Chondrules and inclusions in a 4.6 billion yr old meteorite

[rjlittlefield]

Bruce,

About the lighting, I'm not sure this will help but I'll give it a try. Sorry about the length -- I couldn't figure out how to make it any shorter. Grab a snack and read carefully, I guess.

---------

It helps to imagine yourself to be shrunk tiny, standing on the subject.

From that viewpoint, what does the illumination look like?

If most of the light comes from a narrow angle -- a light source that is apparently small but bright -- then the illumination is "harsh", or "directional". It will cast dark shadows with hard edges. If the camera sees that illumination reflecting from a shiny surface, then the reflection will be small and bright. Direct sunlight on a clear day is harsh/directional.

On the other hand, if the light comes from a wide angle -- a light source that is apparently large but dim -- then the illumination becomes "flat", or "diffuse". It will cast lighter shadows with soft edges. If the camera sees that illumination reflecting from a shiny surface, the reflection will be large and not so bright. Outdoor light on a foggy day is flat/diffused.

Most light sources are intrinsically small and therefore directional. To make diffuse illumination, you either bounce the light off a big reflector, or shine it through a big diffuser. In either case, whatever portion of the reflector or diffuser is lighted up, becomes another light source from the standpoint of the subject. Typically the reflector or diffuser covers a wide angle (from the standpoint of the subject), so that light is a lot more diffuse than the original source.

The light coming from your fiber optic heads will be very directional unless the heads are very close to the subject. To get a feel for this, hold a toothpick near the subject and see how sharp and intense the toothpick's shadows are. (There will be two shadows, one for each light source.)

Most good illumination is a combination of directional and diffuse components. The trick is to balance the relative amounts, and to adjust the position and size of the directional sources.

In terms of image quality, the absolute light level does not matter, except for low-order effects like noise accumulation for long exposures.

However, most rooms provide a fair amount of diffuse illumination just from light bouncing off the walls. When you turn down the brightness of any added directional sources, you increase the proportion of diffuse illumination. Many subjects, especially shiny ones, look better with diffuse illumination. I'm pretty sure this is why Ken's recommendation to reduce the brightness works for him.

Another way to get more diffuse illumination is to explicitly add some. Ken's other suggestion, to use fluorescents, makes sense here in terms of using a larger and therefore more diffuse light source. But there are other ways involving diffusers and reflectors.

White pingpong balls are made of very good diffusing material, meaning that however much light goes in, most of it comes out again, at pretty much all angles.

So, you can add some nice diffuse illumination -- from the standpoint of a small subject -- by shining light onto a big chunk of pingpong ball material that is very close to the subject. You can also create somewhat directional illumination in a wide range of sizes and positions by simply illuminating smaller portions of the pingpong ball.

There's nothing magic about a pingpong ball except that it's small, white, diffusing, and quite durable. (Eggshells were the classic tool for macro lighting, but they're not so great for "durable".)

Sometimes pingpong balls are too small, in which case other diffusers like paper, white fabric, or white-painted plastic can be used in just the same way.

In general, with 3D subjects you need some diffuse illumination to keep shadows from going black, and you need some directional illumination to let the user infer shape from shading. If the subject has shiny parts, then it also helps to have some very small (highly directional) source positioned so that the camera can pick up its reflections as "catchlights". Without some bright spots that are recognizable as reflections, the viewer can't tell whether a surface is matte or glossy.

As you say, there are a lot of variables to play with. If I were feeling mathematical, I might say there's at least 13 of them. That's x, y, and z plus two tilts, plus lens extension, for each of your illuminator's two heads, plus illuminator vs room brightness if there's enough ambient light to matter.

Surrounding the subject with a pingpong ball cuts the count a bit. Let's see...each head will illuminate a spot somewhere on the pingpong ball, and you can characterize the spot by latitude, longitude, radius, and brightness, but the ball will exclude ambient light, and, uh, only the relative brightness of the two spots matters, so, um...gee, we're down to only 7 variables!

Seven variables is a lot less than 13, but it's still quite a few, and it's probably too many to think that you're ever going to get the "perfect" lighting.

I'm afraid that's about all the science. Everything else is art -- a field I've never been particularly good at.

I suggest attacking stepwise -- think in terms of one main light, fairly directional, plus one fill light, fairly diffuse. Using the pingpong ball, that means placing one head very close to illuminate a small area of the ball, and pulling the second head back to illuminate a much larger area. But some symmetric subjects look great with symmetric lighting. Again, you can often get to a good combination by moving one light until the illumination looks pretty good, then moving the other until it looks even better.

With luck, what I've written above will be helpful to understand what's going on. But there's no substitute for flying the lights around to see what "looks good".

Hope this helps,
--Rik

[Ken Ramos]

Rik said:

Grab a snack and read carefully, I guess.

Heck, I sent out for a pizza! Pretty good novel there Rik with lost of good information.

[Bruce Williams]

Really appreciate the time you've given to helping out on this one Rik. In many ways lighting is as important as the microscope and the camera to achieving a good result - so I am grateful for your advice especially when it's as clearly and interestingly expressed as your posting.

Will be buying some ping-pong balls asap.

Bruce

[rjlittlefield]

Bruce, you're very welcome. I find that writing about stuff helps me organize my own understanding, so with a bit of luck and time to do it, it's a win-win situation. Good questions are precious things!

All feedback is appreciated, by the way. If suggestions don't work, or ideas are hard to understand, or just plain don't make sense, be sure to let me know, OK? Sometimes I write nonsense, and once in a while I don't notice until it's pointed out.

--Rik