I’ve had my eye on building a shortwave UV display light for a few years.
When my wife started collecting samples that included fluorescent minerals we didn’t have any way to activate them. When she was in school she was able to use on-purpose geological field lights in a school laboratory setting. About that time decent filtered longwave (UVA) flashlights were starting to become generally available. She got an Alonefire flashlight https://www.amazon.com/Alonefire-Flashlight-Portable-Rechargeable-Blacklight/dp/B07SWW5FHB that had a true 365nm LED chip and almost as importantly it had a visible light filter. Unfiltered UV lights have a lot of visible light that washes out fluorescence.
Shortwave lights or UVC have a wavelength of around 250nm. There is just the beginnings of being able to produce this wavelength of light from LED sources but by far the most common source of higher power UVC is mercury vapor bulbs. These are basically fluorescent bulbs, but with no internal phosphor to translate the UV into visible light.
There are some commercial (and expensive) medical devices that both produce UV and filter the visible light. These pieces of equipment are called Transilluminators and they are used to view stains of DNA testing. The UV light lights up different DNA components for analysis. These devices mostly use UVB tubes but more importantly they have a special glass filter that passes UV but blocks most visible light. The gold standard for this glass was Hoya 325C https://hoyaoptics.com/colored-glass-filters/ultraviolet-transmitting-visible-absorbing-filters-u/. The problem is they stopped making it, at least for a while.
I managed to find a surplus transilluminator and snapped it up. It was a Fotodyne Foto/UV 26 3-3035 for which I could not find a single specification for online. I think I paid $75 plus another 60+ dollars in shipping for it a couple of years ago just to get that glass.
At first I tried replacing the 8w midwave (UVB) bulbs that came with it with shortwave UVC bulbs. The results for the specimens we tried were… underwhelming.
I had read about transillumination conversion on a couple of sites. Converting a Transilluminator into a Short Wave Lamp (naturesrainbows.com) Building Your Own Transilluminators – DIY – MinerShop. These gave me the idea on doing this plus the vector on the transilluminator in the first place.
One of the articles mentioned using higher powered bulbs and replacing the matching “ballast” to get a higher power. That certainly seemed to be in my skill range. I did the research to find the bulbs, worked backwards to find the appropriate ballast (and that was a lot more complicated than I expected) and I ordered bulbs, ballast, and socket. When I got them I realized that for all of the electrical engineering I had done, I failed at some basic mechanical engineering. I forgot to check the length of the bulbs I had ordered against the length of my transilluminator. The model I had was significantly different from the one converted in the article.
I didn’t want to give up the power that I had, but the glass was a fixed size and the bulbs were longer. In the end I decided to create my own fixture. I priced some sheet aluminum and looked at different project boxes and cases and there just wasn’t much selection in the length I needed.
I can’t remember where the idea to use a sheet pan came from. At one point I was using one of my wife’s pans to mock up a layout but I don’t remember when or how the idea to form the fixture out of cookie sheets came from.
So the physical plan was to mount the tubes to the back side of a cookie sheet pan https://www.samsclub.com/p/members-mark-aluminum-baking-sheet-pans/132731 . I used 40 mm spacers between them and used the other of the pair of cookie sheets as the reflector and as the mount for the ballast.
I used old PC component mounting screws from the days of CD drives to connect the 40mm spacers. I already had 6-32 nylon insert nuts to secure all of the sockets and clips, but surprisingly I had to order 6-32 x 3/8 screws since the local stores only had 1/2 inch screws. I needed to make sure I kept clear of the bulbs for the top section.
I laid out the bulbs and used a piece of purple construction paper to simulate the size of the Hoya glass. Cutting the opening for the glass was an exercise in “how not to do this”. I initially used my grinder with a diamond wheel. That was a mistake. The diamond doesn’t cut enough and I really ended up mostly melting lines in the aluminum, not to mention having molten aluminum stuck on my diamond wheel. A jig saw with a fine blade was the proper tool and it made quick work of the remaining parts of the cut. Cleaning up the rest of the cut where I used the grinder was much harder. Mostly I used a metal file to trim the parts that went off in the “z” direction.
The clips were installed and the lights test fitted. Then the nerve wracking part. I carefully sited the glass over the hole (which had been cut with a 1/4 inch boundary inside the glass dimension) and then glued the glass down with black silicone seal.
The ballast (which is longer than the light bulbs) was mounted to the top of the reflector panel. I had to turn it diagonal in order to get screws into the mounting ears of the ballast. All AC electrical transitions are inside of a box. At the moment this is a plug in cord, but that will likely change as the installation completes and some lighting control is added. The metal structure of the fixture is grounded as well.
Those bulbs are going to get pretty warm. That’s 4-60w bulbs in there so 360w is a non-trivial heat load. I will have two fans blowing down the length of the bulbs between the pans. The outside is sealed with aluminum duct tape and the light trap and ducting of the vented ends is TBD.
The contraption was sited on the designated shelf and a hole about an inch larger than the glass was cut in the shelf. The particle board is photoreactive with the light so I used the black silicone to blacken that light color in the middle of the shelf.
I don’t have any photos of the testing since we were doing it in the minimum amount of time. We were without protective front glass and only use protective UV safety goggles. We wanted to minimize the exposure to the UV and also the cooling fans weren’t in yet.
Leave a Reply