Diffusing a Macro Twin Flash

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

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Diffusing a Macro Twin Flash

Post by Dalantech »

Re-posting this one from PhotoMacroGraphy.

I've probably spent the better part of a decade trying to wrap my head around flash photography and how to diffuse a harsh point source like a twin flash in as short of a space as possible. For field macro you just don't have a lot of room to work, so any diffusion scheme has to be compact. I also like to use a twin flash as two separate light sources, in a key (one head at the top of the lens) and fill (one head off to the side) configuration because it gives me a lot of control over the highlights and shadows. It also allow me to partially wrap light around the subject so that it doesn't look flat. If you place the flash heads on opposite sides of the subject, or fire them through a single diffuser that's connected to the end of the lens, the light will be too even across the subject and will potentially make the subject look flat. I don't focus stack, but for those of you who do flat light is a composition buzz kill. Get everything in focus and evenly lit and your images will look 2D. So here's what I've learned while loosing my mind trying to diffuse a twin flash:

1) The diffusers that Sto-Fen sells, and the set that Cannon supplies with the MT26EX RT, are better at blocking the light than they are at forcing it to spread out. You'll lose about a stop of light with either of them for a very small gain in diffusion. I'm not going to do a "how to" on this one cause I don't want you to blame me if you ruin the diffuser set that comes with your MT26EX RT. I bought a second set from Canon to test out my theory that they weren't really diffusing the light very well, and as soon as I got them I used a Dremel tool to remove the front diffusion plastic so I could use it as a base for my own design (I just needed the frame and the clips that hold it to the flash heads). Due to some shipping issues I ended up paying over 60 USD for the set, but it was worth it cause I was right. If you follow me down this rabbit hole with the only set of diffusers that you have for your MT26EX RT you do so at your own risk...

2) 1/4 stop white China silk is the only material I've found that can force the hot spot in the MT24EX (and to a greater extent the MT26EX RT because it has a better built in diffuser) to spread out. It's best to use two layers separated by an air gap, preferably with at least a centimeter between them. Putting one layer of silk directly over another will cause the light to drop by at least a stop, so separate them. You can get it at B&H Photo but it's a little pricey and they have to special order it. I've been looking for a better, cheaper source. Note: The MT24EX knock off flash units perform about the same as the MT24EX, so 1/4 stop silk should also work for them but I have no experience with those flashes.

3) Gary Fong's Puffer Plus makes an excellent last diffusion stage because the light transmittal is good and the surface is dimpled, so it acts like a much larger diffusion surface. I've been looking for a similar photographic grade material that's not curved, but so far no joy.

Here's where things get "tricky": There's a difference between soft light and diffused light, and they are not the same. Using a diffuser that is large relative to the subject will give you soft light, and you can see it in the quality of the shadows. But diffusing the light means forcing the light to spread out, and a large diffusion surface relative to the subject won't necessarily do that. Diffused light means forcing the light to spread out so that the intensity of the light across the diffusion surface is the same. You can have soft light (soft shadows) and a hard specular area. This video explains the difference between soft/hard light and diffused/non-diffused light better than I can.

The quality of your light and the angle between the light, subject, and sensor will determine how much detail you can capture. You can easily lose more detail to poor light quality than to diffraction, and if you focus stack you really should be putting some serious effort into your light. Doesn't make sense to spend all that time to create a stack only to blow out detail with poor specular highlights and a lot of micro contrast (reflective areas throughout a scene where the detail is lost due to harsh light).

Here are some sample images that I've taken with my current lighting setup. Always a work in progress cause I'm never really happy with my light quality, but it's never been better than now and I'm only losing 1 and 1/4 stops compared to the bare flash heads.

Tech Specs: Canon 80D (F11, 1/250, ISO 100) + a Canon MP-E 65mm macro lens (3x) + a diffused MT-26EX RT (E-TTL metering). This is a single, uncropped, frame taken hand held. In post I used Topaz Denoise AI, Sharpen AI, and Clarity in that order.

