Modifying a Zeiss HBO 50W microscope light to give more UVB

[JMC]

Safety note - this talks about high intensity UVB light sources. Please do not attempt this type of work yourself without taking the correct safety precautions.

 

Over the last year I've been building a UV transmission microscope to help with one of my projects. One area that needing modifying was lighting, as I wanted to have a good light source in the UVB region at 313nm.

 

As a basis for it I decided to use a mercury xenon light and a small Zeiss 50W came up for sale on ebay so I snagged that. Here's how it looks.

 

In the exit port is a metal tube with a lens at one end to focus the light. This tube and lens can be moved back and forth by turning a black knob on the side of the light.

 

Problem is the original lens is glass, so no good for UVB as it was blocking everything below abut 320nm. It was a non standard diameter so I couldn't find a fused silica replacement. In the end I had a new tube made, threaded on either end for Thorlabs SM1 tubes, and got a Thorlabs 20mm focal length aspheric fused silica lens as a replacement. Original and new tubes shown below.

 

When I originally used this light for UVB imaging I used it without any lens, so obviously lost a lot of light. As a quick check I did some irradiance measurements without the tube and lens, and then with the new tube and lens in the focused position and in the defocused position.

 

In the focused position the irradiance at the brightest area was about 600x that when no tube and lens was present, and at the defocused position, it was still about 30x more intense than without the lens. Interestingly in the focused position it actually produced a slightly fuzzy image of the two electrodes in the mercury xenon bulb and the arc between them when shone at a surface.

 

The image below was taken using a 313nm Edmund Optics OD4 bandpass filter in combination with a Hoya U-340 4mm filter, using a monochrome modified Nikon d800 and Rayfact 105mm UV lens. The light was shone at a piece of white paper inside a box painted with Semple Black 3.0 paint and it was the image projected on this paper that was imaged.

 

In the image above, the lamp is orientated vertically and the electrodes at the top and bottom of the image. This was about as sharp as I could get it, as I'm limited with how close I can get the lens in the tube to the bulb without touching it.

 

More funky was that using that camera and lens I was able to capture UV videos at 313nm and 365nm using EO bandpass filters in combination with Hoya U-340 4mm. They show the arc flickering and moving about slightly on the electodes. I've uploaded these to youtube, and the links are as follows;

 

313nm video -

 

365nm video -

 

The increase in brightness will be useful for my work, but it could actually be too bright now and damage the samples. I will be trying it in the defocused position, as even that should be significantly better than with no lens. I also may get some fused silica diffusers from Thorlabs and mount those at the other end of the tube with the lens on it, to help make a more even light source for microscopy.

 

As mentioned at the top, work like this can be extremely hazardous, producing very very intense UV light down to and below 300nm. Do not attempt this yourself without an understanding of the hazards involved and how to protect yourself from them.

[colinbm]

Show us your spaceman outfit ?

[colinbm]

The videos look good, I would like to have a xenon lamp like this soon, but I don't know where to start or what I need to buy.

I am waiting for some 309nm LEDs at arrive to make a lamp, perhaps two with 9 x 3w 309nm LEDs.

[Stefano]

Jonathan, I like these experiments. I want to image in UVB one day, in particular I want to take UVB photos outside with sunlight.

 

Colin, I too want to buy 310 nm LEDs. I would like to build a lamp/illuminator like you, but they cost too much for me. One day...

 

also I would have to resist the temptation to see how UVB looks like. I really have to. And jokes apart, yes, UVB is dangerous, so be careful when working with it.

[colinbm]

Stefano,

Get a job or work experience with an electric arc welder, you will see all the light you want & you may even get a 'flash' to experience why you would never want to get another one, ever !

It won't cause permanent damage & depends how bad you get it you will be OK in a few days or a week or so.

 

PS, by the way you don't need to look directly at the light to get a 'flash', because your eyes usually close pretty quickly. A 'Flash' usually catches you unawares in the corner of your eye & burns the retina where light doesn't usually fall much & you won't know anyway till later that night if you got one one not.

[Stefano]

My dad once welded without protection and the following night he felt "sand in the eyes", they hurt quite a bit but he recovered.

 

I know what sand in the eyes feels like as I got actual sand in the eyes when I was a child on the beach. Not pleasant at all.

[colinbm]

Stefano, multiply that unpleasantness ten times, even grown men will be bawling their eyes out from pain & the eyes trying to heal then selves of the pain.

[Stefano]

I definitely would never work with a setup like Jonathan's without protection. You will get burns (unless you are there for a few seconds I guess).

 

Mine was kind of a joke, although I really do wonder how UVB appears to us. Some studies show it looks gray: https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.quora.com/When-people-can-see-ultraviolet-what-is-the-color-they-see&ved=2ahUKEwiC09r-v7DtAhWBxqQKHVMMA7MQrAIoAnoECBQQAw&usg=AOvVaw0lzc0-RX49k4CvR9VGaD_n&cshid=1606953620504

 

So yes, they actually made people look at UVB light down to 302 nm. What's interesting is that I should be able to see UV to about 310 nm, as I have young eyes (although this may not be a sufficient condition), in that study people saw that deep (in fact, even deeper) and Wikipedia says:

 

"[...] Under some conditions, children and young adults can see ultraviolet down to wavelengths around 310 nm".

 

Probably it looks gray.

 

Hope I didn't go too much off-topic.

[JMC]

Thanks Colin. This lamp is mercury xenon and as such has a very 'peaky' emission spectrum, not ideal for normal imaging. I like it because it gives strong lines at 313nm and 365nm in the UV which as I am doing imaging of sunscreens in the UVA and UVB region is ideal as I can use narrow bandpass filters to isolate those two regions. For more normal photography though, flashes or LEDs are the way to go.

 

Stefano, UVB imaging in sunlight presents a lot of challenges for filters (as well as camera and lens). I've done some but with monochrome converted camera, quartz/calcium fluoride lens and stacked filters (Invisible vision 308nm and Hoya U-340 4mm). Certainly not straightforward especially if you want to be certain you've blocked most of everything else that isn't UVB.

[colinbm]

Thanks Jonathan

Are these mercury xenon lamps like the automotive xenon headlights, but with UVB transmitting glass ?

[JMC]

Not sure Colin, I don't really know how the car xenon bulbs work. I doubt the car ones have mercury in them though.

 

This is what the lamp uses - https://www.slclightingonline.com/products/osram-hbo-50w-ac-l1

 

To give you an idea of size it's about 2 inches long.

 

Main issue with them is lifespan. Even if used properly, lifespan is quoted as 100 hours, so it's an expensive light to run.

[colinbm]

Thanks Johathan

Yes the car HID headlights are marketed as xenon, but are mercury, but have a non UV glass.