How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

mkbn
Posts: 61
Joined: Thu Jan 25, 2024 10:58 pm

How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by mkbn »

While I'm going crazy on my own with little tweaks and trying to compare results frame to frame, I thought I'd ask if anyone has thoughts on shapes you would use to:

1. Diffuse the flashes (shape and distance from diffuser to subjects)

2. Reflect light for efficiency to hit the subject

3. Trap stray light and keep it out of the front lens element?

What I've got so far is 3 parts: an inner light shielding cone (testin various shapes, miht add reflective material outside of tip of cone) to cut out on glare, an outer diffuser, and a new clip on reflector. Also considering some on-flash reflectors to bring more of the wasted light towards the reflector, and a behind-the-objective flat reflective surface or cup to bring some aiming forward.
PXL_20240210_065109291.MP_copy_800x600.jpg
PXL_20240210_071731031.MP_copy_800x600.jpg
Screenshot_20240210-091822_copy_768x845_copy_307x338.png
Before I go too crazy, anyone got any obvious tips on reshaping my diffusers, or reflectors, or what would you do if you could have material wherever you wanted it to move light around?

I'm happy to keep playing, but would love to hear things that seem off base or obvious improvements I may not have the knowledge for :)

Babylonia
Posts: 91
Joined: Tue Mar 20, 2012 10:56 am

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by Babylonia »

It's always a matter of fiddling around with light shapers, with a high DIY content.
Yet there are many useful attributes in the market to be found at Amazon and AliExpress, for low prices.
With some fantasy it can be optimized for your own needs.

To get light of flashes just more diffuse and gentle e.g.:
https://www.amazon.com/s?k=Flash+Diffuser+Dome+Bounce (and follow up pages).

Also there are Velcro band hinged "transparant" barn-doors. (Falcon Eyes - including coloured gels & reflective / black piece).
Simply place a piece of black paper cut to size within the transparent holder, and you have a small black barn-door to keep light away.
Can also attached around a lens barrel to keep light away.

(Colour of Velcro band normally is black).
Barn-door.jpg
-

From a handicraft shop, pieces of foam board, to stick adhesive silver foil on it, and cut in smaller cards (A5 - A6 & smaller).
"Kneaded eraser" = putty rubber - not for erase things, but for sticking small props or a small reflective cards around a set-up.
(To stick to a piece of metal "block" and hold as a "stand" for a small reflective foam card).
But as an alternative or "extra" attributes as for putty rubber, you can also make use of - handy magnetic "Third hands"
(As you are already using): https://www.aliexpress.us/item/3256804755320812.html

Also be available within a handicraft shop (or Photo attribute shop):
White opaque polyester foil / translucent milk white polyester film. (Also as sheets to be packed as a notebook).
For all kinds of softening the light coming from behind.

Ikea LED light (black or white coloured fixture), when using continuous light.
https://www.ikea.com/us/en/p/naevlinge- ... -30449885/

When using this kind of "straight" small continuous IKEA light, just very close to the translucent milky cup around the lens.
You can modelling the light extensively in gentle "light / dark" areas.

Hope it gives a little bit of inspiration.
-
Last edited by Babylonia on Sat Feb 10, 2024 12:00 pm, edited 1 time in total.
Greetings from Holland

Beatsy
Posts: 2095
Joined: Fri Jul 05, 2013 3:10 am
Location: Malvern, UK

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by Beatsy »

You are travelling a familiar road but it looks like you have a good handle on the important issues.

Like you, I tinker endlessly but along the way I found a fairly general purpose solution that I still use. I described it in post that touches on a few of the points/issues you raise. Assuming you haven't read it already, it's here...

viewtopic.php?t=43490

One tip: when trying various lighting and diffusion setups, it helps to use a small ball bearing as a test subject. The reflections off the ball surface will show you precisely where the light is coming from, and how it's diffused. This very clearly shows hot spots and areas/directions that provide no illumination, or too much. It's often surprising how little of the "sky" (from the subject's PoV) is actually lit. More sky lit = more diffusion.

Cheers
Beats

mkbn
Posts: 61
Joined: Thu Jan 25, 2024 10:58 pm

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by mkbn »

Babylonia wrote:
Sat Feb 10, 2024 11:34 am
It's always a matter of fiddling around with light shapers, with a high DIY content.
Yet there are many useful attributes in the market to be found at Amazon and AliExpress, for low prices.
With some fantasy it can be optimized for your own needs.