ImageAsian Beetle on a Sunflower Leaf by John Kimbler, on Flickr

Tech Specs: Canon 80D (F11, 1/125, ISO 200) + a Canon MP-E 65mm macro lens (2x) + a diffused MT-26EX RT (E-TTL metering). This is a single, uncropped, frame taken hand held. In post I used Topaz Denoise AI, Sharpen AI, and Clarity in that order.

ImageMetallic Beetle by John Kimbler, on Flickr

Tech Specs: Canon 80D (F11, 1/250, ISO 200) + a Canon MP-E 65mm macro lens (over 2x) + a diffused MT-26EX RT (E-TTL metering). This is a single, uncropped, frame taken hand held. In post I used Topaz Denoise AI, Sharpen AI, and Clarity in that order.

ImageBeetle Eating Daisy Pollen by John Kimbler, on Flickr

I still have some bumblebee shots that I took a few days ago, and I'll add a frame when I get around to processing them. Glossy black curved surfaces still give me a fit, and I think for some subjects I'm gonna have to get use to ISO 800. Although I'll get more noise, and might lose some detail in the shadows due to a reduction in dynamic range, resolving compound eyes in the specular area better might be worth it.

If anyone has anything to add please jump in. The more I think I know about flash photography the more I realize that I still have a lot to learn.

Edit: As promised here's the bumblebee shot:

I went to my sister in law's house to photograph the Bumblebees in her Lavender. She lives in the mountains above Itri, Italy so I got there at 5:30 in the morning hoping I could photograph them before they got active. Unfortunately it was 20C (68F) and they were pretty much hyperactive. I found a few that were having a tough time getting their metabolism going but they didn't give me much time to shoot. Due to the pandemic I wasn't able to go to her place earlier in the year.

Tech Specs: Canon 80D (F11, 1/250, ISO 200) + a Canon MP-E 65mm macro lens (over 2x) + a diffused MT-26EX RT (E-TTL metering). This is a single, uncropped, frame taken hand held. In post I used Topaz Denoise AI, Sharpen AI, and Clarity in that order.

ImageBumblebee on Lavender by John Kimbler, on Flickr

I like the fact that all of the compound eyes are rendered, even in the harshest part of the specular highlight. I don't like the fact that some of them look flat because I've lost some detail due to the light blowing it out. Due to the angle of the light the shot is a worse case example and I may get better results without the Kaiser Adjustable Flash Shoe on the key. Still better than what I've been able to do in the past, so at least it's a step in the right direction.
Last edited by Dalantech on Sun Jul 19, 2020 7:21 am, edited 6 times in total.

aveslux
Posts: 83
Joined: Tue May 28, 2019 6:13 am

Re: Diffusing a Macro Twin Flash

Post by aveslux »

Hi John, very interesting as always.

How much do you think the size of your subject matters here, I notice you shoot smaller subjects at higher magnifications than I tend to, does the relative size of your diffusers make a lot of difference, whats your light like at 1x?

Have you considered using something like Rosco Cinegel #3026 in place of the Gary Fong, you would have to make a frame to mount it on, but it could be made whatever size you needed.

Increasing the ISO also opens up the exposure to more natural light at 1/250 as well as making the flash faster, this might not be an issue however you might get some natural light bleeding in from highlights if the subject is well lit by the sun, the natural light will be "slower" and blur during the exposure. A faster sync speed would help but 1/250 is the limit.

My other thoughts are it would be nice to have some sort of system for building frames on for experimenting with setups. Outside of 3d printing or something.

