I'd like to be able to determine the intensity of UV radiation from different sources like CCFLs, the sun under various conditions, UV torchlight, etc. The sensor doesn't have to be very linear, only approximately so. It's not necessary to measure the absolute intensity either - knowing the relative values is enough.
Will a UV LED work? I don't have any with me, so I can't just try it out.
Probably but you might need a piece of Wood's glass from a black light in front of it so that only the UV can get to the sensor.
Otherwise blue and violet light will also generate photoelectrons.
-- Regards, Martin Brown
Ah yes, there's that. Even if I manage to get Wood's glass - which I doubt - I wonder how the IR component of sunlight will affect a UV diode.
use a bit of exposed photographic film.
IIRC the incoming radiation needs to be higher frequency than the LED is designed to ouput, so you'd use blue to sense UV but not RGB. I'm not certain on that point though.
ISTR someone using a combination of 2 permanent marker pens to make a uv filter. Might find it on youtube.
...All these bits of info need checking.
NT
Could using a phosphor and a visible light detector work?
-- Cheers Clive
On a sunny day (Mon, 3 Aug 2020 15:15:43 +0530) it happened Pimpom wrote in :
There are several substances that light up bright blue in visible light when exposed to UV:
You might (and probably should) be interested in what sort (wavelength) of UV you are measuring as well. For example, a wavelength of 350nm is UV, and is good for curing some glues or exposing the photoresist on a PCB, but it is fairly useless for killing germs which requires more like
250nm wavelength. So, without giving more information about the purpose of your measurement, it is possible that the suggestions you receive will be correct but not really useful and perhaps more misleading than no measurement at all, of for example it leads you to believe that something has been sterilised and in fact it has not.
Hi pimpon, IDK. You somehow have to select what wavelength 'band' you want to see. This could be with some UV filters, or spectrometer and a broad band detector (photodiode) or pick a detector with some wavelength response that you can use.
I think of Forest Mimms for led's as detecotrs...
George H.
You and others have brought up the question of which UV band I'm interested in. I have no specific application in mind at the moment. I just wanted to have a general idea of how UV levels differ between sources and, in the case of the sun, with the weather and its position in the sky.
If I had to choose filter and sensor types for wavelength, I'd start with near UV in the mid-300s nm.
My problem, as usual, is that there's little chance of getting the filter materials mentioned by people on the internet, especially with the pandemic disrupting everything. To make things worse, with the tension between India and China, I don't know if I'll still be able to order from AliExpress.
Whereabouts are you? There should be a fully calibrated UV photometer at a local or national observatory. I used to follow the one in Brussels.
Start from here and you can probably find one near you:
You just need to find a scrap black light that no longer works and carefully fracture the outer envelope in a stout bag behind a plexiglass screen. It isn't as dangerous as imploding a TV tube but depending on the size of the bulb it can be interesting. Mine was a 6" diameter.
Schott glass or Edmund optics will sell you a real UV filter for a price. (probably about $50) UG3 or UG11 (much more expensive)
Their much "Improved" website is infuriating to use :(
-- Regards, Martin Brown
On 2020-08-03 19:03, Martin Brown wrote: [...]
I see what you mean. Why do they do that? To me it means "Get out of here, fast."
I erase all cookies systematically every session, anyway.
Jeroen Belleman
Thanks for your interest. But I live in one of the most isolated regions of a third-world country and all those sources are out of reach for me. Besides, this is something I want to try out once and then shelve indefinitely, so I want to spend as little as possible on it.
Well, this is still colorimetry; the separation of a bit of spectrum is the first part, and detection is second.
Either a prism (quartz passes UV) or grating, with some kind of beam-forming (maybe just a tube with interior painted black) will do light separation, and a sensor can be either a bolometer or (remember, vosible light has been filtered) a phosphor that is excited by all the UV range of interest. The phosphor can be detected with regular-old visible light ICs intended for photography and such.
Expect, if you select a small part of spectrum, dim light and noisy data.
A bolometer consisting of a quartz fiber suspending a thermocouple loop, in the focus of a mirror, will deflect in a magnetic field; this was a classic method of a century ago, and could detect a cahdle light at a mile distance... electromechanical solutions and mirrors are not to be despised. That's a bit less practical than finding a UV diode that works in photoelectric sense, but it's such an elegant experimental arrangement!
A light chopper (logic controlled shutter, or a small fan with some phase-detection) would combine with the right kind of data logging to make a good small-signal amplifier, requiring only a bit of FFT masssaging to pull small signals out of noise.
Look at Sparkfun.com or Adafruit.com for UV sensors, They are also stocked on Digi-key or Mouser. About $6.50 US. for a tiny PCB. Maybe something similar in your country.
On a sunny day (Mon, 3 Aug 2020 13:52:30 -0700 (PDT)) it happened whit3rd wrote in :
Indeed, wonder how far into the UV a normal prism of an old pair of binoculars will go, by rotating it you can make a nice spectrometer.
Would the prism in binoculars be made from dispersive glass?
John
Yes. The way that the prisms are used for near perpendicular entry and exit of the light rays and a total internal reflection there is very little dispersive effect at all. UV doesn't go through glass very well.
The lenses of binoculars need to be achromatic doublets. There is no such thing as non-dispersive glass. Only clear glasses with different dispersions such that a well chosen pair can make a workable achromat (matched focus for two wavelenghts) or apochromat (three).
One of the exotic solutions to the problem uses single crystal calcium fluorite as an optical element which is brittle as hell to work with.
The OP might stand a chance with the photoelectric effect from a cerium metal "flint" sold for outdoorsmen to strike energetic sparks onto tinder. That has a metallic workfunction of about 3eV which is ~UV photon range. Might be worth a look in the SciAm Amateur Scientist archives in case it has been done before.
I think the same sort of flint material is used in some cheap disposable lighters but I have never tested it. Most rare earth metals are in the same ballpark for WF so any of them ought to work.
-- Regards, Martin Brown
Or through optical cement.
Flints are usually mischmetal IIUC. Interesting idea though--might work if he can get the surface clean enough.
One possibly useful fact: free electrons are stable in nitrogen gas, so you can collect them easily with a grid.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
There are flame sensor tubes, and the poor man's version is an NE-2 neon bulb. I've seen that published someplace but can't lay my hands on it at the moment.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
I designed some flame detectors, to check for flame-out on steamship boilers. A flame-out can have serious consequences if you don't shut off the fuel pump pretty soon.
The sensor was a golf-ball sized quartz tube. It ran at basically line voltage. It was blind to full sunlight but would fire from a match held a couple of feet away.
Might have been Hamamatsu.
The ships are mostly diesel now. Steam plants are efficient but too complex.
-- John Larkin Highland Technology, Inc Science teaches us to doubt. Claude Bernard
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.