Jump to content
UltravioletPhotography

340 nm LED (first impressions)


Stefano

Recommended Posts

Hi,

This time I went down a bit with the wavelength, and tested a 340 nm LED, found on ebay. The seller claims are:

  • 3 W power (in)
  • 60-70 mW power (out)
  • 340-345 nm wavelength

The LED is rated at 500 mA, and the forward voltage (from the seller) is 4.6-5 V. The viewing angle is quite wide, at maybe 150° (I didn't measure it yet).

 

That's how it was delivered:

 

Paper container (with bubble wrap inside), with a "ball" of bubble wrap inside

post-284-0-01378600-1584049403.jpg

 

Inside the "ball" there was what looked like an anti-static bag and some kind of protective material (like a spongy styrofoam). The LED was inside the bag of course.

post-284-0-53910500-1584049711.jpg

 

The LED came with two blobs of solder (I will not solder wires in those places anyway)

post-284-0-44187000-1584050033.jpg

 

So... this LED created a lot of questions.

 

That's what I expected:

  • Green color when white-balanced with a typical UV sunlight white-balance;
  • Dark glass (not black, but noticeably darker);
  • Weak output (that's normal for this kind of LED, they are inefficient below 365 nm).

As you will see, the reality turned out to be (mostly) different.

 

About the color... this is the color of a 365 nm LED torch, with white-balance;

 

(243, 255, 7)

post-284-0-14331400-1584050215.jpg

 

It is a greenish-yellow, no surprise here.

 

...And that's the LED, seen through a 3 mm thick ZWB1 filter:

 

(255, 255, 6)

post-284-0-49809800-1584050319.jpg

 

That's almost a pure yellow. I think that it shouldn't be like that, maybe something is wrong with my camera.

 

Comparison:

post-284-0-60696100-1584050412.png

 

Regarding color... I had a curiosity to satisfy: how does the human eye see light at this wavelength? It happend (mostly accidentally) that I saw a 365 nm LED, with no protection. It appeared violet, especially at low power. So, I thought that 340 nm would have appeared violet too. And... I did it. Running the LED at low power (~1 mW output), and filtering it with the same 3 mm thick ZWB1 filter I used before (the LED emits a bit of white light), I briefly looked at it. I noticed two things:

  • My eyes have a huge focus shift down there (that was expected);
  • And... the LED appeared BLUE! Not violet, but a rather nice shade of blue, a bit on the violet side.

L cones (red) lose almost all sensitivity in the shortest wavelengths, and the only cones still sensitive enough are the "S" (blue) type. I will not do that again, and I will always use eye protection when working with this LED.

 

That's the approximate color I saw (50, 0, 255)

post-284-0-72155300-1584051551.png

 

So, light appears violet only in a well-defined range, maybe ~350-430 nm. If you want to see the same color I saw, but MUCH more safely, look at a 435-440 nm LED or at the 435.8 nm mercury G-line.

 

Running the LED at the maximum rated current of 500 mA required 4.2 V, for a total power consumption of 2.1 W. LEDs emitting at short wavelengths require high voltages (see here https://www.ultravio...ng/page__st__20). At 340 nm, photons have an energy of ~3.6 eV, compared to 3.4 eV at 365 nm.

 

Another surprise came with imaging. The LED is quite weak, but it can easily light up paper, highlighters, and all sort of fluorescent objects. The colors seem the same of those under 365 nm light. I needed f/2.8, ISO 80 and 60 s of exposure to capture correctly exposed images, with the LED ~20 cm away from the subject. I initially used my ZWB2 (2 mm) + chinese BG39 (2 mm) stack, which worked well, but then I switched to the ZWB1 filter above (used alone). Unless otherwise specified, the settings are f/2.8, ISO 80 and 60 s exposure for all images. They aren't white-balanced.