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Re: Diffusing a Macro Twin Flash

Post by Dalantech »

aveslux wrote:
Wed Jul 01, 2020 4:09 am
Hi John, very interesting as always.
Thanks, and thank you for a very detailed reply!
aveslux wrote:
Wed Jul 01, 2020 4:09 am
How much do you think the size of your subject matters here, I notice you shoot smaller subjects at higher magnifications than I tend to, does the relative size of your diffusers make a lot of difference, whats your light like at 1x?
The relative size of the diffusers makes a huge difference -in the shadows. My light quality is good at 1x, but it's also my worst diffusion due to the working distance (2.5" at 2x verses 4" at 1x). Even the 2" difference between a 60mm macro lens and a 100mm one makes a big difference. Unfortunately I don't do enough shooting at 1x to have a really good sample set. Here's a shot at 1:1 with the previous diffuser design:

ImageSwallowtail Butterfly at Life Size by John Kimbler, on Flickr
aveslux wrote:
Wed Jul 01, 2020 4:09 am
Have you considered using something like Rosco Cinegel #3026 in place of the Gary Fong, you would have to make a frame to mount it on, but it could be made whatever size you needed.
I may have experimented with something similar a few years ago, and it didn't work very well. But if B&H ever gets it back in stock I could take another look. The dimples on those Puffer Plus diffusers mimics a larger diffusion surface and Cinegel just doesn't have the same properties. But firing through two layers of silk before the Cingel might be interesting (that's something that I haven't done).
aveslux wrote:
Wed Jul 01, 2020 4:09 am
Increasing the ISO also opens up the exposure to more natural light at 1/250 as well as making the flash faster, this might not be an issue however you might get some natural light bleeding in from highlights if the subject is well lit by the sun, the natural light will be "slower" and blur during the exposure. A faster sync speed would help but 1/250 is the limit.
I know that ISO effects the flash and natural light exposure. Most of the time I'm shading the subject because even if I don't get close to the natural light exposure I want to avoid those rainbow colored natural light specular highlights that are caused by the sensor recording movement due to the natural light in the scene.
aveslux wrote:
Wed Jul 01, 2020 4:09 am
My other thoughts are it would be nice to have some sort of system for building frames on for experimenting with setups. Outside of 3d printing or something.
Building diffusers may force me to buy a 3D printer. If I had the CAD file for the MT26EX RT diffusers then I'd order the printer in a heart beat. I've tried working with the folks that make diffusers for the macro twin flashes but they don't seem interested. There's some money to be made in a diffuser that has multiple sections that snap together so you can try different diffusion materials.
Last edited by Dalantech on Sat Jul 11, 2020 10:54 am, edited 1 time in total.

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Re: Diffusing a Macro Twin Flash

Post by Dalantech »

aveslux wrote:
Wed Jul 01, 2020 4:09 am
Have you considered using something like Rosco Cinegel #3026 in place of the Gary Fong, you would have to make a frame to mount it on, but it could be made whatever size you needed.
Light Panels Half White Diffusion might work better, since it's a half millimeter thick plastic that can be cut with a heavy duty pair of scissors. It also has similar properties to Cinegel. The stop loss is 3/4, so not too bad.

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Re: Diffusing a Macro Twin Flash

Post by Dalantech »

I see what's happening now -when I try to edit the original post to fix minor typos or grammar issues the script creates a new post where the entire OP is quoted with my edits in place. Not sure if that's intended behavior or an "undocumented design feature"...

Pau
Site Admin
Posts: 6053
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Re: Diffusing a Macro Twin Flash

Post by Pau »

As admin. I'm able to delete any unwanted post like I did with your former one asking for it.

But because I don't understand the issue I prefer to wait until other admins read it.

Is possible that you're clicking on the quote button in place of the edit one (the pencil) ?
Pau

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Diffusing a Macro Twin Flash

Post by rjlittlefield »

There are two buttons -- one for editing and one for quoting. They're easily confused because after the upgrade they're just graphics, no text. You want the one that looks like a pencil, not the one that looks like a double-quote mark. Hovering your cursor over the pencil will probably produce a tooltip that says "Edit post". That one should let you do an edit-in-place, versus creating a new post that starts with "[quote" .

I see that Pau and I were posting at the same time...

--Rik

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Re: Diffusing a Macro Twin Flash

Post by Dalantech »

rjlittlefield wrote:
Wed Jul 01, 2020 1:38 pm
There are two buttons -- one for editing and one for quoting. They're easily confused because after the upgrade they're just graphics, no text. You want the one that looks like a pencil, not the one that looks like a double-quote mark. Hovering your cursor over the pencil will probably produce a tooltip that says "Edit post". That one should let you do an edit-in-place, versus creating a new post that starts with "[quote" .

