Wraparound diffuser size

johan · Post by **johan** » Fri Jan 13, 2012 3:56 am

Hello,

I'd like to put out a general question about wraparound diffuser sizes and whether larger is better. I'm in the somewhat unfortunate position of having all my stuff on one corner of my desk, which is in what used to be known as the study but nowadays is more a children's playroom

- and I only have the smallest of spaces to work in. As in 50cmx50cmx50cm... So larger diffusers are prettymuch out the question.

Larger is of course a relative term and what I mean is a kitchen bowl size diffuser as opposed to the toothpaste tube size that I have to use. Is there much advantage? I notice some eminent mebers here going for the larger size in the various setups that I've seen. The question I suppose really boils down to whether the larger ones mean more bounced diffused as opposed to shine through light.

Thank you.

canonian · Post by **canonian** » Fri Jan 13, 2012 6:12 am

You could try pingpong balls, yoghurt bottles or The DIY chewing gum diffuser .
I use them on microscope objectives and gives off a nice omnidirectional light.
I sometimes use a big diffuser, but it needs plenty of light, thus longer exposures, thus more room for vibration.
There are also a lot of nice idea's and results shown here, taking up not too much space.

EDIT: woops, just looked at your rig and the growing collection of diffuser things perched on the printer! Nothing new I mentioned here, still I hope it's useful info....

ChrisR · Post by **ChrisR** » Fri Jan 13, 2012 11:13 am

As far as I've found, it's the angle of reflector seen by the subject, which matters. So if your subject is a pinhead you can simulate the sky with a tiny dome.
Also too as well, how evenly are you lighting the diffuser... you might have a mirror "behind" the subject to bounce light back on to the diffuser’s rear side (as in ping pong ball) so you need to think of where your flash (EG) is, if it's the sole light source. Closer in, and the directly lit side of the ball sees a wider light source, but the light bouncing off the mirror has to go significantly further (relatively) so it's dimmer than on the front. Inverse square law applies.
And itt depends how well the diffuser actually diffuses - or spreads light within its thickness. Thick things like foam cups are better than pp balls - but they lose more light...

rjlittlefield · Post by **rjlittlefield** » Fri Jan 13, 2012 12:05 pm

As ChrisR says, it's the angles that matter. Small has the advantage that it's simpler to get bright light onto the subject, which reduces exposure times. Large has the advantage that you have more room to manipulate the subject and/or the lights. But if the angles are the same, so is the result.

I have two commonly used diffusers. One of them is a piece of Kleenex tissue just big enough to wrap conveniently around the subject; the other is a roughly 4"x5" cone that goes over the head of my flash. I still use pingpong balls, but not as commonly as I used to because I find them mechanically less convenient.

--Rik

Charles Krebs · Post by **Charles Krebs** » Fri Jan 13, 2012 12:24 pm

One thing to keep in mind is the use of a small diffuser relative to a large subject. (In other words the diffuser makes a very "tight fit" around a large subject). If you have a large subject with a diffuser that is very close to the subject you can experience light fall-off (inverse square) depending greatly on how the surface of the diffuser is illuminated. For example if you put a light on the part of the diffuser that is very close to the subject so that if creates an intense spot, you will clearly notice the light fall-off on the subject as you move away from the parts that are very close to the diffuser. This can work for you or against you, depending on the lighting effect you wish to establish.

johan · Post by **johan** » Fri Jan 13, 2012 1:02 pm

Charles, thank you, this was exactly what made me ponder this and makes me suspect bigger is better (if evenness is your thing). In my head I had it like this, if you take two imaginary cylindrical diffusers to shoot a 3mm bug, the first being 1cm radius and the second being 10cm radius, with the inverse law being what it is, light falloff over say 1mm on the 1cm radius is far more than 1mm on a 10cm radius. As in 1mm is a massive 10% of 1cm but only 1% of 10cm. and since we're actually talking the square of this it's even more pronounced. Hope that makes some sense...

