary" reflection part that I did not think of because I did not observe it. Of course, what I do not see does not mean it is not there, probably those are just weaker and hence my modified question:

"if you shine a laser at 45 degrees at a glass of certain thickness, do you see multiple dots on the other side of glass, and how intense are they?"

I do not think this is the case with my setup
Image

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

by "secondary" reflection, I mean the extra light beam in Rik's modified diagram of mine, not sure if it is intense enough if it does exists (in theory, yes) to form a visible ghost. The reason I am not sure if it exists is that I do not see multiple dots if I shine a laser (that does not mean they are not there).

Image

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

For the life of me, it is so hard to demonstrate anything with 1.1mm glass. So I searched on Youtube. Here is the video

In this video, the glass is really thick, using the width of glass -- from the direction of where laser is shined. I see two beams out of the SAME side as laser, ie, on the C side of my diagram, so this is fine. However, I could not find an incident where the "secondary" beam shows, ie, two or more beams coming out of the other side of the glass, maybe the glass is too thick (width of the glass, in the demo), but still do not see "them" on the wall. Maybe the video is not used to show that secondary beam on the other side of glass.

So, I think my original diagram represents the real world much better, and it is really what I observed, not sure why the secondary beam, by Rik's diagram, seems elusive, maybe someone with deep knowledge in optics can explain.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by rjlittlefield »

mjkzz wrote:
Fri Jan 21, 2022 5:22 pm
1. If I shine a laser onto a glass at 45 degree angle (same as shining straight to a 45 degree angled glass), and the glass is 1mm thick, with Rik's analysis, do you see multiple dots on projection plane, spaced at 1.414mm (square root of 2) apart? Is this really true for beamsplitters?
The spacing is not exactly that, because the angles change when the light enters the glass. But as to the multiple dots, the answer is simply "yes".

I set up a casual demo just now, consisting of a sheet of window glass, roughly 3 mm thick, a laser pointer with an aluminum foil pinhole, a white target, and a camera with macro lens, shooting a picture of the laser dot with and without the glass in place.
ReflectedRaySetup.jpg
ReflectedRayResult.jpg
2. If I shine a laser onto a glass at different angle, say, 44.5 degress, do I get the same?
Sure -- what would be special about 45 degrees?
I remember some phenomenon that when light travel in a denser medium (hence I agree with Ray's idea of trying different material), they get trapped and never get out, I do not remember what this is called.
This is called "total internal reflection", and it cannot be made to happen by shining light onto one of the two parallel sides. The reason is that any light entering the glass will be refracted to an angle that is too steep to be totally reflected on the other side. See https://en.wikipedia.org/wiki/Total_internal_reflection , especially figures 5 and 6 which show the angles.
3. If I shine a laser at some extreme angle at the glass, do I even get a dot coming out of the other side of glass?
Have you tried it? Or for that matter, have you tried looking through a window at a steep angle? Is there any angle at which the window glass becomes opaque?
Or maybe those secondary beams are definitely there, just very weak?
Yes, just so!

We can calculate how weak they are. If the glass surface is untreated, then at 45 degrees the reflection is certainly less than 10%, probably closer to 5%. Even if it's 10%, then the reflection of the reflection will be only about 10% squared, roughly 1% of the light transmitted straight. If it's only 5% at one surface, then the double bounce is more like 0.25% of the light transmitted straight.

As a side note, this is very different from the issue of ghosting with a second-surface mirror, where the ghost from the front surface needs only one bounce and is correspondingly brighter.

Anyway, yeah, the double-bounced ray is quite weak, and that one fact is all that's needed to explain why it's not a problem except in the most demanding situations.

All your other convoluted explanations are not needed and mostly aren't even right, but I have no enthusiasm to try explaining where the flaws are.

