Color Temperature Meter Recommendation 7

[Medeski]

Hello ladies and gents,

I'm not sure if this is the right forum for this, but I was wondering if anyone could provide a recommendation for a color temperature meter. I need one that will provide the color temperature in Kelvin as well as the lux/fc reading. I am actually going to be using this to measure the color temperature of LED lights, not for flash. If anyone has any recommendations and possibly a source where I can purchase one, it would be greatly appreciated.

From my search so far I've found a range from $70 to over $1000 and have been unable to find a summary of which ones are good or bad. I would like one that is fairly accurate (approx. +/-200K), but I think my price range would be around $200 to $400 if that's possible. I have a multimeter too, I know some lux meters plug into those, but I'm not sure if any output color temperature as well. My Google-fu has been exhausted, so if anyone has any recommendations, they would be greatly appreciated.

Best,
Medeski

 

[IRstuff]

Color temperature for LEDs sounds like an oxymoron, particularly since most white LEDs have output spectra that don't in any way resemble even a graybody, much less an actual blackbody.

I would think a spectrometer would be more useful and capable. Some open source work:

http://spectralworkbench.org/sets/show/149

http://store.publiclaboratory.org/products/desktop-spectrometry-kit

http://store.publiclaboratory.org/collections/spectrometry/products/foldable-mini-spectrometer

Not sure how good these things are, but, certainly, they're probably tolerable for defining a "color temperature" for a "white" LED.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[Medeski]

Thanks for the reply. The product are fluorescent replacement LED light tubes. The color temperature range is between 3000K and 6000K. Though the LEDs are all initially white, they are coated with phosphors to produce varying color temperatures. The following has a chart of the color temperature (CCT) scale close to what I want to test. Link

 

[VEBill]

"...the LEDs are all initially white..."

No, that's not exactly how it's done.

http://en.wikipedia.org/wiki/Light-emitting_diode#White_light

 

[IRstuff]

VE1BLL is correct, as usual. It can be readily verified by a simple inspection of a so-called "white" LED. They are invariably covered with a yellowish coating, which is the phosphor, unlike other color LEDs, wherein the actual LED die can be seen:

http://www.ebay.com/itm/20PCS-SMD-S...799?pt=LH_DefaultDomain_0&hash=item3a7aced267

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[Medeski]

Thank you for your reply. Regardless of the initial color of the LED, I still need a color temperature meter that can measure from approximately 3000K to 6000K. So if anyone has a recommendation for a good color temperature meter, it would be greatly appreciated.

 

[IRstuff]

I suggest you read up on what color temperature is. It is the temperature of an ideal blackbody whose output spectrum most accurately represents the output spectrum of another emitter. Therefore, a color temperature meter is a spectrometer that calculates the color temperature. You cannot measure color temperature directly.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[VEBill]

The LED with phosphor emits a combination of wavelengths that stimulate our eyes' cones to replicate the human RGB response to white light of different colour temperatures.

Where measuring a black body radiator would be a simple technical challenge (a couple of readings and a chart, done), measuring an LED's colour temperature is not going to be simple nor cheap. You'd need a synthetic eyeball to figure out the human response.

One could probably get within about 500 degrees just by visually comparing the LED under test to photographs of references.

One could use an adjustable incandescent to transfer the colour temperature, and then employ a cheaper meter that is calibrated for black body radiators.

I can't help but agree with IRstuff that LEDs spectrums are non-trivial.

The other option is to look up the PN of the LED in question. The spec sheet should say.

Give this thread some time in case anyone else can help with a pointer to a cheap LED colour temperature meter.

Good luck.

 

[Medeski]

Thank you for your replies. For instance, one color temperature meter is the Kenko KCM-3100 Professional Digital Color Temperature Meter Link, which can measure ambient or flash lighting. The meter will display the CCT (Correlated Color Temperature) in Kelvin. It also costs over 1000 USD. We have received test reports from the manufacturer stating the CCT for the LED replacement tubes, but we want to be able to test it ourselves. If the only way is to spend 1000 USD, then so be it, but it would be great to get a recommendation from someone that has used one. Thanks again.

 

[IRstuff]

I suggest that you read through:

http://www.dvinfo.net/forum/photon-management/493324-color-meter-weirdness.html

It may give some insight and possibly suggest that using the Kenko will be a nontrivial exercise. Note the erroneous discussion about "measuring color temperature." I will reiterate, color temperature cannot be "measured;" you measure spectra and calculate color temperature. As alluded to by VE1BLL, you can use color corrected detectors to simulate the eyeball, which simplifies the calculation, assuming you are attempting to achieve a human perceived color temperature. Not all eyeballs are the same, nor are all tristimulus filters the same.

If the LED is intended to be used to illuminate for a camera, say, then what you calculate for eyeballed color temperature may be off.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[Medeski]

Thank you for the link, I will give it a read through. We are also looking at getting an integrated sphere measuring device and I've looked at making a hand held spectrometer. Thanks again for both of your comments. I will have to do some more research.

 

[MagicSmoker]

Rather than try to school you in blackbody emitters...

I get this newsletter from Saelig which usually has something interesting in it, hence why I haven't unsubscribed, and today there was an article on a spectrometer using Android phones. I thought "that's clever, but really expensive" and promptly deleted it, but you might find it useful:

http://www.saelig.com/product/TSTEQ...March+Saelig+Newsletter+(A)+&utm_medium=email

 

[MagicSmoker]

Hmmm... it didn't post the link for some reason...

Link

 

[Medeski]

that's clever, but really expensive

Wowza. You ain't kidding. Thank you for the link.

 

[IRstuff]

There's

http://www.ebay.com/itm/USB-Broad-r...787?pt=LH_DefaultDomain_0&hash=item337cda003b

Probably a whole bunch within your apparent budget, including my previous cites.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[SPLatManOz]

Have you tried a colour temperature app for your smart phone? Android has at least one, no doubt that other brand will have them as well.

I wouldn't expect them to be super-accurate, but a good free way of exploring your idea. Mine certainly gives sensible readings on daylight, incandescent, and fluoros.

I don't think it makes much sense to try and define the CT of a monochromatic LED. For a white LED, yes, because it emits broad-spectrum light from a phosphor, much like a fluorescent lamp. Color mixes from RGB LEDs will give a reading, which will be an indication of subjective CT.

Incidentally, there are also magnetometer and gravitometer apps. I found the magnetometer useful for exploring an idea in the lab, relating to sensing nearby magnets, and played with the gravitometer on a cruise ship. "Unfortunately" the weather didn't get rough, so the readout was not very dramatic.

David Stonier-Gibson
[URL unfurl="true"]http://splatco.com[/url]

 

[MacGyverS2000]

One could probably get within about 500 degrees just by visually comparing the LED under test to photographs of references.

I'd bet you could get within a couple hundred degrees without even trying... there's a world of difference in color over 1,000 degrees, so splitting that up into 5-6 bins should be no problem at all.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[djs]

You might try a colorimeter used to calibrate video displays (TVs). An excellent unit is the X-rite i1Display Pro. You can use the freeware Color HCFR software on a PC. These units are used to calibrate the white "color" on video displays based on Kelvin temperature.