I see that Pau and I were posting at the same time...

--Rik
Thanks Rik! Could have sworn I clicked on the edit icon and not the quotes. I either need more coffee or more beer ;)

MarkSturtevant
Posts: 1946
Joined: Sat Nov 21, 2015 6:52 pm
Location: Michigan, U.S.A.
Contact:

Re: Diffusing a Macro Twin Flash

Post by MarkSturtevant »

A widely used kind of diffuser material are the thin sheets of foam wrapping material, such as what you can get as padding around electronic products. It can be also had from office supply stores. It does make good diffuser material, but you do have to come up with measures to cut down on the hot spot.
The dimpled plastic type of diffuser looks interesting. There are products that have a similar texture. One that comes to mind right away are the very cheap translucent plastic panels for fluorescent ceiling lights. These have a diamond shaped dimpling pattern rather than circular one, but I expect it would work well anyway. So now I'm wondering about trying that.

Very interesting thread! I am watching it.
Mark Sturtevant
Dept. of Still Waters

enricosavazzi
Posts: 1474
Joined: Sat Nov 21, 2009 2:41 pm
Location: Västerås, Sweden
Contact:

Re: Diffusing a Macro Twin Flash

Post by enricosavazzi »

MarkSturtevant wrote:
Thu Jul 02, 2020 5:24 am
[...]
The dimpled plastic type of diffuser looks interesting. There are products that have a similar texture. One that comes to mind right away are the very cheap translucent plastic panels for fluorescent ceiling lights. These have a diamond shaped dimpling pattern rather than circular one, but I expect it would work well anyway. So now I'm wondering about trying that.
[...]
I tried these years ago, as well as other materials (e.g. the rubbery transparent sheets to pad the bottom of Ikea kitchen cabinets and drawers). They generally work as diffusers, but at close distances (less than 10-20 cm) from the subject they create an array of small hotspots of their own, especially if the light source is small like a LED. So these materials can be used perhaps as a first diffusing layer, but require an additional layer of more uniform diffusing properties closer to the subject and separated from the first layer by a few cm.
--ES

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Re: Diffusing a Macro Twin Flash

Post by Dalantech »

MarkSturtevant wrote:
Thu Jul 02, 2020 5:24 am
A widely used kind of diffuser material are the thin sheets of foam wrapping material, such as what you can get as padding around electronic products. It can be also had from office supply stores. It does make good diffuser material, but you do have to come up with measures to cut down on the hot spot.
There is such a thing as soft light with a hard specular highlight. It happens when you use a diffuser that's very large relative to the subject, but it doesn't actually force the light to spread out. So it's a large light surface with poor diffusion qualities, and that accurately describes the material you mentioned and why people run into issues with it. It can also add an odd color cast to the light. Not recommended.
MarkSturtevant wrote:
Thu Jul 02, 2020 5:24 am
The dimpled plastic type of diffuser looks interesting. There are products that have a similar texture. One that comes to mind right away are the very cheap translucent plastic panels for fluorescent ceiling lights. These have a diamond shaped dimpling pattern rather than circular one, but I expect it would work well anyway. So now I'm wondering about trying that.

Very interesting thread! I am watching it.
I've tried similar materials for diffusing fluorescent lights and LEDs and they work great -for diffusing fluorescent lights and LEDs ;) I've seen materials that were really good at diffusing a constant light source that a flash would just blow right through, or that simply blocked the light better than diffusing it. Doesn't hurt to experiment with them though, just pay close attention to how much light you're losing compared to a bare flash. Higher light loss does not equal better diffusion.

aveslux
Posts: 83
Joined: Tue May 28, 2019 6:13 am

Re: Diffusing a Macro Twin Flash

Post by aveslux »

So am I right in thinking your current lighting is

Mt-26EX with 2 heads -> Drilled out diffuser cap as a bracket > layer of 1/4 stop silk > 1 cm gap > layer of 1/4 stop silk > Garry Fong puffer plus - > Subject.

