A Macro Beauty-dish Diffuser

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

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Eric F
Posts: 246
Joined: Tue Nov 11, 2008 1:38 pm
Location: Sacramento, Calif.

Post by Eric F »

Craig,

I picked up a Rosco Diffusion Kit awhile back (cost about $30 U.S.; they make you buy from a local retailer...). The kit contains 24 letter size pieces of diffusion material, which I have been playing around with, looking for the best option for my needs. I'm far from exhausting all possibilities (several types contain 1/2 & 1/4 strength pieces -- so the combos are numerous!). So far, my favorite is indeed "Tough Rolux" (which is labelled #3000 in my set).

Using these pieces is really simple, and my favorite method is using a two-piece setup, as follows: (1) make a transparent cone (from clear Mylar is good) with a shape like Andrew & Chris devised for the Beljan RMS adapter cone, with the smaller hole just large enough to either fit tightly on the end of a lens or objective, or to slip loosely around a narrow, basal part of the lens (near the RMS screw end) -- like a skirt. (The latter method requires the cone to be in place before the lens is screwed on.) (2) Make a diffuser cone of the same size (or longer), and slip this over the outside of the plastic cone (diffuser cone always has to be in place before the plastic cone is attached). The plastic serves as a nice, stiff platform for the diffuser "paper" to rest on; a single plastic cone can be used as the base for a variety of different diffuser cones - so it is easy to experiment with various pieces in the Rosco Diffuser Kit. One variable is working distance (between tip of lens and subject), and I have found that a single "short" plastic skirt will nicely hold diffuser cones of different lengths.

There are, of course, many other possible ways to place the diffuser material in position; the main challenge is to decide which material to use!

Holiday Cheers,

Eric

The BAT
Posts: 111
Joined: Wed Sep 02, 2009 5:32 pm
Location: Ballarat, Australia

Post by The BAT »

Eric F wrote:Craig,

I picked up a Rosco Diffusion Kit awhile back (cost about $30 U.S.; they make you buy from a local retailer...). The kit contains 24 letter size pieces of diffusion material, which I have been playing around with, looking for the best option for my needs. I'm far from exhausting all possibilities (several types contain 1/2 & 1/4 strength pieces -- so the combos are numerous!). So far, my favorite is indeed "Tough Rolux" (which is labelled #3000 in my set).

Using these pieces is really simple, and my favorite method is using a two-piece setup, as follows: (1) make a transparent cone (from clear Mylar is good) with a shape like Andrew & Chris devised for the Beljan RMS adapter cone, with the smaller hole just large enough to either fit tightly on the end of a lens or objective, or to slip loosely around a narrow, basal part of the lens (near the RMS screw end) -- like a skirt. (The latter method requires the cone to be in place before the lens is screwed on.) (2) Make a diffuser cone of the same size (or longer), and slip this over the outside of the plastic cone (diffuser cone always has to be in place before the plastic cone is attached). The plastic serves as a nice, stiff platform for the diffuser "paper" to rest on; a single plastic cone can be used as the base for a variety of different diffuser cones - so it is easy to experiment with various pieces in the Rosco Diffuser Kit. One variable is working distance (between tip of lens and subject), and I have found that a single "short" plastic skirt will nicely hold diffuser cones of different lengths.

There are, of course, many other possible ways to place the diffuser material in position; the main challenge is to decide which material to use!

Holiday Cheers,

Eric
Hi Eric,
I'm sorry but I'm completely confused? What is it that you are trying to diffuse? You say that you are fitting the diffuser cones tightly to the lens and then vary the distance of the diffuser material to the subject?
I thought that this thread was about fitting diffusers to flash heads to control lighting of the subject?
Maybe some pictures to help out a 'dumb colonial' like myself. . . .hehehe
Bruce

Eric F
Posts: 246
Joined: Tue Nov 11, 2008 1:38 pm
Location: Sacramento, Calif.