 

Not really the best way to run the LED, but it works and it dissipates heat well. I then made contacts with the alligator clips by holding them with my hand. I will eventually mount it properly to the heatsink.

post-284-0-00211500-1584052922.jpg

 

I noticed two things:

  • Typical glass does NOT darken, it is still very transparent;
  • Apparently, glass fluoresces emitting UV light (UVIUVF). You can see it glowing in my photos.;

Glass piece

post-284-0-70748300-1584053127.jpg

 

My hand

post-284-0-61729200-1584054108.jpg

 

Glass jar (340 nm)

post-284-0-36428200-1584053419.jpg

 

(365 nm), with a torch ~1.5 m away

post-284-0-25383500-1584053501.jpg

 

Same, but only 15 s of exposure

post-284-0-68185400-1584053559.jpg

 

Last, this is the link if you want to try it (again, I have no affiliation with the seller): https://www.ebay.com...60eff6adc8ac2df.

I also noticed that my LED, this one https://www.ebay.com...Ks2aloJPxecaFYg, and this one from Thorlabs https://www.thorlabs...tgroup_id=6071# all seem to be the same exact one, more precisely this one https://www.aptechno..._AAP63_60mW.pdf

 

That's all for now, maybe I will post other experiments in the future.

Link to comment
Andy Perrin
Usually the lens of your eye absorbs most of the UV (hence the stories about people seeing UV after having the lens removed for cataracts). I wouldn't expect you to see anything at all directly at 340nm, but what you might see is fluorescence of either the lens or fluids in your eyeball. Anyway, I'm glad you are donating your eyes to humanity, Stefano. Very kind.
Link to comment

Usually the lens of your eye absorbs most of the UV (hence the stories about people seeing UV after having the lens removed for cataracts). I wouldn't expect you to see anything at all directly at 340nm, but what you might see is fluorescence of either the lens or fluids in your eyeball. Anyway, I'm glad you are donating your eyes to humanity, Stefano. Very kind.

The lens should start absorbing below 400 nm, and the cornea below 360 nm, according to Wikipedia. I saw a sharp image of the die, so probably it wasn't fluorescence. People have reported seeing X-Rays as a blue glow coming directly from the inside of the eye. Nobody (hopefully) tries to see X-Rays by pointing an X-Ray tube in front of their eyes today.
Link to comment
When I was a child, I used to stare at the sun for minutes, until I saw it as a green circle... glad I stopped doing it.
Link to comment
Andy Perrin
The lens should start absorbing below 400 nm, and the cornea below 360 nm, according to Wikipedia. I saw a sharp image of the die, so probably it wasn't fluorescence.

That conclusion is the opposite of the one you should draw? If the lens absorbs the light, it won't hit your retina to be seen. If you saw a sharp image of the die, it isn't focal shifted, so it is probably visible light fluorescence. What we have here is a nice case of confirmation bias.

Link to comment
And I think that my camera does that as well, I don't think it goes deep into UV. Are you sure that 340 nm UV should appear green?
Link to comment

 

If you saw a sharp image of the die, it isn't focal shifted

I had to refocus. Once I did that, the image became sharp. It was blurry without adapting my eye to it.

 

Also, I couldn't see anything with a pair of polycarbonate goggles (and the ZWB1 filter). I am pretty sure I saw UV.

Link to comment
Andy Perrin
Well, I obviously don’t advise double-checking! But it seems very unlikely, especially at such a short wavelength.
Link to comment

Well, I obviously don’t advise double-checking! But it seems very unlikely, especially at such a short wavelength.

Keep in mind that I looked directly at a small light source, and my eyes are still young. Children, according to Wikipedia may be able to reach UVB at 310 nm. Isn't it impressive that I can see remote control LEDs? I'm not that good with sound, I tested myself and I am deaf above 16250-16500 Hz. Some people have no problem reaching 18000 or above.
Link to comment
Andy Perrin
I don't think I ever did a careful white balance (off PTFE) with that filter, just informal white balances off pavement.
Link to comment
If someone wants to build a 340 nm UV flashlight, a piece of Hoya U-340 would be absolutely perfect to filter it. If a manufacturer made a 4 dies version, we could have a respectable 200+ mW of output power at 8.4 W power consumption. Very inefficient, but it would be not that bad.
Link to comment

I think I should try a narrowband filter with this LED. JMC obtained darkish glass at 335 nm, if I recall right (I have to find where).