To get light of flashes just more diffuse and gentle e.g.:
https://www.amazon.com/s?k=Flash+Diffuser+Dome+Bounce (and follow up pages).

Also there are Velcro band hinged "transparant" barn-doors. (Falcon Eyes - including coloured gels & reflective / black piece).
Simply place a piece of black paper cut to size within the transparent holder, and you have a small black barn-door to keep light away.
Can also attached around a lens barrel to keep light away.

(Colour of Velcro band normally is black).
Barn-door.jpg
-

From a handicraft shop, pieces of foam board, to stick adhesive silver foil on it, and cut in smaller cards (A5 - A6 & smaller).
"Kneaded eraser" = putty rubber - not for erase things, but for sticking small props or a small reflective cards around a set-up.
(To stick to a piece of metal "block" and hold as a "stand" for a small reflective foam card).
But as an alternative or "extra" attributes as for putty rubber, you can also make use of - handy magnetic "Third hands"
(As you are already using): https://www.aliexpress.us/item/3256804755320812.html

Also be available within a handicraft shop (or Photo attribute shop):
White opaque polyester foil / translucent milk white polyester film. (Also as sheets to be packed as a notebook).
For all kinds of softening the light coming from behind.

Ikea LED light (black or white coloured fixture), when using continuous light.
https://www.ikea.com/us/en/p/naevlinge- ... -30449885/

When using this kind of "straight" small continuous IKEA light, just very close to the translucent milky cup around the lens.
You can modelling the light extensively in gentle "light / dark" areas.

Hope it gives a little bit of inspiration.
-
Thanks a bunch for the inspiration and rundown, that's very helpful!

I'm starting off with a little mini positionable reflector set for the speed lights, so I can aim things and not esste light. Other ideas are percolating around right now :)
Screenshot_20240212-122326_copy_170x229.png
Beatsy wrote:
Sat Feb 10, 2024 11:52 am
You are travelling a familiar road but it looks like you have a good handle on the important issues.

Like you, I tinker endlessly but along the way I found a fairly general purpose solution that I still use. I described it in post that touches on a few of the points/issues you raise. Assuming you haven't read it already, it's here...

viewtopic.php?t=43490

One tip: when trying various lighting and diffusion setups, it helps to use a small ball bearing as a test subject. The reflections off the ball surface will show you precisely where the light is coming from, and how it's diffused. This very clearly shows hot spots and areas/directions that provide no illumination, or too much. It's often surprising how little of the "sky" (from the subject's PoV) is actually lit. More sky lit = more diffusion.

Cheers
Beats
Thanks!!!! Yep, my brain is always tinkering even when offline from a project, so these things are fun.

I hadn't seen yours explicitly (though I checked out the setup thread), so thanks for linking me!


One thing I hadn't figured out was the diffuser size and distance to subject, but I'm tinkering around with different sizes.

rjlittlefield
Site Admin
Posts: 23508
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by rjlittlefield »

mkbn wrote:
Mon Feb 12, 2024 12:29 pm
I hadn't seen yours explicitly (though I checked out the setup thread), so thanks for linking me!
If "setup thread" means viewtopic.php?p=55311#p55311 , then yep, it's in there, currently 8th from the bottom. But it's a lonnggg list -- easy to miss stuff.

--Rik

Espresso
Posts: 11
Joined: Sun Feb 04, 2024 6:18 am
Location: Sweden

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by Espresso »

In a totally different field of physics I came across two different philosophies
regarding the size of emitter/receiver and distance to the object that was in the
"center". Far away the sources must be large not to be a point source but then it
is hard to move them and takes up a lot of space. Close up you use less power but you
can easily move them around. You can use halogen or LED and continous light if
most of it hits the target and is not illuminating the entire room. All depends
on the size of the object but a couple of optical fibers can give enough light
for a square millimeter.

Is totally diffused light (like a cloudy day outdoors) the "norm" or is directional
variation wanted?