Post by Eric F »

Hi Bruce,

My apologies for the confusion! I'm afraid I inadvertently changed the nature of John's "Macro Beauty-dish" thread. My diffuser is for use on a camera attached to a stand, not for a free-roaming camera used to capture the fantastic field shots that John makes. I was too anxious to respond to Craig's comments about Rosco "Tough Rolux" and stopped thinking clearly (again...happens a lot these days I'm afraid!). Craig is responsible for my getting involved with the Rosco diffuser system--from comments he made about Rosco many months ago. I contacted the company after his initial thread, and have been very pleased with their products, so I wanted to add my two cents. I should have started a new thread to do this (maybe Rik can switch my remarks to a more appropriate thread?)

I can still make some snaps if you are interested. But my small diffuser "skirts" are quite like a half ping pong ball diffuser--but offer more options re. diffuser strength using the various Rosco papers. No way you could use them to photograph live insects I think, as the subject has to be "inside" the base of the diffuser cone for it to work properly.

Eric

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

Post by rjlittlefield »

Eric F wrote:maybe Rik can switch my remarks to a more appropriate thread?
Since things are now clear here, I think I'll leave well enough alone. When a thread gets broken apart, it tends to get even more confusing because of implicit references to other parts of the original thread. Besides, reorganizing is kind of like remodeling -- it's hard to know where to stop. My comment about Kleenex tissue, now a page and a half back in this thread, originated in wrap-around diffusers also.

--Rik

augusthouse
Posts: 1195
Joined: Sat Sep 16, 2006 1:39 am
Location: New South Wales Australia

Post by augusthouse »

Eric,

Thankyou for the info and update regarding your use of the Rosco diffusion series. :D
Your description doesn't leave too much to the imagination; but it would be good to see some of your 'skirts' in a new thread when you get a chance to take some 'snaps'.

Craig
To use a classic quote from 'Antz' - "I almost know exactly what I'm doing!"

The BAT
Posts: 111
Joined: Wed Sep 02, 2009 5:32 pm
Location: Ballarat, Australia

Post by The BAT »

Hi Guys,
Just a quick look at my new diffuser.
It has been made from a 75mm white plastic kitchen funnel and a plastic can sealer of the same diameter. The can sealer was originally purchased to keep large cans of dog food fresh after they had been opened. . . :lol:
Anyway, 5mins with the dremel tool and the funnel fitted neatly onto my SB-400 flash and then a couple of layers of window 'privacy' film took care of the diffusion...
Image
Image
Image
Let me know what you think. . . .
Bruce

morfa
Posts: 554
Joined: Mon Oct 12, 2009 2:14 pm
Location: Stockholm, Sweden
Contact:

Post by morfa »

Hi!

It's been most interesting to read about the different diffuser materials in this thread.

Ironically, one of the points I was trying make in my original post was that the design of the diffuser rather than the diffusion material that is the key factor to getting pleasing results (given my set of requirements – see original post). So, without taking anything away from the great, informative discussion on different diffuser materials – here are some additional thoughts in line with the reasoning in my original post:

This isn't always the case but I in my experience, if you want a relatively small, durable diffuser, that can handle a wide range of working distances (subject – diffuser) for work in the field it's better to focus your efforts on trying to spread the light out using some amount of reflection rather than trying to find the magic material to put in front of the flash head. For studio or table-top work of motionless subjects it's not a big problem to position the diffuser very close to the subject – for field work on the other hand this is much more of an issue. Also, for field work, you don't want the diffuser to be "deep" (long distance between flash head and diffuser surface) since this makes a flexible lighting system more difficult to achieve. For ths reason you want the diffuser to extend as short as possible in front of the flash head. But this means the diffuser must deal with a narrower, more concentrated light beam. Again – this is not a big issue under controlled circumstances!

A couple of observations I've made when experimenting with different field diffusers:

1. Perfectly even diffusion over a small area often gives less satisfying results than less even diffusion spread out over a larger area.
2. At the same (diffuser area) : (subject distance) ratio a less even light pattern can be vastly preferable to the perfectly even one if the brightest parts of the diffuser is spread out towards the edges.
3. There is often a trade-off between light output (transmission) and light "even-ness". I.e. the diffuser with a very evenly spread light pattern demands more power to achieve the same exposure as the diffuser with a less even pattern.

