A cheap do-it-yourself point light source for simple spectral comparative UV transmission studies

[Kai]

When photographing in the invisible, it is of course essential to know in which spectral range you are recording. For this purpose, light sources, filters, lenses and sensors can be measured very precisely with spectrometers or monochromators. There is a good deal of material on this here and elsewhere.
 

Unfortunately, not all components been measured yet. And not everyone has scientific equipment available. 
A good approach is then to take a photo of a spot or narrow strip of the light source used (e.g. sun/slit) through a suitable optical grating. In this way, meaningful comparisons of entire systems or even components can be made. However, depending on the composition of the air (humidity, particles) and the distance traveled by the light through the atmosphere (time of day, geographic latitude), sunlight arrives at a very individual spectrum - especially in the range below 400 nm.

So I was looking for a light source that I could easily access, that was point-like and also produces UV in the range well below 350 nm (albeit with more lines than mercury).
My solution is a small, cheap spark generator. A 6V battery with low internal resistance serves as the energy source, the spark is generated between small titanium and molybdenum electrodes.

Material


Assembled device


Grating, lens and camera

The combination of the spectral lines of both metals results in a relatively even distribution of the emission lines from 320 nm far into the VIS range. The spark is set very short (max. 1 mm). I record the spectra at a distance of 1-2 m through an optical grid, which I adapt in front of the lens. With a UV filter you can measure in a bright room, without a UV filter you need an absolutely dark room.

 

Here is a comparison of three light sources: 1. An Osram Xenophot 100W halogen bulb (average lifespan of only 50 hours); 2. Sunlight and 3. Ti-Mo Spark.
In this case, one sees that the UV limitation in sunlight comes from the sunlight, which the optical system appears to be sensitive to at shorter wavelengths as well. 


Comparison of different light sources: Xenophot 100W, sunlight, Ti-Mo-spark
(500D FS at 1600 ASA, enlarger lens, filter, DVD grating, slit, light source)


In the second example one can see the difference between Baader-U and Schott DUG11X filters:




In the meantime I have measured dozens of lenses and filter combinations and learned so much about my own equipment. 

 

[colinbm]

Nice set-up Kai
The information on the spark generator & the spark, generated between small titanium and molybdenum electrodes, could be explained a bit more please.

[Nate]

Thanks Kai for the setup, just wondering what the center tube is. Transformer?

[Kai]

Hi Colin,
I'm glad if you're interested in this!
There are many possibilities - depending on your the requirements and manual skills. An example can be seen in the pictures. It's probably easier with cardboard and hot-melt glue.
The plastic holders shown here are printed from PLA and can be rotated in the wooden board. This allows the spark size to vary nicely. The electrodes can then be clamped into the crocodile clips.
Titanium and molybdenum wire are available - like all other components shown - e.g. at Amazon or ebay.
Good luck with crafting ;) 

[Kai]

Hi Nate,
yes, the high voltage generator is glued into the black plastic tube. Such a (ready-to-use) high-voltage source can be had for around 10 euros/dollars at Amazon or ebay.
Advertising claim: "DC 6-12V, 1000kV high voltage generator, pulse generator".
It is strong enough for a few mm long sparks. I think they will find application in electric fly swatters.
Have fun experimenting ;) 

[ulf]

I think that the HV-generator might draw more current than the switch  can handle comfortably.

I suspect that the spark act as a wide band radio noise emitter too that can give problems for Wi-Fi, Bluetooth and normal radio transmissions.

[Stefano]

I think it's illegal to generate radio waves like that in most countries. You could build a Faraday cage around it (or put it in a microwave, but I don't know how well it would shield EM waves. Microwaves are designed to shield well their operating frequency, which is 2.45 GHz. Higher frequencies can leak).

[Andy Perrin]

It will definitely make radio waves but shielding down to microwaves would cut all the regulated sources of interference. Higher frequencies wouldn’t matter (and would likely be absorbed by walls). I think the solution is just to put a small faraday cage around it. Also, it makes radio waves only during the spark which would not be very frequently— how often do we need to do these tests anyway?

[Stefano]

53 minutes ago, Andy Perrin said:

Also, it makes radio waves only during the spark which would not be very frequently

I don't know how that device works, but I imagine it makes several sparks per second, almost continuously. I don't think you image a single spark at a time.

 

If you use it for a couple of minutes it shouldn't be a big problem. I would still build a Faraday cage around it.

[Andy Perrin]

You misunderstood me, I mean, he will not be USING it frequently so the sparks will not be frequent. 

[Stefano]

20 minutes ago, Andy Perrin said:

You misunderstood me, I mean, he will not be USING it frequently so the sparks will not be frequent. 

OK, in that sense. Yes, once in a while it shouldn't bother anyone.

[enricosavazzi]

A few years ago there was some discussion on the Internet of home-made substitutes for very short-duration (one to a few μs) ultrafast electronic flash. For example, see http://www.njnoordhoek.com/?p=735, although slightly different variants were discussed. I started to build one but never got around to complete the circuit, in part because it used a physically large and rather expensive high-voltage capacitor to store a charge that may have been quite enough to kill the operator if something went wrong, and I had some doubts that it could be made sufficiently safe to use in practice.

 

One of the characteristics of this device is that it used a guided arc rather than an open-air arc. Both spark electrodes were made with thick copper wires encircling a thick glass tube or rod, and the spark was guided by the glass surface of the tube, which supposedly made the spark more stable and repeatable. Possibly a comparable guided arc (with electrodes placed closer to each other than the multiple-centimeter-wide gap discussed at the above link) could be used with the simple high-voltage transformer discussed in this thread. Also, a quartz tube would be a logical choice for generating UV.

[Kai]

Very exciting article, Enrico! Many thanks.
I'm afraid my tinkering skills don't reach that far...

I came up with the setup shown here when I was looking for a simple, fast and do-it-yourself metal analysis method. My experiments then led in the direction of atomic emission spectroscopy with spark excitation. Titanium and molybdenum provided particularly line-rich spectra in the range from 300-400 nm. If you combine both, you get the results I have shown.

Here is another example of two 1:2.8 135mm lenses. One is a revuenon special and one is an auto revuenon. Both differ significantly in terms of UV transmittance. This is understandable as Revue was just a trade name and the lenses came from different manufacturers.
The original images also show small focal length differences and by varying the distance setting one get an impression of the extent of the focus shift due to the sharpness of the spectral lines. 

Incidentally, the front lenses of both lenses show a less effective, brownish coating.

[Kai]

@ Andy and @ Stefano:
Well, I don't live in an apartment building with many W-LAN routers. But even in my home, people would react sensitively to disturbances in the WiFi network. The radio emissions are apparently meaningless in practice.
And of course the spark generator only runs for a few seconds. I don't have that many lenses and filters to examine... 

[Doug A]

@Kaithis is super. Your apparatus really shows the UV difference between the two lenses. Well done. The Revuenon Special isn't so special when it comes to UV response.

 

Thanks for sharing,

Doug A