I have read that one can make a wavelength calibration source from a flourescent lamp starter, but I don't know how to do it. I tried connecting one directly to 120V AC house current--it glowed momentarily and then popped (burned out). Does anyone have plans for building such a thing?
You need to include a series resistor to limit the current. It was the Swiss amateur Richard Walker (The author of the "Spectral Atlas for Amateur Astronomers" ) who discovered some of these starters could be used as calibration lamps. He has a writeup on it here.
(Note different starters have different gasses. The ones used in Shelyak's instruments are RELCO 480. which contain mainly Ne and Ar. They use a special constant current power supply derived from 12v which prolongs the life compared with a simple mains connection.
They are still avaialble here in europe eg
though US suppliers probably don't stock them as they are 240v rated. Not sure if there are any 120v equivalents. You can also buy them as spares for their spectrographs from Shelyak instruments
I am part of a group that has been 3D printiong a spectrograph and have had some luck with DIY cal lamps. If you would contact me privately I will send along what we have found.
You need a line-rich spectra for calibration. There are 2 inexpensive options:
I use a germicidal lamp (low pressure Hg) which has great lines from 250 - 1013nm. (see UVTools 4W M12 version or alternate sources for $20-30, and you can remove the cover and get the full coverage of lines.It is made for fluorescence and the cover over the bulb eliminates the visible light) Unit is self contained, uses 4 AA batteries and just turn on and use.
Neon Glow Lamps are rich between 500 - 800 and then a great line at 1176nm. It can be bought off of Amazon (CEC Industries C2A Neon Indicator Bulbs, 105 to 125 V, 0.25 W, Wire Terminal Base, T-2 shape (Box of 10)) for $6 for 10. All you need is a 40-50kohm resistor in line and connect to 110V (double resistor for 220V). Produces a bright light with great lines. So very simple solder one wire of 110V wire to a 50Kohm resistor, solder resistor wire to lamp, solder other lead of lamp to 110V wire.
110V--- 50Kohm -- Ne bulb -- other 110V lead
CAUTION/DANGER -- BE CAREFUL OF EXPOSED 110V WIRES! DEADLY!!! (Just need to say this!!!)
Combined, you get lots of lines. I usually use about 13 lines that are greatly resolved from 250 to 1176nm.
Hope this helps
If working at low resolution even the Balmer lines in a hot star can be an affective means of calibrating spectra but beyond that the Ne/Ar Relco starters are probably the most common solution and have the advantage of compactness so they can be built in. They are used in commerical instruments such as Shelyak LISA,ALPY and LHIRES and Starlight Xpress for example. I have even run them off something as simple as the battery powered flash circuit from a disposable camera but for long term reliability and safety you really need to run them off a high voltage constant current controlled supply though .
Many neon lamps also have some traces of Argon though and if you expose long enough you can see these and weak Ne lines too which give coverage in regions where strong Neon lines are lacking. There are some examples of other lines identified from a Neon lamp (as orginally fitted to my LHIRES, though that now has a Relco Ne Ar lamp) here on my website