=> An inversely center weighted light pattern can give more pleasing "diffusion" (within citation-marks since technically the light might actually be less diffused this way than with other designs) as well as being more energy efficient.

Another thing I thought I should mention:

It is easy to be fooled into thinking the light output is very even across the diffuser. This, for instance looks pretty good at a quick glance:

Image

However, the truth is completely hidden in the level 255 peak to the right in the histogram.

To be able to judge the light pattern you need to make sure nothing is blown out. Judging from a properly exposed image of the same diffuser it's obvious that this diffuser is utterly lousy.

Image

(This was shot in RAW-mode with my Fujifilm S5pro set to it's maximum "400%" dynamic range. As you can see the diffuser center is almost an entire "dynamic range" brighter than the edges and this means the diffuser doesn't do much good at all)

/John

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

Post by rjlittlefield »

John, thanks for bringing us back on track. Sometimes we need to be reminded of what the important points are!

--Rik

augusthouse
Posts: 1195
Joined: Sat Sep 16, 2006 1:39 am
Location: New South Wales Australia

Post by augusthouse »

Excellent thread!

John, in the earlier version of your elipitical diffuser the 'V' reflector at the center has the 'foil' facing the flash head; whereas, in the second version the 'V' foil/center weight appears to be modified and is facing outward when attached to the transparent positioner. Is this a design modification or is it displayed as such for demonstration?

Also, the testing of the diffuser's efficiency. The 'file info/camera data' indicates that the initial test image was taken at f/11, 1/250 sec. Could you confirm that I am reading these details correctly? Just want to make sure I understand this testing process accurately and the other design factors mentioned.

Thanks for sharing this design and your detailed posts.

Craig
To use a classic quote from 'Antz' - "I almost know exactly what I'm doing!"

morfa
Posts: 554
Joined: Mon Oct 12, 2009 2:14 pm
Location: Stockholm, Sweden
Contact:

Post by morfa »

Craig,

The fold is facing the flash head in both – it's an optical illusion that it seems to be reversed in the newer version.

Here is another one that perhaps is easier to make sense of
Image

As, regards to the shooting data – that seems to be the exif but I'm afraid this data is both wrong and uninteresting:

1) Exif says it's shot with a 45mm/f4.5 but I remember I shot it with a 28mm/f2.8 – I just didn't bother to dial in the correct focal length/f-number into the camera when I snapped the test shots.

2) The exposure is controlled entirely by the flash system, via iTTL and there is no way of knowing what output the flash used. The image would've looked much the same whatever exposure settings I'd used!

If I were to do a proper test of the output efficiency (and that would be interesting!) I would need to set the flash manually. Unfortunately this is not possible with the SB-400, at least not with my camera body.

/John

Planapo
Posts: 1581
Joined: Tue Nov 07, 2006 2:33 am
Location: Germany, in the United States of Europe

Post by Planapo »

Ah yes, from that last picture I understand how the inner part is made and set in. I plan to tinker such a "beauty dish" myself, and I will name it "the blåbär diffuser". :wink:

Thanks John and all the others for this helpful discussion!

--Betty

DQE
Posts: 1653
Joined: Tue Jul 08, 2008 1:33 pm
Location: near Portland, Maine, USA

Some ad-hoc thoughts on flash reflector and diffuser design

Post by DQE »

Would the following design work with similar goals as the DIY flash systems discussed, probably including Lord V's design?

1. Parabola-shaped reflector, with high specular reflectance, ideally mirror-like in its qualities. If diffusing materials are used on the reflector's surface, some of the benefits and efficiency of the parabolic reflector would presumably be lost.