(the "different field" was microwave antenna measurements - nearfield or farfield)

Babylonia
Posts: 91
Joined: Tue Mar 20, 2012 10:56 am

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by Babylonia »

mkbn wrote:
Mon Feb 12, 2024 12:29 pm
Babylonia wrote:
Sat Feb 10, 2024 11:34 am
Also there are Velcro band hinged "transparant" barn-doors. (Falcon Eyes - including coloured gels & reflective / black piece).
Simply place a piece of black paper cut to size within the transparent holder,
and you have a small black barn-door to keep light away. Can also attached around a lens barrel to keep light away.
I'm starting off with a little mini positionable reflector set for the speed lights, so I can aim things and not esste light.
By the fast 3D software design of such a self made part, you made of it already,
it seems you do have a great hobby to make such parts yourself. ;-)

Image

​Of course it is great that you can manage to make such parts yourself. Don't want to stop people from doing that.

For all other users who do not have a 3-D printer, for the low costs of these parts that you can buy straight away.
Think about all energy, time and perhaps also costs you can save by simply buying it.
https://www.aliexpress.us/item/2251832682115571.html

-
Greetings from Holland

mkbn
Posts: 61
Joined: Thu Jan 25, 2024 10:58 pm

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by mkbn »

rjlittlefield wrote:
Mon Feb 12, 2024 12:40 pm
mkbn wrote:
Mon Feb 12, 2024 12:29 pm
I hadn't seen yours explicitly (though I checked out the setup thread), so thanks for linking me!
If "setup thread" means viewtopic.php?p=55311#p55311 , then yep, it's in there, currently 8th from the bottom. But it's a lonnggg list -- easy to miss stuff.

--Rik
Yep! That's the thread I meant. And I certain haven't gone through each one yet, I've just been looking at posts in chunks and letting the reasons for various things sit in my mind for a bit.
Espresso wrote:
Mon Feb 12, 2024 1:48 pm
In a totally different field of physics I came across two different philosophies
regarding the size of emitter/receiver and distance to the object that was in the
"center". Far away the sources must be large not to be a point source but then it
is hard to move them and takes up a lot of space. Close up you use less power but you
can easily move them around. You can use halogen or LED and continous light if
most of it hits the target and is not illuminating the entire room. All depends
on the size of the object but a couple of optical fibers can give enough light
for a square millimeter.

Is totally diffused light (like a cloudy day outdoors) the "norm" or is directional
variation wanted?

(the "different field" was microwave antenna measurements - nearfield or farfield)

Ah, likely totally diffused vs directional light will be the most helpful, but I could see myself wanting to have some directionality for certain compositions as I learn.

I'm using flash speed lights for the moment, which I all of a sudden have 4 of, but I'd like to try some continuous lighting at some point. I have random materials (granite slab, metal, al extrusion, etc) to explore making things solid for continuous, but I don't want to get too deep into anything at the expense of the rapid learning right now, so I'm sticking to speed lights.

One thing I was curious about and that I've seen discussions of is a fiber optic modifier for one speed light. I know collimation would be best, but I also saw this curious reddit 3d printed adapter that the person used a cheap toslink cable cut up for:

https://www.reddit.com/r/3Dprinting/com ... ter_macro/

Since the 3d printing and design stuff is no problem, I was almost curious if I used my tiny tt350 to inefficiently funnel light right up to the objective in channels, if i'd gain some fill light just due to things being so close.

Some of my subjects have proved very difficult to illuminate in a diffuse manner as they're reflective spheres, but also dark, but also surrounded by translucent elements. I chatted about this briefly in another thread and Rik linked a helpful discussion on illuminating knife shavings.

Examples, heavily sharpened (I was playing with how much is too much):
cexsharpr2024-02-1116-47-29(B,R8,S1)_copy_768x817.jpg
exsharpo2024-02-1116-52-46(B,R8,S1)_copy_600x604_copy_768x773.jpg
2024-02-1123-22-51(B,R8,S1)_copy_600x272.jpg
There are also subjects with far, far more dark spores on them packed to the absolute brim, and those have been a nightmare so far. Learning, though! :)

And what I'm currently playing with for setup:



(Still playing with other diffusion material, as well as this one in different thicknesses

and configs)
PXL_20240212_195900048.MP_copy_800x600.jpg
Babylonia wrote:
Mon Feb 12, 2024 2:28 pm
mkbn wrote:
Mon Feb 12, 2024 12:29 pm
Babylonia wrote:
Sat Feb 10, 2024 11:34 am
Also there are Velcro band hinged "transparant" barn-doors. (Falcon Eyes - including coloured gels & reflective / black piece).
Simply place a piece of black paper cut to size within the transparent holder,
and you have a small black barn-door to keep light away. Can also attached around a lens barrel to keep light away.
I'm starting off with a little mini positionable reflector set for the speed lights, so I can aim things and not esste light.
By the fast 3D software design of such a self made part, you made of it already,
it seems you do have a great hobby to make such parts yourself. ;-)

Image

​Of course it is great that you can manage to make such parts yourself. Don't want to stop people from doing that.

