Are there any inexpensive spectrometers?

[LightningInAJar]

TL;DR Summary

Inexpensive spectrometers?

I was wondering the cheapest way to measure all the wavelengths from NIR to UVC that comes to earth. Both varieties and intensities. I am curious how sunsets and sunrises appear different and how things change midday throughout the year.

 

[Drakkith]

Visual spectroscopes are available for very cheap but don't allow for much in the way of measurement. I have two in my closet that I occasionally pull out to look at various lights, the Sun, etc.

A quick look at amazon didn't show me any spectrometers for less than about 200 USD, though you might be able to find one if you spend a little more time looking than I did.

If you're a DIY kind of person you could try making one yourself with a diffraction grating, a cheap webcam, and some cardboard. You could probably make one for less than 20 USD.

 

[sophiecentaur]

Google amateur astro spectroscopy. It's a popular field for people who already have a scope and a dslr.

 

[LightningInAJar]

Google amateur astro spectroscopy. It's a popular field for people who already have a scope and a dslr.

I don't have a telescope.

 

[Drakkith]

I don't have a telescope.

You don't need one. Anything with a focusing lens can be used, such as a webcam. You can get new ones for under 25 USD and I expect that if you look hard enough you can get a used one for virtually nothing.

 

[LightningInAJar]

Webcams can't detect ir or uv?

 

[Drakkith]

Webcams can't detect ir or uv?

Some can. There are plenty of cheap cameras that can detect IR since night vision cameras are very widespread. UV is not as prevalent, but there isn't as much UV light as there is visible and IR anyways.

 

[sophiecentaur]

There are plenty of cheap cameras that can detect IR

Apparently the filter on the front of DSLR camera sensors can be removed fairly easily. (But not replaced that easily.)
Mechanically, it may be easier to use a DSLR lens (a relatively long focal length) to focus an image on a webcam sensor. This will give a well dispersed spectrum, compared with the usual wide angle of a webcam lens. (This is along the lines of imaging with a telescope.)

 

[Godot_]

If you want linearity across the spectrum, cheap camera photosensors* really aren't that good. Then you might try to find some old, used lab photometer. You might even get one of those for scrap value, as they are often discarded when the (expensive) lamps break down.

But if you're just doing that out of curiosity / as a hobby, the tips above will probably do nicely.

Beware: Normal glass optics and mirrors might absorb quite some UV light. For measurements at 280/340/360nm, which are a daily thing in the MolBiol lab, cuvettes made from special quartz glass or special plastics are used, which don't have that issue. So if you're seriously interested in UVC, as noted above, this _will_ become an issue. In the near IR, this should be not so relevant, but eventually, glass will also absorb that.* I remember having photographed (ca. 2005) some wee flowers, orange with white specs in the centre, to illustrate something else for a talk: Most had five petals, but some had only four... When I gave that talk and asked whether someone saw something strange, the first answer was: "They're purple in the centre..." And yes, they were - I had failed to notice when prepping my slides.

Fortunately I realized why, and could answer that: The sensors of my mid-price pocket digicam are sensitive to near UV, and optics with low chromatic aberration should be pretty permissive to near UV, too. And as bees and other insects have a visual range up to 300-ish nanometers... ...they do see that as a "yummy nectar here!"-sign, which we just can't appreciate.

 

[Drakkith]

If you want linearity across the spectrum, cheap camera photosensors* really aren't that good. Then you might try to find some old, used lab photometer.

What do you mean by linearity? Equal response across all wavelengths?

 

[Godot_]

What do you mean by linearity? Equal response across all wavelengths?

Yep.

After all, you don't / the OP doesn't have a reference light to calibrate the sensor. So weaknesses in certain wavelengths a) hence can't be compensated without some calibration-based compensation, b) might not be obvious, and c) might thus lead to wrong conclusions.

 

[LightningInAJar]

It would be nice to have a digital meter that gives me an intensity measure of a range of wavelengths. I'm not looking to take pictures of the sky. Just know the wavelength ratios throughout the day and year. I would like to correlate light levels of different kinds to mood and physical activity of people based on self reporting.

 

[ohwilleke]

Summary:: Inexpensive spectrometers?

I was wondering the cheapest way to measure all the wavelengths from NIR to UVC that comes to earth. Both varieties and intensities. I am curious how sunsets and sunrises appear different and how things change midday throughout the year.

It depends on what you consider inexpensive.

I have a friend who has a private sector business with a scientific laboratory that buys used equipment from universities and government research labs for his own lab when the state of the art university and government labs replace their equipment. Periodically, these institutions hold low key auctions to sell their outdated surplus equipment.

He typically pays 60-80% off of retail, for equipment that is typically 10-20 years behind the state of the art equipment, but is entirely adequate for his purposes.

On the other hand, it certainly isn't bottom dollar for minimally sufficient DIY gear that you personally piece together from parts.

Another place to look is Ali-Aba which is basically a business to business e-commerce portal. This often allows you to get wholesale prices for goods rarely sold to retail consumers in brick and mortar storefronts from places like China for goods shipped to you. Sometimes, you need to buy in bulk, but not always. I have inventor clients that mostly source their materials and equipment that way.