One point does merit further comment:
lights reflected by the glass reaching subject are surprisingly parallel and directional
Imagine yourself at the subject's location, looking upward at the glass. What you see looks just like the light source, as if it were positioned above shining down, and possibly clipped by the edges of the reflecting glass. If light striking the subject is surprisingly parallel and directional, then either that's what the light source is putting out to be reflected, or the reflector is small and/or distant so it's clipping the reflection. So, if you want less parallel and directional illumination, you need a larger/closer reflector, and a light source that is correspondingly large and diffuse, so as to fill the reflector as seen by the subject.

In any event, nothing about my analysis assumed omnidirectional light sources. If some of the black paths are not actually occupied, the red paths still act the same.

--Rik

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by rjlittlefield »

rjlittlefield wrote:
Fri Jan 21, 2022 9:07 pm
We can calculate how weak they are. If the glass surface is untreated, then at 45 degrees the reflection is certainly less than 10%, probably closer to 5%. Even if it's 10%, then the reflection of the reflection will be only about 10% squared, roughly 1% of the light transmitted straight. If it's only 5% at one surface, then the double bounce is more like 0.25% of the light transmitted straight.
There is an interesting and possibly relevant point hiding in here.

The calculation described above assumes that the glass surfaces are untreated, and it concludes that the double-bounce is a tiny fraction of the light transmitted straight through. Assuming 5% reflectivity per surface, we get more precisely that the straight-through transmission would be 0.95*0.95, while the double-bounce transmission would be 0.95*0.05*0.05*0.95. The ratio of those values is the 0.25% previously mentioned.

However, if we substitute glass that has been half-silvered on one side and untreated on the other, then the ghost image gets relatively a lot brighter. With one surface of the glass half-silvered, the straight-through transmission becomes 0.95*0.5, while the double-bounce transmission becomes 0.95*0.5*0.05*0.5. The ratio between those values is 2.5% -- an order of magnitude stronger than without the silvering.

When talking about reflector systems like this, people often jump on the idea that for maximum light through the whole system they should have a 50% beamsplitter. That is true, but when there is potential ghosting due to double-bounce, using a half-silvered mirror makes the ghost relatively much more visible.

--Rik

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

I set up a casual demo just now, consisting of a sheet of window glass, roughly 3 mm thick, a laser pointer with an aluminum foil pinhole, a white target, and a camera with macro lens, shooting a picture of the laser dot with and without the glass in place.
Thanks Rik, interesting, I will try to shine the laser at 45 degrees, according to geometry, the secondary dot should be 1.414*thickness of glass. Yours appear to be far apart due to shining angle.
Sure -- what would be special about 45 degrees?
Nothing special, I tend to do things gradually, from slight change to extreme change. What I am trying to do is to lead readers to think about it, along with your diagram (more on that later). With that glass there, NOT all rays in your diagram will reach camera, so image captured will be different than, say, similar setup without the glass. This difference is significant enough than, say secondary reflection.
We can calculate how weak they are. If the glass surface is untreated, then at 45 degrees the reflection is certainly less than 10%, probably closer to 5%. Even if it's 10%, then the reflection of the reflection will be only about 10% squared, roughly 1% of the light transmitted straight. If it's only 5% at one surface, then the double bounce is more like 0.25% of the light transmitted straight.

As a side note, this is very different from the issue of ghosting with a second-surface mirror, where the ghost from the front surface needs only one bounce and is correspondingly brighter.

Anyway, yeah, the double-bounced ray is quite weak, and that one fact is all that's needed to explain why it's not a problem except in the most demanding situations.
OK, 1% means 6.7 stops and 0.25% means 8.7 stops, I would say, the ghost is still busted. :D
All your other convoluted explanations are not needed and mostly aren't even right, but I have no enthusiasm to try explaining where the flaws are.
:P Maybe at least point out what my convoluted stuff are, this way, readers can learn, too. See, I enjoy this kind of discussion and grateful to learn. Thanks in advance.
Imagine yourself at the subject's location, looking upward at the glass. What you see looks just like the light source, as if it were positioned above shining down, and possibly clipped by the edges of the reflecting glass. If light striking the subject is surprisingly parallel and directional, then either that's what the light source is putting out to be reflected, or the reflector is small and/or distant so it's clipping the reflection. So, if you want less parallel and directional illumination, you need a larger/closer reflector, and a light source that is correspondingly large and diffuse, so as to fill the reflector as seen by the subject.


I think your way is one way, this is also related to what I have to discuss about your diagram. :D
Last edited by mjkzz on Sat Jan 22, 2022 4:45 am, edited 1 time in total.

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

In any event, nothing about my analysis assumed omnidirectional light sources. If some of the black paths are not actually occupied, the red paths still act the same.
This is what I am saying, without omnidirectional light source, just like what you acknowledge if some black paths are not there, you will NOT get a diagram like you drew. So I think it is kinda at least misleading if people do not read carefully. So, it, the diagram in that form, has a lot to do with omnidirectional light sources.

With colimated light sources, you would only get one path in that diagram.

Back to my setup, if you look at it, I have a tube blocking a lot of light from sides and the tube is relative tall. Though the LED panel has diffuser on it, and lights are being reflected downwards, it creates sort of light source with smaller divergence, and yet, the light is about 1 meter away. This is reflected in what I observed.

To adjust this, I can change the diameter of the tube or height of it, the difference is significant, instead of changing size of light/reflect or distance.

[edit]
Furthermore, because of the tube, the cone in your diagram will be narrower than without, light reaching camera are cut further, therefore, I can sort of reduce your diagram into mine. As matter of fact, I was afraid to draw any path not "head on", but of course, mine lacks the secondary one. But really, what I observed behave almost like what I drew.
Last edited by mjkzz on Sat Jan 22, 2022 4:43 am, edited 1 time in total.

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

rjlittlefield wrote:
Sat Jan 22, 2022 12:29 am
rjlittlefield wrote:
Fri Jan 21, 2022 9:07 pm
We can calculate how weak they are. If the glass surface is untreated, then at 45 degrees the reflection is certainly less than 10%, probably closer to 5%. Even if it's 10%, then the reflection of the reflection will be only about 10% squared, roughly 1% of the light transmitted straight. If it's only 5% at one surface, then the double bounce is more like 0.25% of the light transmitted straight.
There is an interesting and possibly relevant point hiding in here.

The calculation described above assumes that the glass surfaces are untreated, and it concludes that the double-bounce is a tiny fraction of the light transmitted straight through. Assuming 5% reflectivity per surface, we get more precisely that the straight-through transmission would be 0.95*0.95, while the double-bounce transmission would be 0.95*0.05*0.05*0.95. The ratio of those values is the 0.25% previously mentioned.

However, if we substitute glass that has been half-silvered on one side and untreated on the other, then the ghost image gets relatively a lot brighter. With one surface of the glass half-silvered, the straight-through transmission becomes 0.95*0.5, while the double-bounce transmission becomes 0.95*0.5*0.05*0.5. The ratio between those values is 2.5% -- an order of magnitude stronger than without the silvering.

When talking about reflector systems like this, people often jump on the idea that for maximum light through the whole system they should have a 50% beamsplitter. That is true, but when there is potential ghosting due to double-bounce, using a half-silvered mirror makes the ghost relatively much more visible.

--Rik
What if the "first surface" (surface facing light or subject) is treated to reduce reflection at first layer, then silver, the 2nd surface is treated to reduce reflection?

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by mjkzz »

However, if we substitute glass that has been half-silvered on one side and untreated on the other, then the ghost image gets relatively a lot brighter. With one surface of the glass half-silvered, the straight-through transmission becomes 0.95*0.5, while the double-bounce transmission becomes 0.95*0.5*0.05*0.5. The ratio between those values is 2.5% -- an order of magnitude stronger than without the silvering.
Indeed interesting, however, one question: I can understand when light travels from air into medium, we can use the reflectivity of the [edit AGAIN] silver (coated on first surface). Once it gets inside, hitting the second surface (the one away from source of light), I understand we should use reflectivity of [edit AGAIN] glass. But when light travels from glass, denser medium, to air, which reflectivity do we use?