Here's a wikipedia link to parabolic reflectors. This design requires that the flash gun's light source be approximately at the focus of the parabola, as in a searchlight. Unfortunately, flash guns usually are not point sources of light.

http://en.wikipedia.org/wiki/Parabolic_reflector

Perhaps a parabolic trough would work better with a typical rectangular flash gun's shape and size:

http://en.wikipedia.org/wiki/Parabolic_trough

2. high-efficiency diffuser sheet at the exit surface, whose purpose is to convert the light to a diffuse beam (light goes uniformly into a hemisphere) instead of the parallel "spotlight" beam the reflector creates. How diffuse the light beam should be is a matter of taste and preference, hopefully proven by careful side-by-side comparisons with a variety of subjects.

The size of the exit surface of the parabola (a high-efficiency, highly effective diffuser) should be consistent with the illumination goals of one's macro photography and the anticipated distance to the subjects.

Unfortunately(?), a larger reflector may start interfering with one's ability to take photos inside a bush or flower in some instances. I know I have some trouble with my MT-24 twinflash getting caught up in leaves, etc, while I'm looking through the viewfinder.

The main point is that the shape of the reflector should be designed to create an approximately uniformly illuminated diffusion sheet/exit surface. Otherwise, one has to design compensators as Morfa has done, yet such compensators may cause some light absorption. One does not want too many reflections in a reflector since each reflection eats some light, depending on its surface reflectance.

Just my 1.5 cents/Euros and a few ad-hoc thoughts. I hope they at least stimulate some useful thought.
-Phil

"Diffraction never sleeps"

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

If the flashy thing is a long thin cylinder, then it it's a point-source ish, if viewed from the end. So the reflector can be a parabola. The one I broke up yesterday looked parabolic to me!
But the flashgun in the diffuser box above doesn't have the tube exposed and the reflector is already in there, so I think it just has to be developed empirically.

I once worked at a research lab (Thorn EMI) where the lighting division were using maths to work out the best reflector shapes, for things like lighting bays. For tubes (flourescent) they came up with were a sort of short-middle W shaped paraboloid.
I later found myself making "fish fryers" and "swimming pools", which were the names some gave to huge powerful flat lighting sources for studio use. They were more like the box above, (though about 3 metres by 2 or 4,) in that they had discrete sources (like 10 x 1000 Joules of Elinchrom studio flash heads) inside. The heads were pointed backwards at a white rear surface, and the front had layers of drafting film with spaces between. We had to do that at the edge because we didn't want a hard edge to appear in reflections. This was film days of course, so we had to go over the surfaces with flash meters to see how even things were.
With a couple of kilowatts of modelling lights, fans had to be built in too. Plenty of light to focus with though, even at f45, with your loupe under a dark cloth - a bit like a Live View screen!

dmillard
Posts: 636
Joined: Tue Oct 24, 2006 7:37 pm
Location: Austin, Texas

Post by dmillard »

morfa wrote:
A couple of observations I've made when experimenting with different field diffusers:

1. Perfectly even diffusion over a small area often gives less satisfying results than less even diffusion spread out over a larger area.
2. At the same (diffuser area) : (subject distance) ratio a less even light pattern can be vastly preferable to the perfectly even one if the brightest parts of the diffuser is spread out towards the edges.
3. There is often a trade-off between light output (transmission) and light "even-ness". I.e. the diffuser with a very evenly spread light pattern demands more power to achieve the same exposure as the diffuser with a less even pattern.

=> An inversely center weighted light pattern can give more pleasing "diffusion" (within citation-marks since technically the light might actually be less diffused this way than with other designs) as well as being more energy efficient.

/John
John -

I really like your ice cream diffuser (plus it gives me one more excuse to eat ice cream :) ). Just an idea - could you incorporate fiber optic filaments in your diffusion head , e.g. these currently on eBay, to further eliminate the central hot spot by selectively piping more light to the edges of the lid?

Regards,
David

Eric F
Posts: 246
Joined: Tue Nov 11, 2008 1:38 pm
Location: Sacramento, Calif.

Post by Eric F »

Just to note: I have added a discussion and some photos on using "Rosco Diffuser Kit" papers to the "How can I diffuse illumination" thread in the Tecniques FAQs forum.

Eric

Post Reply Previous topicNext topic