For all other users who do not have a 3-D printer, for the low costs of these parts that you can buy straight away.
Think about all energy, time and perhaps also costs you can save by simply buying it.
https://www.aliexpress.us/item/2251832682115571.html

-

Ah yeah for sure!! I love AliExpress quite a bit. I'm thankful to easily be able to make things in the browser and print them quickly (the panel pictured is about 20 minutes of print time), as I'm not quiiiite sure what I want yet and also because time is greater than money in supply for me, often, especially if learning happens. Sometimes after I figure out what would be ideal with prints, i buy something, sometimes the solution I whip up is more useful to me in some way.

rjlittlefield
Site Admin
Posts: 23508
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by rjlittlefield »

Espresso wrote:
Mon Feb 12, 2024 1:48 pm
Is totally diffused light (like a cloudy day outdoors) the "norm" or is directional variation wanted?

(the "different field" was microwave antenna measurements - nearfield or farfield
Some directional variation is helpful to bring out the 3D structure of a subject.

But if the illumination becomes too directional, then you get into a regime sometimes called "partially coherent", in which an assortment of wave interference effects become evident. Here at photomacrography.net, we also speak of "utilized aperture" effects in which light enters only a portion of the entrance pupil, resulting in unexpected effects on DOF, perspective, and diffraction. One of those effects, particularly evident at higher magnifications, is when slightly out-of-focus details of the subject appear to move laterally as focus is changed. The direction of apparent movement depends on local surface orientation, so nearby features can move in different directions, resulting in what I often call "squirming around". For some further discussion and illustration of these effects, see viewtopic.php?p=58613#58613 , viewtopic.php?t=19582 , and viewtopic.php?p=149187#p149187 .

As a general rule, the angular spread of the incoming illumination should be somewhat larger than the angular span of the imaging lens's entrance cone. With a 10X NA 0.25 objective, that means a light source that spans at least 29 degrees wide (=2*degrees(asin(0.25)). The safe approach is to start wide and then narrow down if desired. It's also helpful if the illumination pattern is "apodized" so it does not have hard edges, and this argues for a wider maximum span into which the illumination bright areas can gradually fade.

--Rik

mkbn
Posts: 61
Joined: Thu Jan 25, 2024 10:58 pm

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by mkbn »

rjlittlefield wrote:
Mon Feb 12, 2024 3:36 pm
Espresso wrote:
Mon Feb 12, 2024 1:48 pm
Is totally diffused light (like a cloudy day outdoors) the "norm" or is directional variation wanted?

(the "different field" was microwave antenna measurements - nearfield or farfield
Some directional variation is helpful to bring out the 3D structure of a subject.

But if the illumination becomes too directional, then you get into a regime sometimes called "partially coherent", in which an assortment of wave interference effects become evident. Here at photomacrography.net, we also speak of "utilized aperture" effects in which light enters only a portion of the entrance pupil, resulting in unexpected effects on DOF, perspective, and diffraction. One of those effects, particularly evident at higher magnifications, is when slightly out-of-focus details of the subject appear to move laterally as focus is changed. The direction of apparent movement depends on local surface orientation, so nearby features can move in different directions, resulting in what I often call "squirming around". For some further discussion and illustration of these effects, see viewtopic.php?p=58613#58613 , viewtopic.php?t=19582 , and viewtopic.php?p=149187#p149187 .

As a general rule, the angular spread of the incoming illumination should be somewhat larger than the angular span of the imaging lens's entrance cone. With a 10X NA 0.25 objective, that means a light source that spans at least 29 degrees wide (=2*degrees(asin(0.25)). The safe approach is to start wide and then narrow down if desired. It's also helpful if the illumination pattern is "apodized" so it does not have hard edges, and this argues for a wider maximum span into which the illumination bright areas can gradually fade.