Furthermore, in case of total internal reflection, does it matter what material is used outside, say, the fiber glass?

Also, say we have a mirror with thickness of 8mm and with normal everyday glass that has reflectivity of 8%, what happens if we look at something in the mirror at an angle, with high reflectivity of silver (aluminum most likely) and 8%, we might see ghost image if we use reflectivity of silver, but do we?

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by rjlittlefield »

mjkzz wrote:
Sat Jan 22, 2022 4:13 am
Yours appear to be far apart due to shining angle.
The secondary dot in mine is a few mm away from the primary dot, consistent with the thickness of glass. I did not measure it precisely. Perhaps your judgement of scale is set off because I ran the laser through a pinhole so the primary dot is much smaller than a typical laser pointer would produce.

I would say, the ghost is still busted.
You may be surprised, but I agree with you on this point.

The purpose of my analysis and diagrams was to illustrate what causes ghost images, in cases where they are seen.

I am not surprised that your setup does not produce visible ghosts. Axial illumination by shooting through angled glass is a classic technique. I have used and discussed it myself, for example at viewtopic.php?t=15820. And in fact, I wrote in that discussion (10 years ago) that
thin is important mostly to avoid introducing aberrations. There would be no problem if the beam-splitter were in an infinity portion of the optical train, where the image-forming light is organized into pencils of parallel rays. But located in front of the lens like this where the light is focusing in a cone, even glass that's perpendicular to the optical axis will introduce spherical aberration. (Think of cover glass thickness under a high-NA microscope objective.) Glass that's not perpendicular to the optical axis is even worse, since the path length through the glass not only increases but becomes different on opposite sides of the aperture.

To illustrate, here are three images that I shot just now. These are actual pixels from a Canon T1i and Canon 100 mm f/2.8 L IS USM macro lens at very close to 1:1, same as the original stacks were shot. However these were shot at f/2.8 to maximize the effect of the glass. As you can see, the thin glass has only a small effect on image formation, but the thick glass really messes things up.

Left = no glass; middle = 1.15 mm at 45 degrees; right = 3.1 mm at 45 degrees.

Image

...
I didn't worry about reflections of the subject. Those are formed by a double-bounce from the two glass surfaces, so their intensity will be under 1%, too small to matter here.

However, anything on the side opposite the light will reflect in a single bounce, which basically means it's just as important as the subject.

The issue of aberrations may be in play in other people's reports of ghosts. If you're predisposed to think of reflections and not aberrations, it would be simple enough to describe the image degradation shown above as a ghost, rather than a blurring caused by aberration.

At this point I really don't know what is responsible for the reports of ghosts. Perhaps something in this discussion will help people who see them, figure them out.

You asked:
Also, say we have a mirror with thickness of 8mm and with normal everyday glass that has reflectivity of 8%, what happens if we look at something in the mirror at an angle, with high reflectivity of silver (aluminum most likely) and 8%, we might see ghost image if we use reflectivity of silver, but do we?
Of course. The ghost in this situation is well known and easily documented. Here is a snapshot that I took just now, showing a glancing reflection from a second-surface hand mirror.
GhostInSecondSurfaceMirror.jpg

--Rik

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Axial Lights II (Ghostbuster?)

Post by ray_parkhurst »

rjlittlefield wrote:
Sat Jan 22, 2022 12:28 pm
The issue of aberrations may be in play in other people's reports of ghosts. If you're predisposed to think of reflections and not aberrations, it would be simple enough to describe the image degradation shown above as a ghost, rather than a blurring caused by aberration.
Yes, this is the case with my inputs. I was not sure exactly how to describe the effects you show in your thickness sweep, so in @mjkzz's previous "Axial" post, I stated that the cause was "ghosting/CAs/Distortion", when indeed it is probably other aberrations in play.

One thing I have found though is that the aberrations even on 1.1mm glass are substantial, which is why I was so interested in shooting with 0.7mm and even the 0.17/0.2mm cover glasses.

Post Reply Previous topicNext topic