What do I need to be able to display light spectrums?

Started Feb 3, 2014 | Discussions thread
xpatUSA
xpatUSA Forum Pro • Posts: 17,298
Re: What do I need to be able to display light spectrums?

Wayne Larmon wrote:

I want to measure the characteristics of light sources. Spectrum plots, CRI, color temperature, etc.

I have already made a simple spectroscope with a slit made from box cutter blades and a piece of diffraction grating but that only shows me how spiky a light source is. It doesn't show me spectrum shape, color temperature, or CRI.

This probably doesn't need mentioning, but I am not a scientist nor an engineer. So talk slowly when you reply, please.

I use the cheap approach, have a look at my post on LL here:

http://www.luminous-landscape.com/forum/index.php?topic=76990.msg616112;topicseen#msg616112

It is fun, but very hard to beat the manufacturers published data which is usually expressed with relative power on the X-axis. Your model of camera has a large impact. The camera incorporates filtration to take out UV and IR which is OK provided that visible wavelengths are no attenuated; not always the case. The next hurdle is the raw converter and it's transformation of the data from raw via XYZ, obfuscated by WB, and finally to the color space of your choice.

What I'm saying here is that the pretty rainbow-colored spectrum that you see in your image is probably not what the sensor saw.

I have a Sigma DSLR from which I have removed the hot mirror and intend to use it in future to capture lamp spectral data. I also have several utilities that can take raw data and present it as an image with no other processing or just as a simply-rendered RGB with no processing other than conversion.

Foe spectrum shape, ImageJ is very useful as it allows you to draw a line through a spectral image and give you a graph of pixel luminance. As to calibration, the spikes in most lamps are due to elements such as mercury or phosphates and their wavelengths are all available on-line.

As to CRI, it can be estimated once the wavelengths of any peaks are know but the original parameters are band-based and manufacturers can shift those peaks around by phosphor selection or doping alteration until they get better CRI numbers (not that I'm suggesting that they do, and I'm really proud of my CRI 95 Philips overhead fluorescent tube!).

As to the color temperature, obfuscated as it is by 'tint', I'm less able to offer anything but we've plenty of knowledgeable folks here that could.

-- hide signature --

"Engage Brain before operating Keyboard!"
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-G1 Sigma SD15 Panasonic Lumix DMC-GH1 Panasonic Lumix DC-G9 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS +12 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow