[UVC SAFETY WARNING] UV-C light sources & filters

[colinbm]

The Sun, from outside of our atmosphere.

[lukaszgryglicki]

OK, but for people living on Earth surface?

 

[Andy Perrin]

Probably excimer lasers.

[dabateman]

7 hours ago, lukaszgryglicki said:

Simple question:

What is the most powerful UV-C light source?

Maybe a power plant substation. You can see bursts of UVC off some badly installed or old high KV power lines. 

I stll need to image the substation near me. But need to find a good angle and be careful not to draw too much attention. 

I was photographing flowers on the grass near Walter Reed Navy base, and the cops blocked 4 lanes of traffic to ask me some questions once. The person whom called in must have been upset, with the rush hour traffic and wanted to wait longer. 

[Andy Perrin]

I guarantee excimer lasers have higher intensity than flashes from substations…I mean, they are frickin laser beams, as Dr Evil would say. 

[Doug A]

@dabateman UV C arcs at a substation is fascinating. Is this something that is always happening or time sensitive, like a lightning bolt?  

 

Thanks,

Doug A

[Andy Perrin]

Doug, I haven’t seen it but I didn’t think it was arcing (that would be bad!). Instead it’s like aurora, from ionized air. So more of a consistent glow. There were pics from a few years ago on here from someone who imaged them. Maybe I’m wrong though. I definitely haven’t seen it with my eyes.

[dabateman]

12 hours ago, Doug A said:

@dabateman UV C arcs at a substation is fascinating. Is this something that is always happening or time sensitive, like a lightning bolt?  

 

Thanks,

Doug A

I don't know,  I haven't tried to image it yet.

My understand was uvc is used to detect problems on the lines. So maybe shouldn't be continuous.  Radio interference can also be used, as a pin of damaged cable with high voltage will emit in the radio frequency. 

[Andy Perrin]

Not continuous but not arcs either - nfoto got the pics here:

https://www.ultravioletphotography.com/content/index.php?/topic/763-power-lines-in-uv/#comment-2435

Quote

Most of the UV-visible spots are on the insulators, while some are on the cables themselves. I shot many frames and the cable spots are not in the same position. Many appear to be flares or jet-like like you see of eruptions on the Sun.

 

[enricosavazzi]

For what is worth, there are special X-ray-activated phosphors that emit UVC, although these are still experimental materials:

https://www.nature.com/articles/s41377-018-0089-7.pdf

 

As for generating X-rays, there are unconventional methods like unrolling a roll of adhesive tape in a vacuum (unrolling the tape in air generates electrostatic sparks and a little VIS and UV, but doing it in a vacuum has a whole different effect - the X-rays are strong enough that this method has been proposed as a way to make radiographs of human subjects in regions where electricity or X-ray machines are not available or unreliable).

[colinbm]

Enrico, there was a guy on Flickr that made his own X-Ray machine & posted some wonderful pictures, until he got the knock on the door & his gear confiscated  

[enricosavazzi]

Yes, I guess that it would be difficult to camouflage an X-ray tube as something else, especially now that TVs no longer require a giant cathode tube. The anode that emits X rays also requires some hard-to-get materials. The oldest tubes used a steel anode, which became visibly corroded after just a few hours of operation.

 

When I was a teenager in high school, people were just becoming aware of the dangers of X rays exposure. Until a few years earlier, a lab technician in the same school used to make regular demonstrations of the penetrating power of X rays generated from a large cathode tube (some 20 cm in diameter as far as I remember, with an anode-to-cathode gap of some 10 cm) by placing his hand between the tube and a fluorescent screen.

[Andy Perrin]

Oh wow, that poor tech. 

[lukaszgryglicki]

Does it make any sense to consider this kind of UV-C lightning: https://www.ebay.pl/itm/263910516555?hash=item3d724ae74b:g:grUAAOSwmBVhJwNH&amdata=enc%3AAQAHAAAA4BwX35fpz5mWZoacXvcsoLZ4wSFQT%2BniJHKRUqXBH6b4nP%2BCQ5ed0tzG0mz7ZKW6svQ3DIH8TOb2i9gFkMIgc%2FZE4K5f6bbuoQr1FS0VrEfREIPIu3YlDl%2FYgkmdgqqGXZv7j3g2IZYs1crbbbgYtmscsaoPpFnWbZfjRYQGAdbN%2FLMTUr%2FfngUJyo3h%2FV6Jral%2FpsET0teS1mNCG40EjTgjWvcvyYPfsSm%2B8riGBnq2E%2BiT%2FWntGeFdS1ERGAgtcKWLs3KR%2B0Lu5ROdodHeB%2BDmLn4XblXqcze0YhcZQSGc|tkp%3ABk9SR9TpzpPbYA ? (it's from China so...) 

8W input transmitts to ?? W UV-C output?

 

 

[lukaszgryglicki]

Also considering this https://www.faruvclite.com/faruvc222/

Requested quote and specs.

 

[lukaszgryglicki]

Seems like I can get some UV-C filter that will probably work for both 222nm and 254nm:

- https://www.edmundoptics.eu/p/228nm-cwl-25nm-fwhm-50mm-mounted-diameter/22998/

And then adapter to mount this 50mm filter in 52mm filter ring: 

- https://www.edmundoptics.eu/p/2-diameter-threaded-filter-holder-for-imaging-lenses/26485/

Then it should be usable for imaging in mercury vaport lamp or KrCl excimer lamp...

I just asked them for exact chart for OD blocking above 300nm till 1200nm.

 

[Andy Perrin]

In terms of blocking, for the Edmund filter expect similar performance to Reed’s results in the other thread. It will leak IR around OD4 most likely. Andrea also had major issues with her Edmund 340nm filter with reflections. I don’t know if this will have a similar issue or not. 

[lukaszgryglicki]

IR OD4 is expected (probably even more becaus ethey only say average OD4) - but this is *while passing* required UV-C band, I think I will give up searching *anything* that passes wideband UV-C and blocks everythign else - after a long long search I think there is no such thing at all.

 

[Andy Perrin]

Yes, but the camera is much much much less sensitive to UV-C than to IR, so to have the UV-C "win" the contest, you need much better than just OD4 blocking, unless your light source happens to have no IR in it.

 

I agree with your conclusion that the perfect UV-C filter doesn't exist. Probably your search should be directed towards light sources that won't have too much IR.

[JMC]

Lukas, if you get the Edmund filters, please do share your experiences (especially with regards to blocking).

 

Here's an example for you about the challenge with blocking unwanted wavelengths.

 

In my UV microscope, I use a mercury xenon lamp, and image (mainly) at 365nm and 313nm. The light passes through the microscope and sample and the a photoeyepiece before hitting the sensor.

 

I place filters on top of the photoeyepiece as the last step before the light reaches  the sensor to remove the chance for things like fluorescence. At 313nm I use 2x Edmund Optics 10nm bandpass filters stacked on top of each other, as one was not enough to remove enough of the unwanted wavelengths to give me a 'pure' 313nm image. These filters are each OD4, and each passing over 70% of the 313nm light.

 

You will be working in relatively unexplored territory that far into the UV so finding solutions wont be straightforward unfortunately.

[lukaszgryglicki]

Yes, so my idea is that light source will not have IR so I still hope one edmund filter will be OK.

I already pasted two links - they are for 222nm and 253nm.

I think that if I use mercury lamp then I need to have anything that passes 253nm best and then it should have as good as possible blocking of other mercury lines:

184.45		ultraviolet (UVC)
253.7		ultraviolet (UVC)
365.4	I-line	ultraviolet (UVA)
404.7	H-line	violet
435.8	G-line	blue
546.1		green
578.2		yellow-orange
650.5		red

 

So I want 184 & 253 lines and I *need to block as good as possible* 365, 404, 435, 546, 578, 650.

 

For 222nm KrCl excimer lamp I'm not sure what other frequencies are emitted in addition to 222nm line - I know it emits some 250-280 because there are special filters to block this dangerous frequencies (called 222nm pass filters by China company), but I have no knowledge if they emit UV-A, UV-B, Visible, IR? Anybody have emission spectrum example of KrCl lamp?

 

In any case I will wait for my GFX 50R mono - Dan said he already has quartz coverglass for me (it will be mono naked sensor with only quartz glass above it said to pass from 180nm). I will see if entire thing is worth continuying when I see how this camera "see" KrCl excimer lamp and low pressure Hg lamp without any filters...

 

[Andy Perrin]

Lukas, I think you should try to do this in steps. Like first do UVB imaging and work out the issues, then go deeper gradually. Taking a “moonshot” approach of going straight to 222nm will probably fail and you won’t be able to understand why. If you do it step by step it will be more clear where the issues are. 

[dabateman]

On 9/12/2022 at 3:22 PM, lukaszgryglicki said:

Yes, so my idea is that light source will not have IR so I still hope one edmund filter will be OK.

I already pasted two links - they are for 222nm and 253nm.

I think that if I use mercury lamp then I need to have anything that passes 253nm best and then it should have as good as possible blocking of other mercury lines:

184.45		ultraviolet (UVC)
253.7		ultraviolet (UVC)
365.4	I-line	ultraviolet (UVA)
404.7	H-line	violet
435.8	G-line	blue
546.1		green
578.2		yellow-orange
650.5		red

 

So I want 184 & 253 lines and I *need to block as good as possible* 365, 404, 435, 546, 578, 650.

 

For 222nm KrCl excimer lamp I'm not sure what other frequencies are emitted in addition to 222nm line - I know it emits some 250-280 because there are special filters to block this dangerous frequencies (called 222nm pass filters by China company), but I have no knowledge if they emit UV-A, UV-B, Visible, IR? Anybody have emission spectrum example of KrCl lamp?

 

In any case I will wait for my GFX 50R mono - Dan said he already has quartz coverglass for me (it will be mono naked sensor with only quartz glass above it said to pass from 180nm). I will see if entire thing is worth continuying when I see how this camera "see" KrCl excimer lamp and low pressure Hg lamp without any filters...

 

Lukas, Your forgetting that the Germicidal bulbs are not point line lasers with distinct lines. There is a curve a light. This isn't the best image, but one I have on hand to show you the UVC curve of light coming from a 15W germicidal bulb:

 

[lukaszgryglicki]

Right, thanks for this, really helpful.

I decided to wait for my GFX to arrive and then see how it "sees" mercury lamp and KrCl lamp (already have both)

Then I will decide what to do.

Probably will buy one bandpass filter for 222 and one for 254 (both from Edmund - they make 50mm filters with adapter so they can be made into standard 52x0.75mm filters).

Those "dedicated" FWHL 10nm filters offer highest transmission, I'm just not sure how good they block in Visible & IR.

 

[lukaszgryglicki]

So when I use my mono Nikon, UV-Nikkor and

- Hoya U-340 4mm thick - I can see through window with maybe 1 stop loss or even less.

- When I use IV308 - I cannot see through window but when I do images outside (on the balcony) I'm getting a bit light foliage (I've already pasted those photos). Same happens with CFA Nikon - even lighter foliage and this looks like IR leak. I wonder why light foliage when window is almost black or 6 stops darker.

 

Now when I stack this Hoya 340 & IV308 window:

- window still pitch black (with mono) - can't see anything even using ISO=1600, f=4.5, t=30s (all on UV-Nikkor) so even blacker than it was (window has 3 glasses).

- when used on mono full spectrum camera I need around f=11, t=30s, ISO=100, so I've fixed t=30s, ISO=100 and was able to get images with f range 5.6 till 16, depending on current sun conditions (we have around 50% sunlight now). Foliage much darker.

- when used on color full spectrum camera: I need f=5.6 max f=8, then I need t=30s and I need ISO at least 3200 even 6400! Foliage finally a lot darker - so in previous examples color photos were mainly IR while mono photos were mix of UV-B and IR in unknown ratio. Now with H340+IV308 finally UV-B, I don't see anything IR-like... quality is very bad but even if I do f=4.5, t=30s then I still need ISO 3200 and image is noisy and f=4.5 is also dreamy - I could possibly do bulb but will probably need 20-30 minutes exposure which would add longe exposure noise anyway, so probably something like f=8, t=15min ISO=1000 would be needed with CFA, while mono is just fine with ISO=100 and t=30s or even 20s with f=8 or even 11.

 

color shots: *the question is - is it finally UV-B color as passed by CFA or rather IR leaks?*

white balance not changed:

white balance changed manually - I couldn't white balance to any photo element due to high iso noise (also still see leaks but this is ISO 6400 and 30s and I black taped everything around my camera and it still leaks a bit):

 

mono images:

That's it. Note how all windows/glass are pitch black everywhere