Back when I was graduating from the university, I was working on an experiment for dark matter direct detection called CRESST. This experiment uses particle detectors based on scintillating calcium tungstate (CaWO4) crystals. I stilI have a broken crystal as souvenir and always wanted to build a display stand that excites the crystal's fluorescence.
I realize that people probably will not copy this exact build since calcium tungstate crystals are not commercially available and also the UVC LEDs I used are quite expensive. However, it might help you if you are planning to build a display stand for other fluorescent minerals like amber or fluorite.
Step 1: Gather Materials
- fluorescent CaWO4 crystal
- small project box (e.g. conrad.de)
- 278 nm UVC LED (e.g. Crystal IS)
- LED starboard (metal core PCB) (e.g. Lumitronix)
- thermal pad (e.g. Lumitronix)
- heatsink (e.g. Lumitronix)
- step up module (e.g. ebay.de)
- LED boost driver (e.g. ebay.de)
- LiPo battery (e.g. ebay.de)
- slide switch
- 0.82 Ohm 1206 SMD resistor
Fluorescence in calcium tungstate can be excited at wavelengths < 280 nm. This is quite far in the UV and LEDs at this wavelength are usually quite expensive (~150 $/pc). Luckily, I got some 278 nm SMD LEDs for free as they were left over engineering samples from the company I work at. These type of LEDs are usually used for disinfection.
WARNING: UV light can cause cause harm to the eyes and skin. Make sure to have proper protection, e.g. UV goggles.
According to the spec sheet the LEDs have an optical output power of ~25 mW, an operating current of 300 mA and a high forward voltage of ~12 V. Since this means the LEDs dissipate about 3 W of heat they need to be mounted to a proper heat sink. Therefore, I purchased a metal core PCB (starboard) with the right footprint, a thermal pad and a small heat sink. As LEDs can easily be damaged by too high currents they should be operated with a constant current driver. I got a very cheap constant current boost driver board based on the XL6003 IC which also steps up the output voltage. According to the datasheet the output voltage should not be higher than 2x the input voltage. However, since I wanted to power everything from a 3.7 V LiPo battery, I added another step up converter that increases the battery voltage to ~6 V before the LED driver. The output current of the LED driver is set by two SMD resistors connected in parallel on the board. According to the XL6003 datasheet the current is given by I = 0.22 V/Rs. By default there are two 0.68 Ohm resistors connected in parallel which amounts to ~650 mA. In order to lower the current, I had to replace these resistors with an 0.82 Ohm resistor which will give ~270 mA.
Step 2: Mounting the LED
In the next step I soldered the LED onto the starboard. As already metioned it is important to get a PCB with the matching footprint of your LED. Soldering on a metal core PCB can be difficult as the board dissipates the heat rather well. To make soldering easier it is recommended to put the PCB on a hot plate but I also managed to do without. The LED should be coupled to the board with thermal paste. After soldering I attached the starboard to the heatsink using the thermal pad.
Step 3: Connect Electronics
I glued all electronic components to the bottom plate of my enclosure. Note that the heatsink gets quite hot so it is useful to use a glue which can withstand high temperatures. The battery gets connect to the step up module which increases the voltage to about 6 V. The output is then wired to the LED boost driver which is connected to the LED. A slide switch was added after the battery but you might want to do the soldering only after you have mounted the slide switch in the next step.
Step 4: Modify Enclosure
I made some modifications to the encluse using my dremel tool. A slit-shaped hole was put into the top for the LED light to escape. In addition, I put some openings in the side for ventilation. Another hole was made for the slide switch which was fixed with hot glue. I am not very happy with the look of the enclosure as the holes look pretty rough. Luckily most of them are not visible. Next time I will probably make a custom box using a laser cutter.
Step 5: Finished!
After closing up the enclosure the project was finished. The crystal can be placed on the slit at the top and is excited by the LED from below. The fluorescence emission is quite bright. Note that all light is really coming from the crystal as the UVC light is invisible.
The build can certainly be improved in a few ways. First of all the thermal managment of the LED is not great and the heat sink gets quite hot. This is because there is very little ventilation since the heat sink was mounted inside the enclosure. So far I did not dare to run the LED longer than a few minutes. Secondly, I would like to make a nicer enclosure next time using a custom laser cut box made from black acrylic. In addition, a LiPo charger module with microUSB plug can be added so that you will not need to open the box for recharging.
Participated in the
Make it Glow Contest 2018