Charles Krebs · Post by **Charles Krebs** » Fri Jan 13, 2012 2:55 pm

Charles, thank you, this was exactly what made me ponder this and makes me suspect bigger is better (if evenness is your thing)

Yes, if "evenness is your thing" you want to be wary of too small a diffusion dome, or a diffusion "tent" that is very close to one part of the subject. But no need to get too carried away with diffuser size. In your limited workspace you should still have no problem with even larger subjects.

if you take two imaginary cylindrical diffusers to shoot a 3mm bug, the first being 1cm radius and the second being 10cm radius, with the inverse law being what it is, light falloff over say 1mm on the 1cm radius is far more than 1mm on a 10cm radius. As in 1mm is a massive 10% of 1cm but only 1% of 10cm. and since we're actually talking the square of this it's even more pronounced.

Your numbers are not correct, but you do have the concept, which is what really matters.

rjlittlefield · Post by **rjlittlefield** » Fri Jan 13, 2012 4:01 pm

In my head I had it like this, if you take two imaginary cylindrical diffusers to shoot a 3mm bug, the first being 1cm radius and the second being 10cm radius, with the inverse law being what it is, light falloff over say 1mm on the 1cm radius is far more than 1mm on a 10cm radius. As in 1mm is a massive 10% of 1cm but only 1% of 10cm. and since we're actually talking the square of this it's even more pronounced. Hope that makes some sense...

Hhmm... This stuff is a bit tricky to discuss. What you've said does "make sense", but it also overlooks some important aspects, and the difference can drag you off in unproductive directions. Let me explain...

It's critical to recognize that the inverse square law directly applies to point sources, but not to extended sources like diffusers. To calculate what happens with those, you need to integrate the inverse square law across the whole diffuser, and when you do that, some unexpected behavior turns up.

It turns out that the light intensity inside a uniformly lit sphere is the same everywhere -- not brighter near the surface as you might think based on the inverse square law. The same is true for a uniformly lit infinite plane and a uniformly lit infinite cylinder. With a finite plane or cylinder, or less than a full sphere, the illumination is not completely uniform, but it's still a lot more uniform than you'd think based on inverse square. So, if evenness is your thing, the important aspect is to light the cylinder itself evenly. Diameter does have an effect, but it's far less important than simple "inverse square" would suggest.

I'm struggling a bit, trying to figure out what's the most productive advice to give.

Charlie's comments are of course dead on. If the illumination is not even, then closer is brighter, approaching inverse square if the illumination is small. That can be a powerful friend or enemy.

On the other hand, there are a lot of other sources for uneven illumination, and in my experience most of them have to do with angles. With your wasps and mineral specimens, my guess is that almost all of the unevenness has to do with the angles that the light is hitting the surface, as opposed to the surface's distance from the diffuser.

So, in terms of advice for what I think you're interested in (getting some specific effect with a 3D subject), I'm still going with angles as the most important thing to consider. But I could be wrong about that, and I'd be interested to see more examples of what you're getting and what you don't like about them.

--Rik

Charles Krebs · Post by **Charles Krebs** » Fri Jan 13, 2012 5:48 pm

Rik said:

Hhmm... This stuff is a bit tricky to discuss

Agreed! Because there are so many variables... the lighting effect you are after, the surface reflectance of your subject, the size and shape of the diffuser, and the positioning and size of the light source(s).

"Angles" play a huge part. But I know for sure that I have had trouble with fall-off in some shots (and have used it to good effect in others). So when you light a subject it's probably a good idea to keep both in mind.

As for the "fall-off" aspect, consider the following illustration (based in part on your hypothetical 3mm subject and 10mm round diffuser). The left panel would most closely follow the "inverse square" relationship. If the distances were as shown, you would experience near a 1.8 stop difference between the head and rear end of the subject. With a partial circular (or dome) diffuser the "inverse square" relationship can't be simply used. Certainly you would not have the same 1.8 stop difference (head to tail) but you would likely still see some noticeable fall-off. The third panel would obviously provide even more even illumination, but this can be varied considerably by the placement and size of the lights used to illuminate the diffuser, as well as its shape... cylindrical or dome (spherical).