building an homemade spectrometer

I'm not sure how you can simply sense the light if your eye can't see it. I have this spectroscope.

Mike

I suspect that since a 'digital camera' was mentioned by the OP, that they want sensitivity in the CCD there for that wavelength range, as well. Probably not using their eyes, directly.

Jon

Not specific to your question, but ocean optics do some low cost spectrometers. There are some tech articles on their site.

At least one of their products is a low cost PC connected spectrometer along the lines of what you describe.

I have built a few cheap, DVD-based spectrophotometers. You can use heavy construction paper and elmer's glue for the box and baffles and a flat razor blade to either cut a slit or else use themselves in pairs to make precision optical slits to let the light in. Also, a shoebox can be used. Or build something cheap from materials you are comfortable with. These work pretty well with digital cameras, too. I've not been interested in 380-410 and I have a sneaky suspicion that the plastic used in the DVD will absorb a lot of it. But if your source is bright enough or your camera sensitive enough, that may not be such a bad problem. The optical lens of the digital camera may also be a problem. You'd have to test to see. But you need to do that, anyway. So if you have a source that you know emits in the

380-410nm band (a 405nm LED?) you could at least try it out and see how well it measures out against the stated curves for it. You will definitely see the problem, if there is one.

You can test with DVD-RW, DVD-R or DVD+R. I know for certain that the DVD-R has a terrible, nasty absorption notch in the red (they include a dye that absorbs there) and that the DVD-RW seems to not have that problem. But you are on the weakly visible side of blue, so I suppose you don't care about that, anyway.

I think Edmund Scientific carries (well, it used to be cheap) some diffraction gratings, too. You might consider those.

What exactly are you trying to do?

Jon

Thanks a lot, Do you have an idea about price of the spectrometer you mentioned ? GB

I'm just trying to build an Uv source lightmeter. Thanks for yours ideas. GB

It was 10 yrs ago. The lab I was working in had one. I remember it being

Well, you could use a CD-R to reflect the light into a spectrum...

Why don't you use a UV filter glass then?

Regards, Martin Brown

Yup that and a photodiode should work fine.

George h.

Not really gonna work; the blue and UV sensitivity of silicon photodiodes is poor. A day-glow painted screen to shift the UV light to red, followed by a photodiode, would be OK.

The only good sensor for wideband is the humble bolometer (or thermopile); you can make a black item and look at it heat up easier than you can sense IR+visible+UV by other methods. My favorite variant is photoacoustic spectroscopy, actually: you let the heating and expansion create an acoustic signal and then pick it up with a microphone.

Have you considered using a diffraction grating?

You'd need some kind of mechanical thing to zero in on the wavelength of interest, unless you have a mongo light sensor that's sensitive to UV.

Good Luck! Rich

A CD-R R/O is slightly cheaper...

A CD-R R/O disk makes for a slightly less expensive reflection grating..