 

[sophiecentaur]

It’s a question of whether the OP wants a fun DIY project or to buy something and start getting results tomorrow.

 

[Godot_]

[I want to] Just know the wavelength ratios throughout the day and year. I would like to correlate light levels of different kinds to mood and physical activity of people based on self reporting.

This starts to sound quite a bit like a psychology / psychophysiology science project...

...you might or might not be aware that this topic has been researched to quite some extent. And that phototherapy is a first line therapy for "winter depression", and also is used with several other forms of depression.

So I did a quick PubMed search and found several research papers relating to that, and looked through the PMC hit list. Some of these are open access, some arent, but are publications from journals that might or might not be accessible in your local university library. As the 17-hit list filtered from 93 PubMed Central hits might be too long, so I recommend you do the search yourself again:

Go to the NCBI website, and search for "mood light spectrum psychology", choose the 93 hist on PubMedCentral, and tick all the hits where the brief abstract doesn't contain "autistic / bipolar / schizophrenic spectrum". You might tick more than 17 papers, but then you'll still get an overview of what's been already researched there.

https://www.ncbi.nlm.nih.gov/search/all/?term=mood light spectrum psychology
should immediately take you to the search result page...

HTH

P.S.: If you're really doing this for a university research program, you might just ask for a suitable budget to accommodate the price for a ready-to-go spectroscope. These should also give you the proper wavelength in addition to the intensity, which a DIY thingie won't...

You might need to attach a fisheye lens to that, so keep that UV absorption in mind. UV-transparent lenses might not be all that cheap, too...

 

[jim mcnamara]

I think the OP may want to try color temperature. To start off.

Color temperature is a way to describe the light appearance provided by a light bulb. It is measured in degrees of Kelvin (K) on a scale from 1,000 to 10,000. Typically, Kelvin temperatures for commercial and residential lighting applications fall somewhere on a scale from 2000K to 6500K.

This works outside as well, example so-called "North light" measures at about 7000.
Example color temperature phone app - I think it is free, which certainly matches "cheap":
https://play.google.com/store/apps/details?id=com.bti.lightMeter&hl=en_US&gl=US

There are also hand held meters. Not free. This only works AFAIK mostly in the visible range of light. But it is a start.

Also note: some types of 'energy efficient' glass block NIR (near infrared) to increase the R value of a window, for example.

You know, simple questions get simple answers sometimes. If you told us what your ultimate goal was
I bet you will get more good answers, because right now we are all poking around in the dark, so to speak.
Guessing.

 

[ohwilleke]

It’s a question of whether the OP wants a fun DIY project or to buy something and start getting results tomorrow.

Totally agree.

 

[LightningInAJar]

I am more interested in natural light and what wavelengths are reaching Earth at different times in my local area. Maybe display the data in an art project or compare against human mood throughout year. I just want accurate data and from there I hope to have better clarity on how to use it, but I have a general idea.

 

[Godot_]

Well, you might want to take a look here for sunlight wavelengts, and also here for technical specs of glass. Don't worry, it's only some curves. Which show you that UVC really doesn't play a role...

That raises your shortest relevant wavelength to >250nm... ...quite a bit of which, however, is still absorbed by standard bulk glass. And makes the standard digicam CMOS sensors somewhat adequate (they still have weaknesses in the UV range). So, apart from the optics, you'll get away cheaply.

Spectroscope building kits are available from 5€-ish, like this DIY cardboard kit. Take a look at the third image in the first half's link to see what you'd get. Or for a bit more (<20€) you can get e.g. this handheld hobby spectroscope, which, again, will probably only work the visual range. Also, they only give you a spectrum, no quantitative measurement, nor full sky view.

You can pimp that up for some with an affordable ocular cam like this one here. The linked one has a built-in UV/IR filter. Getting to the data for those parts of the spectrum, though, might hence be possible, by just removing the filters. This setup would be semi-quantitative... ...but you might have to experiment quite a bit with how this is displayed, by adjusting the distance between your diffraction grate and the cam. Maybe even find some computer savvy person to read those data. Also, you probably won't get any precise wavelength for your measurements.

But... ...if you're into an art project, this might even be better. Colorful spectra should be quite the eye-catcher!

So the question you'd have to answer is: Can you live with a qualitative display - od do you really need to quantify your light intensities?

Keep in mind that much of the light absorption due to cloud cover is in the IR range (which you can see in the very first link in this post). And there actually are some Bio/Med/Psych papers proposing / finding that this really does influence people's mood...

Also be aware that the sunshine and exposure duration are relevant for mood, too. So the daylight & sunshine hours, and the time your "test subjects" spend outdoors will have to be figured into your analysis, too.

Disclaimer:
The links are not intended to advertise specific products, by a specific manufacturer, or via a specific retailer! They're merely meant as examples to illustrate what the market does offer.

Should that violate the forum rules, I'll immediately remove them.