That is what I also have, basically copying your setup but also extending the flash head brackets a bit because I also use a 100mm lens as well as the 65mm MPE

What would you say the gaps are between the flash bulb and the 1st layer of silk?

Stupid idea I had how was about a bit of foil on the centre of the puffer plus to reflect the hotspot back into the diffuser so it reflects back around and back out? Maybe cost too much light though.

enricosavazzi
Posts: 1474
Joined: Sat Nov 21, 2009 2:41 pm
Location: Västerås, Sweden
Contact:

Re: Diffusing a Macro Twin Flash

Post by enricosavazzi »

aveslux wrote:
Thu Jul 02, 2020 7:25 am
[...]
Stupid idea I had how was about a bit of foil on the centre of the puffer plus to reflect the hotspot back into the diffuser so it reflects back around and back out? Maybe cost too much light though.
[...]
Something similar is actually done in some speedlight diffusers, which are made of a single layer of diffusing material, with an additional patch of the same material stuck at the center.

Studio softboxes also are typically made of two layers of diffusing material separated by 20-30 cm, the layer closer to the flash bulb covers only the center of the softbox, while the one farther from the bulb covers the entire opening of the softbox.
--ES

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Re: Diffusing a Macro Twin Flash

Post by Dalantech »

aveslux wrote:
Thu Jul 02, 2020 7:25 am
So am I right in thinking your current lighting is

Mt-26EX with 2 heads -> Drilled out diffuser cap as a bracket > layer of 1/4 stop silk > 1 cm gap > layer of 1/4 stop silk > Garry Fong puffer plus - > Subject.
That's pretty close :)
aveslux wrote:
Thu Jul 02, 2020 7:25 am

That is what I also have, basically copying your setup but also extending the flash head brackets a bit because I also use a 100mm lens as well as the 65mm MPE

What would you say the gaps are between the flash bulb and the 1st layer of silk?
From the flash heads to the first layer it looks like there is about a centimeter gap. You should only have to extend the flash heads for the 100mm -that extra 2" of working distance will work against you when using a flash, and I don't recommend any macro lens in the 100mm focal range due to the difficulty in getting good specular highlights. Better off using the MP-E when shooting macro and just use the 100mm for closeups (although there are better lenses for that kind of work as well).
aveslux wrote:
Thu Jul 02, 2020 7:25 am
Stupid idea I had how was about a bit of foil on the centre of the puffer plus to reflect the hotspot back into the diffuser so it reflects back around and back out? Maybe cost too much light though.
Not sure what that would do to the specular highlight (would there be a stripe in it?) and it's completely unnecessary. Currently my specular highlights at 1x are butter smooth after a couple of mods to my old diffuser design. Not gonna go into specifics cause at the end of the day I gotta have something unique for myself. Should be pretty easy to figure out though.

Dalantech
Posts: 694
Joined: Sun Aug 03, 2008 6:57 am

Re: Diffusing a Macro Twin Flash

Post by Dalantech »

enricosavazzi wrote:
Thu Jul 02, 2020 7:54 am
aveslux wrote:
Thu Jul 02, 2020 7:25 am
[...]
Stupid idea I had how was about a bit of foil on the centre of the puffer plus to reflect the hotspot back into the diffuser so it reflects back around and back out? Maybe cost too much light though.
[...]
Something similar is actually done in some speedlight diffusers, which are made of a single layer of diffusing material, with an additional patch of the same material stuck at the center.

Studio softboxes also are typically made of two layers of diffusing material separated by 20-30 cm, the layer closer to the flash bulb covers only the center of the softbox, while the one farther from the bulb covers the entire opening of the softbox.
I experimented with that first idea when I was using the MT-24EX. Ended up ditching it in favor of just using multiple thin diffusion layers to force that hot spot to spread out, which works better than a few thicker diffusion layers. Managed to smooth out the specular highlight while at the same time losing less light.

Post Reply Previous topicNext topic