Several posts mention how shallow the depth of field is when scanning - which made me ask myself whether that would help to reduce the need for diffusion when a tablet is used as a backlight. Distance could also help to reduce the capture of dust on said tablet.
I tested with a Basic Film Carrier 120 Mk 2 (Negative Supply) on an iPad 10 and Canon EOS R7 with EF100mm Macro lens set to f/8.
Test 1: 12mm (1/2 inch)
Inner carrier only (with extra rubber feet) on the iPad - makes a distance of about 12mm.
Left: focused on the iPad. Right: focused on the film surface.
At 300%, we can easily see the raster of the iPad ⌠and itâs still apparent on the right.
â 12mm distance is not enough with this setup. And definitely not with a wider lens either!
Test 2: 22mm (7/8 inch
Outer and inner carrier (with extra rubber feet each) make a distance of about 22mm.
Left: focused on the iPad. Right: focused on the film surface.
At 300%, we can easily see the raster of the iPad ⌠and itâs gone on the right.
â 22mm distance is enough with this setup. Not with a wider lens though.
This short test reveals that - with a 100mm lens used at f/8 - a distance between negative and carrier of greater than 20mm should be enough to hide the raster of an iPad 10. This means that an additional diffusor is not needed. This also means that exposure times donât get as long as they would be with additional diffusion.
Under very extreme âcorrectionsâ e.g. a very steep tone curve and exposure set accordingly in Lightroom, the raster of the iPad can still be seen at 300% with captures taken with the bigger distance of 22mm. The raster canât be seen under normal viewing conditions though.
Thatâs a valuable test. It suggests that the design of these film carrier systems with integral panels is a bit marginal in terms of film to panel distance when using something like a 55mm or 60mm Micro-Nikkor, or 50mm Sigma Macro and other frequently used lenses for film copying with shorter focal lengths. I settled on around 45mm for my DIY scanning box and a 75mm lens.
indeedâŚand some macro lenses arenât really what their label says.
The Canon EF 100mm f/2.8 Macro has an equivalent FL of about 75 mm when set to 1:1.
For scanning 645 negatives, the lens is at around 86mm. Read about it here:
I have had good experiences with these âLight Enhancement Sheetsâ for my scanning. They break up the patterns in the light while enhancing brightness at the same time. Well worth the $20 for me.
I am fairly certain they are just 3M Light Enhancement film (or similar knockoff) used in 2 layers for extra light uniformity. Creates a collimating effect and increases brightness.
I wondered what the material might be, good to have some alternatives to try. My CS-Lite came with one or maybe two but I havenât tried them with my current lens, but I should try. Now they are sold as accessories and the implication is that they might be used on âany diffused light sourceâ rather than just the CS-Lite,
Lots of products on their site awaiting delivery later in the month, I suppose one could hazard a guess as to why that might be:
I said that Iâd settled on 45mm between my panel and the film but that was not because of any real experimentation, just how it turned out and I knew I wanted it to be at least that. My DIY film scanning box is still the prototype so the dimensions were fixed by the picture frame that I spotted in a charity shop that inspired me to construct it.
I do wonder sometimes if having the panel further away, and masked, could create some kind of âcollimatingâ effect and these sheets might help with that. Indeed on launch they referred to it as a âlight collimating sheetâ
When the light is moved away from the negative, details in and on the negative become crisper. The light is not collimated in any way, but the effect is comparable.
DO NOT EVEN TRY THIS
If weâd use the sun as a backlight, weâd get parallel rays of light and hight power, high CRI light too. Weâd also risk to irreversibly damage our eyes and gear.
Worth trying with B&W negative then. I also have Nikon PB-4/PS-4 setup so it would be easy to use the CS-Lite and just move it away in stages, Iâve removed the PS-4 glass diffuser.
That should work. You should see differences quite early, and as long as the light is in view of your capture, lighting should bee even, but dropping with every step you move the light away.
Thanks, Iâd seen your post, itâs probably what got me thinking. Iâll try it with Vladâs Test Target, that should show up any meaningful improvement. Iâve got a condenser enlarger head I could fire at it as well but that starts getting more complicated.
I never paid attention to the definition of collimation until now, thank you. 10 x FL.
So the âLight Enhancing Sheetsâ (LES from now on) do not collimate so much as they waste fewer light photons by reflecting wide angles back and sending the pproperly oriented photons in the correct direction and you need two to account for X and Y axis. Essentially the LES should be called Light Efficiency Sheets instead.
So now, based on your post from 2021, I want to test raising my LES a bit from the light source and perhaps try other diffusion, too. Generally I would like a brighter lightsource than the CS-Lite, but it works very well for a great price at the moment. Thinking about the Skier but would love to find someone who has done lumin or lux tests on a few sources at proper working distances.
Note that by moving a finite size light source closer or further away from the negative, one changes also the homogeneity of the light falling on the negative, and there are optimal distances in terms of homogeneity as I point out here:
Using an iPad as backlight can have its merit, specially for tuning the hue of the light or as a source of (wide band) r, g and b light. But: Depending on FL of the lens and how far away the iPad is from the negative, the raster can be seen on scans and/or converted images.
The objective of my thread was to find out, how far the source needs to be removed in order to make the raster unnoticeable without additional diffusion.
Other than that, physics apply to any light source. Power density diminishes with the square of the distance (2 * distance â 1/4 power density) and for the most even lighting, the source needs to be faaaar away. And then, it doesnât matter how big the source is. At pragmatic distances, there is often a sweet spot where drawbacks interfere in the least harmful way.
Caution: DO NOT use the sun as a direct backlight.
Itâs faaar away, yes, but has too much power and can damage your eyes ad gear
