Jump to content
UltravioletPhotography

UV trichromes with a set of LED chips - what wavelength for red?


Fandyus

Recommended Posts

One of my favorite posts on this forum is this one by Bernard.

https://www.ultravio...lour-uv-and-ir/

I'd love to replicate something like this myself, at least to an extent. I wanted to get myself a set of LED chips with different wavelengths, I would then light up objects with them and create trichromes.

 

This is my current shopping cart.

post-350-0-43195200-1624288047.jpg

400-405nm/390-395nm for red

380-385nm for green

365-370nm for blue

What I'm not sure about is the red channel, I was thinking perhaps I could get something even more longwave to accommodate for the fact that my reach won't go that deep, definitely not as deep as Bernard's for example. And if I do settle for one of those options, which one should I pick? I guess I could also get both since the 400-405nm one only costs about 3 USD. I don't know if there would even be a difference between those two though. Is the 5nm difference worth it?

 

Edit: Update, I just purchased the LED chips. I kept the blue and green options and I went with a 395-400nm for red. If all I get are some shades of yellow, beige and orange, that's fine with me. It's still a more sense-making way to image things in UV than using the odd transmission curves of the bayer array filters. Not saying I'm done with that type of UV photography, but this is I feel superior because it ties into the way we perceive the world already.

Link to comment
Andy Perrin
No, 5nm is not worth it IMO. Honestly, I’m not sure it’s worth doing a trichrome without at least 100nm from one end to the other of the band you are trying to cover because you won’t get a lot of colors in the trichrome if it’s too narrow. We are already kind of pushing it using just UVA. Visible band is 700-400=300nm. (You can make a decent argument that we should be doing this by energy or frequency rather than wavelength since that’s what the color really depends on, but if you don’t go too far from the visible range it’s probably ok to talk about wavelengths.)
Link to comment
One day I would like to do it with 310 nm, 340 nm and 385 nm LEDs. Probably you will get some colors using 340, 365 and 395 nm LEDs too, or even 365, 385 and 405 nm (I consider 405 nm light violet and not ultraviolet, but that is not important). Colors definitely depend on the subject, imaging UV-absorbing materials that start absorbing between 360 nm and 390 nm can give you yellows/oranges/reds depending on the wavelength choice. But Andy is correct saying that a wider spectrum will produce a wider color palette.
Link to comment

No, 5nm is not worth it IMO. Honestly, I’m not sure it’s worth doing a trichrome without at least 100nm from one end to the other of the band you are trying to cover because you won’t get a lot of colors in the trichrome if it’s too narrow. We are already kind of pushing it using just UVA. Visible band is 700-400=300nm. (You can make a decent argument that we should be doing this by energy or frequency rather than wavelength since that’s what the color really depends on, but if you don’t go too far from the visible range it’s probably ok to talk about wavelengths.)

 

So do you think I should just spend my time and money elsewhere? And if not, do you think I should get a 400-405nm or a 390-395nm chip?

Link to comment

One day I would like to do it with 310 nm, 340 nm and 385 nm LEDs. Probably you will get some colors using 340, 365 and 395 nm LEDs too, or even 365, 385 and 405 nm (I consider 405 nm light violet and not ultraviolet, but that is not important). Colors definitely depend on the subject, imaging UV-absorbing materials that start absorbing between 360 nm and 390 nm can give you yellows/oranges/reds depending on the wavelength choice. But Andy is correct saying that a wider spectrum will produce a wider color palette.

 

Thanks for sharing what you think. Good luck with the trichrome you mentioned.

Link to comment
Andy Perrin

So do you think I should just spend my time and money elsewhere? And if not, do you think I should get a 400-405nm or a 390-395nm chip?

Hm, you might want to wait on the project until you have gear that can go a little further into the UV? Alternatively, you could go into the 400nm range to bring some more color in. It's your money, but since the overall cost starts to mount up, it might make more sense to spend the cash on stuff you need more first.

Link to comment
Andy Perrin
Hm. Yeah if the soligor really does have good reach then I’m not sure why he says he is limited. You don’t need IR suppression with the LEDs probably. Maybe the sensor glass is an issue?
Link to comment
Andy Perrin
Nah, I already have a Kuri. Unless it’s a sharper lens, then I don’t need another 35mm. I own a quartz lens if I need reach.
Link to comment

Hm. Yeah if the soligor really does have good reach then I’m not sure why he says he is limited. You don’t need IR suppression with the LEDs probably. Maybe the sensor glass is an issue?

 

The only thing that limits me is that I don't exactly have the money to get LEDs that are under 365nm. But when did I say I was limited? I don't think sensor glass is an issue. I have a converted Canon EOS 1100D from eBay, I am not sure how far it reaches but it's probably not too bad.

Link to comment
Andy Perrin
But when did I say I was limited?

You said,

What I'm not sure about is the red channel, I was thinking perhaps I could get something even more longwave to accommodate for the fact that my reach won't go that deep, definitely not as deep as Bernard's for example.
Link to comment

 

You said,

 

I see, sorry for the confusion. What I meant by that is that I simply don't have monochromatic sources that reach this deep, nor do I have dichroic filters that would be a bandpass at shorter UVA wavelengths.

Link to comment

Filters/light sources are probably the greatest cost for UV TriColour. Not all cameras can reach that deep, some have a UV-absorbing coverglass (example: https://www.ultravioletphotography.com/content/index.php/topic/3608-sony-a7iii-conversion-to-multispectral/page__view__findpost__p__31603), and anyway the coverglass often absorbs significantly down there.

 

Suitable lenses are not hard to find, and you already have one (Soligor).

 

So yes, filters are the most expensive part. They are often small (25 mm or less), and big 50-52 mm filters are very very expensive.

 

UV LEDs may be a bit cheaper, but I think you should use a filter at least when using a 310 nm LED, because I wouldn't be surprised if you saw UV-induced UV fluorescence at that wavelength.

Link to comment
Andy Perrin
UV-induced UV fluorescence might actually be a very interesting thing to investigate at some point.
Link to comment
  • 2 months later...

Chinese UV Leds are something tricky one.

I've bought a bunch of 365nm 3W power Led for 1$ per pc. And now, as it turned out, they are far from be 365nm. In reality they are such (for comparison solid 3W Led from Nichia:

post-367-0-74777200-1630268358.jpg

Besides obvious wavelength slip, Nichia has a real 40% and even 50% light output power, while Chinese "imitations" have only 1-2% light efficiency.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...