Hello,
I would like to know what can be recommended as a diffuser for use in a DIY scan light source.
I would like to use a good/high quality diffuser if possible.
But is there any difference from the general diffuser?..
Your input would be appreciated.
Easiest thing would be to get what is available in places near you and try out a few things.
When I did such tests, I used a piece of frosted/milky plastic and found that the position between the light and the object can make a difference.
The diffusor looks fairly neutral, but it changes colors slightly. Watch out for this while shopping.
I have already read the post you wrote, and it helped me a lot.
Thanks!
Watch this video. I’ve never tried using those diffusers, but you might have or get a cheap LCD TV.
Thank you for the information!
Here in the UK, the very best light diffuser for optical use (scanning) is from Perspex.
The material is a cast acrylic, code is 1TL2, range is LED Spectrum, and it passes 51% of the light (so loses just 1 stop) and has a flat curve across the visible spectrum (and beyond), so there will be zero color casts.
There are various sources to get this, some places will pre-cut to your req’d size, some will drill holes as you need.
If you are un the US, then there us a Plexiglas equivalent - check their website for LED/light diffusion material.
These diffusers are quite expensive (for diffusers) - you canb get far cheaper parts, but they will have a nasty colour case that is non-flat, so you’ll fight to remove the cast as it’s not even.
By “expensive”, I’m talking maybe <£20 for an A4 sheet… so not that expensive at all really, for a one-off.
Hope this helps
Andrew
Thank you for your detailed explanation. 
I will search for the diffuser you mentioned.
I found a seller on ebay.
I am going to order 3mm and 5mm thickness for testing.
Both sizes are A4. The total price is ÂŁ10.20.
Thanks!
p.s. Oh, this acrylic is the same one used in the Essential Film Holder.
can you let us know which seller on Ebay it is?
This is the lighting setup I use when capturing 4x5 black and white negatives. (1x2 multiple captures).
I use a 3mm thick acrylic plastic sheet.
Although you can’t see it in this photo, I put a rubber pad between the acrylic plastic sheet and the LED panel.
I use 2 pieces of Tru Vue museum glass.
When I scan, I turn off all lights in the room. Also, I scan 4x5 films with a large black mask.
Very nice, very simple set up!
Both work well. Consider light loss - the 3mm loses around 1 stop of light, 5mm loses nearly 2 stops.
…and adding several layers of diffusion will improve diffusion, but will also waste more stops…
An excellent diffusion material is “matte drafting mylar.” The brand “Dura-Lar” comes in pads. I use two layers, separated.
Note: With this product, most diffusers, and most all diffused light sources, there’s some texture on the surface. Keep your film a bit away from the surface so it’s well out of focus.
Andrew, can you please be a bit more specific on Perspex LED Spectrum 1TL2?
I’m asking because it’s said that 1TL2 is LED optimized. It seems like this is a contradictive claim since most LEDs do have some spike in blue. Does it mean that 1TL2 tries to correct the spike in blue?
You say that the 1TL2 “has a flat curve across the visible spectrum (and beyond)”. I can’t find any data. Did you do any measurments? Does it mean that the spectrogram of white light coming through 1TL2 would look more like a flat horizontal line?
I’ve just found a solution by Schott. The used to make Opalika glass diffusers. They did publish a spectrogram and it’s almost a flat line. Here it is
I think that to an extent the optimisation for LED has more to do with the avoidance of hot spots for LED signage. That’s also good for film copying of course. However the Spectrum ITL2 is claimed to pass the entire visible spectrum uniformly.
Thank you. I saw this guide but these marketing papers are unconvincing to me. I’m just trying to find the best solution regardless its material.
I think the easiest way is to compare two spectrograms. Does anybody who has ITL1/ITL2 also have a spectrophotometer to measure the spectrum of incoming/outgoing light?