--Rik
I notice the squirming all the time when using a "modeling" flaslight (lol) which doesn't contribute to exposure at all @ iso 100 when focusing in and out, fascinating to think about it being partially related to strong light and the 'utlilized' aperature. Thread #2 is very interesting, though I have to click through and read more than the first page soooon. I don't totally follow your last sentence (in arguing for a wider maximum span), so I'm off to google some terms and read more :)

Espresso
Posts: 11
Joined: Sun Feb 04, 2024 6:18 am
Location: Sweden

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by Espresso »

To see if memories of a "cloudy day illuminator" that I saw as a demo of lighting
for mashine vision I found that my braincell was correct. (Must give it beer tonight
as a reward.)
https://www.cognex.com/what-is/machine- ... s/lighting
Maybe you can get some ideas for solutions from all the examples shown.

One common light source is LEDs used in many products but not available in small numbers
to us that want to experiment. If you like me does autopsies on most dead equipment
very useful LEDs can be found in the background illumination of laptop display panels.
Good CRI and high efficiency that the panel manufacturer has selected and paid for.
Laptop-LED.jpg
This is an old LED contraption used for macro-photography and as ring-light for a
binocular microscope I use when soldering tiny components (tiny = one or two millimeters).
Changing the direction of the light is nice when the shadows give 3D information.
The object is a deodorant bearing, if cut in half makes a diffusing dome.
(some call steel balls a "ball bearing" and so this is a "deodorant bearing")
Macro-LED.jpg
This can easily be scaled down by using SMD LEDs closer to the object.

Moving the light around by flipping switches.
LED-1.jpg
LED-2.jpg
LED-3.jpg

rjlittlefield
Site Admin
Posts: 23508
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by rjlittlefield »

Espresso, thank you for the images. They raise another issue that often we forget to talk about.

Earlier I mentioned the phrase "partially coherent", and noted that "As a general rule, the angular spread of the incoming illumination should be somewhat larger than the angular span of the imaging lens's entrance cone."

Expanding on that issue, I'll provide this snippet from a much earlier discussion
Asha wrote:
Thu Apr 11, 2013 8:47 pm
From Goodman's "Introduction to Fourier Optics", 3rd ed, p.135, through a relationship of angular size of the source, entrance pupil and angular sprectrum of the subject, an incoherent optical system can behave mostly like a coherent system.
The key point here is that the nature of the subject plays an important role also. If your subject has a highly diffusing surface, then you can get away with highly directional illumination, but more mirror-like subjects require more diffuse illumination to avoid problems. In other words, you need a good amount of diffusion somewhere in the system, either from the illumination, or the subject, or a combination of both. Small subjects generally act more like mirrors, and in addition they get photographed with wider entrance cones, both factors pushing for more diffuse illumination.

In the examples provided by Espresso, the subject is fairly large and highly diffusing; hard directional lighting works great in this situation. The same illumination applied to mkbn's mushroom spores would not show the shapes well.

--Rik

mkbn
Posts: 61
Joined: Thu Jan 25, 2024 10:58 pm

Re: How would you design ideal 3d printed light modifiers (reflection, diffusion, and occlusion?) Ideas? Improvements?

Post by mkbn »

Yeah, thanks indeed, because the photos are tremendously cool to see laid out next to each other. At first, I actually was interpreting it as looking through a lens tube and a demonstration of accidentally using only part of an aperature due to strongly directional illumination. Then I looked more closely.

For my lower mag (<5x) macro photos, some inherent unevenness in my diy diffuser and reflector is usually enough to give a 3d feel, rather than perfectly flat light field. Or if using multiple speed lights, changing power levels for one direction.

For the weird, shiny spore blobs which are not only dark for certain species, but also highly reflective and surrounded by what feels like little translucent mirrors, I'm still learning and just looking for progressive improvement, even if I can't tackle the hardest aspects immediately.

espresso, have you considered emailing your directional and lighting control to www hackaday.com 's tip line? I don't do much SMD soldering at all, but I'd guess that a some people would enjoy the spark of an idea for a quality of life improvement while soldering surface mount components.

On an unrelated note, I quickly printed this monstrosity while I redesign the sub optimal 210mm objective tube I modified. It was quite expected that a 210mm lever at the end of a thinner tube would bounce around a lot, but this quick hack made it so the bounce was not the dominant factor in the motion system, allowing for test stacks. Yes, I imagine proper bellows would be useful, but I'm enjoying playing around and I have so much other stuff lacking in technical understanding that stability isn't the only/limiting factor for me at the moment as I check off boxes of learning.

PXL_20240213_235208393.MP_copy_800x600.jpg

Post Reply Previous topicNext topic