DIY webcam spectrometer - IR filter removal, spectra measurements 

Of course I tried to measure a cheap laser pointer I've bought from ebay several years ago. But it doesn't work any more :(.

@@DiodeGoneWild That's unfortunate. Probably the emission spectrum of NaCl in a flame would also work well.

Hi Mr. @mrnmrn1. Your comment was perfect. Do you know whether is it possible to buy cheap B/W CCD cameras? I´ve tried a lot, but without success. I tried to convert an OV2640 (Very Cheap Camera Sensor), ridding-off the bayer matrix filter. It worked with low efficiency because front illuminated CMOS sensors are not efficient in UV Region. By the other side, backilluminated CMOS Sensors (that are good in UV region) are very difficult to ridoff the bayerfilter without damagging them. Acctually B/W backilluminated CMOS cameras are usually very very expensive, probably because of no mass production.

Good visually, for visible light. Still guaranteed to absorb certain wavelengths.

Yes I also suggested him a hole-camera, but with a true B/W sensor, and cracked-off CCD package window. This one is not a bare-die sensor, just looks like one. It must have a thin layer of glass or epoxy on top. A true bare-die would not survive the pick-and-place, wave or IR soldering, and flux washing procedures.

Several types of Glass passes a significant protion of UV-A and even some UV-B.

thank you ;)

Thank you for your support! ;)

rock salt would be the easiest, it was used in some old spectrometers. You can melt NaCl and it looks like glass.

I copy my comment for Danyk to you, it might help: Yes, the plastic bag, the polycarbonate of the CD and the lens of the camera definitely has losses at certain wavelenghts. Also the color filter matrix on the image sensor subpixels. Some ideas for further improvements: - first, try to use a pressed CD instead of a CD-R. The greenish dye doesn't help much with accuracy - transmissive mode of the CD would probably be better, because in reflective mode, the light passes the polycarbonate twice - you can try reflective mode with a de-laminated CD, but from the back side. This would eliminate the polycarbonate from the light path entirely (for the price of a significant intensity loss) - try some actual diffraction gratings - don't use any diffusor on the slit - convert the camera to a hole-camera by replacing the lens to a very tiny hole, and find its focal point - use an old B/W CCD CCTV camera with manual exposure. You can even crack the glass window off of the CCD, so there will be just air between the bare CCD die and the light source. I can send you some pictures of a cheap old B/W CCTV camera, which uses a bare CCD die out of the factory. Maybe you can find the same model second-hand locally. Hopefully it's not a color CCD in B/W mode (I haven't checked with a microscope if it has a filter matrix). But this cheap one doesn't have manual exposure, so a higher-end model would be better, with a broken-off CCD window.

Yes, probably argon or other noble gas, maybe a misture of more gases to reduce the ignition voltage. It probably produces UV to excite the green phosphor. I guess there's no mercury, otherwise we would see the 436nm blue.

@@DiodeGoneWild see en.wikipedia.org/wiki/Penning_mixture

Aluminium film removed from a pressed CD is a great idea! I suggested him a similar solution, just to use the de-laminated CD's back side in reflective mode. But that would cause a big drop in intensity. Your idea is better, but it's probably essential for the diffraction grating to be perfectly flat, which might be a bit hard to do with the peeled-off piece of aluminium film. The lens problem can be solved by not using a lens, just a tiny hole instead.

@@mrnmrn1 If you cut a DVD you can split it in 2 layers and get a reflective aluminum grating on one side of the plastic.

7:58 wondering than why this isn't blocked by lense !!

@@omsingharjit probably the lens is made of quartz glass

@@SadamHusajn82 or of acrylic plastic both glass and plastic blocks uv light quit Significantly .

The surface of the parabolic mirror has to be pretty good, or else the sprectral lines will be smushed. But replacing or covering plastics with other materials like quartz and aluminium might improve the setup a bit anyway. Diffraction grating is also pretty cheap on Amazon.

@@lesliespeaker668 yes , that mirror will not be perfect but that's not for Taking image that is just to see if camera can pick that wavelength or not by discarding other distorted wavelength temporary .

The spectrometer works by looking at the light intensity along the horizontal spectrum created by the CD. It maps this intensity to a wavelength based on its position. Without a lense, you just get a blurry image and cant figure out the positions of where light is coming from, so you also cant figure out the wavelengths.

@@Basement-Science Then perhaps a pin hole camera would be good?

@@franzliszt3195 I think this is where the linear ccd sensor comes in.

@@franzliszt3195 I guess so, if the sensitivity was good enough. The slit coming into the spectrometer already blocks a lot of light, and an additional pinhole after that blocks even more.

Naiiise

usually a low pressure sodium lamp is used for callibration

@@jg6780 I don’t men calibration for wavelength peaks. i mean calibration for intensity per wavelength

I think you need a webcam that has a manual mode with fixed exposure to do that. If you're serious about building one careful selection of the webcam and using an actual diffraction grating is probably smart.

@@lambda7652 So just use an incadescent lamp and Planck law formula en.wikipedia.org/wiki/Planck%27s_law For each wavelength scale the sensitivity of point to the value which should be according to this law.

@@jg6780 Yes tarts in principle what i wan to suggest. The questin is can the software do it? and if so do you need to have a calibration for every exposure setting of the cam

I don't have any grow light, I have to get some...

I already have a B/W camera from the flat CRT door phone system, but I have nothing to connect it as a USB webcam. UVC tubes from ebay tend to have no warnings :).

@@DiodeGoneWild can you remove the Bayer filter from webcam ? It probably produces this dips in the red part of the spectrum.

www.theremino.com/wp-content/uploads/files/Theremino_Spectrometer_V3.1.zip

I think in the last video he did

@@MassimoTava Ah you're right he did, right at the end. Zero blue in it, good!

né, vůbec ;)

Big thanks for your support!

@@DiodeGoneWild you have one of the best channels on youtube! :)