 

Yes, here (not at 335 nm). https://www.ultravioletphotography.com/content/index.php/topic/3536-composite-uv-imaging-using-multiple-filters/ He got darkening at 350 nm. Various types of glass have various transmission curves, but I believe that my images are not right. Something is wrong.

Link to comment

Here are a couple bandpass filters for you:

152147258592

152147258593

 

Also, what makes you believe your camera is sensitive to 335nm, and your lens transmits 335nm?

Link to comment

Here are a couple bandpass filters for you:

152147258592

152147258593

 

Also, what makes you believe your camera is sensitive to 335nm, and your lens transmits 335nm?

Nothing. I don't know if my lens goes down to 335 nm, but from my tests with this LED it seems that I have a hard time recording 340-345 nm (I think that glass should be darker).

 

How should I use the numbers you provided above?

Link to comment

Well noticing that you live in Italy, I hope your safe from the craziness outside.

Then reading posts like this of looking into UV LEDs I hope your safe from the crazy experiments you are doing inside.

 

Please don't look at UV LEDs. That will not help anyone.

Link to comment

When I tried using 365nm LEDs and measured the wavelength emitted, there was a range from below 365nm up to and in to the visible region. I would presume that a similar effect would be going on with the 340nm and that a range of wavelengths are being emitted, that go well into the long wavelength UV region. The camera is much more likely to be picking up the lower levels of these longer wavelength components, than to be capturing the 340nm part.

 

Sorry, but did I miss what camera and lens you're using? Most lenses will have very little transmission by the time you get down to 340nm, and even the camera sensors are starting to give up there compared to the longer UV wavelengths, because of the absorption of the Bayer filter dyes.

 

Also with your eyes, aren't you just seeing the longer wavelength end of the distribution?

 

Please, don't tell me you're looking into a UV LED at 340nm without any eye protection.

Link to comment

Sorry, but did I miss what camera and lens you're using? Most lenses will have very little transmission by the time you get down to 340nm, and even the camera sensors are starting to give up there compared to the longer UV wavelengths, because of the absorption of the Bayer filter dyes.

Panasonic DMC-F3 (a compact point-and-shoot camera).
Link to comment

Nothing. I don't know if my lens goes down to 335 nm, but from my tests with this LED it seems that I have a hard time recording 340-345 nm (I think that glass should be darker).

 

How should I use the numbers you provided above?

 

Those numbers are eBay item numbers, go to eBay and paste those into the search field.

Those are both surplus filters from the Omega boys. narrow bandpass, and they look like they both have good out of band suppression too. Cheap compared to what you would pay for one of those anywhere else,

and the peak transmission seems rather good for a narrow UV BP filter too.

Link to comment

Please, don't tell me you're looking into a UV LED at 340nm without any eye protection.

Yes, I have to tell you. I did it briefly, for a fraction of a second, but I did it. (and I will not do it again). Just know that 340-345 nm appears blue. I bet you didn’t know that.
Link to comment
Ok, so from what I can find online, that lens has 6 elements in 5 groups. I would be highly surprised if you you are getting much if any light through that lens at 340nm. Which brings me back to thinking it is the longer wavelength light from the LED which you are picking up.
Link to comment

Ok, so from what I can find online, that lens has 6 elements in 5 groups. I would be highly surprised if you you are getting much if any light through that lens at 340nm. Which brings me back to thinking it is the longer wavelength light from the LED which you are picking up.

I think that I am seeing mainly UVIUVF (glass has a blue glow). Things fluoresce behind glass, so there is some transmission.
Link to comment

Well noticing that you live in Italy, I hope your safe from the craziness outside.

Then reading posts like this of looking into UV LEDs I hope your safe from the crazy experiments you are doing inside.

 

Please don't look at UV LEDs. That will not help anyone.

I have a lot of free time now, no school, no meeting with friends, I must stay at home unless I have a valid reason to go outside...
Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...