Open Challenge to Lumens Testing..................

Open Challenge to Lumens Testing

This is sort of Open Challenge to Lumens Testing that is done on LED Lights.

I googled and found information that Lumens are only useful if 2 lights that are being compared have SAME OVERALL SPECTRUM of light.

When we compare LED's, even the Different Tints in Same Bin have Varying Color Spectrum….. leave alone the LED's by Different Manufacturers….

Also to note that More of Yellow will give More Lumens (in most of the Integrating Spheres)

According to My Theory, A True Lumen Rating will be an Integration of Light Intensities measured in Each of the Spectrum Wavelength.

And Yet, Lumens from 2 Different Tint's can be Equal i.e. one putting more Yellow and one Putting more Blue because of Individual Wavelengths having Higher Intensities. The Light with more Blue will be Bright in Perception!!! No matter what.

So, It wont be proper to say that X light has 10 Lumens and Y Light has 9 Lumens, the Light with 9 Lumens will be Brighter! (because of more Blue)

You've raised an interesting point, but most of us don't dive that deep into the specifics of light testing. I think, for the most part, (ie, generally speaking) our flashlights are using perhaps only 10 or so different light sources for leds and maybe twice that for incandescant lights. As long as the same testing methods and light meter are used to compare A to B then most of the time we're getting the same light source compared the same light source. One Lux3 flashlight compared to another Lux3 flashlight is a relatively (and I use that term loosely) level playing field.

Re: Open Challenge to Lumens Testing..............

[ QUOTE ]
ViReN said:

And Yet, Lumens from 2 Different Tint's can be Equal i.e. one putting more Yellow and one Putting more Blue because of Individual Wavelengths having Higher Intensities. The Light with more Blue will be Bright in Perception!!! No matter what.

So, It wont be proper to say that X light has 10 Lumens and Y Light has 9 Lumens, the Light with 9 Lumens will be Brighter! (because of more Blue)

[/ QUOTE ]

But ,IMHO , the more yellow LED will provide better visibility of what is illuminated so they even out. Another interesting point that your post brings up is "What are we seeing when we see something 'Bright'? One use of a flashlight is to disrupt the vision of someone while another use is to enhance the visibility of an object or area. I'd imagine light frequency is relevent to the application.

We're lucky enough to get ratings of flashlights that were actually made in integrating spheres as it is, it is unlikely that frequency specific lumen ratings will be available for flashlights for a loooong time.

Re: Open Challenge to Lumens Testing..............

ViReN has a good point because spectrum balance does affect metered readings and explains why they don't always correlate with what our eyes perceive. I use a solar cell to make base line readings, check output at various battery levels and just to have fun playing around "testing". The solar cells I use definitely are more sensitive to the yellow part of the visible spectrum.

My light output readings loosely follow Quickbeam's charts so I suspect this might explain why some of my incans show unexpectedly higher output than (to my eyes) equally bright LEDs.

Re: Open Challenge to Lumens Testing..............

A Lumens measurements is not a measurement of total radiated power in the visible spectrum, rather it gives more weight to light in the green part of the spectrum than the others. This is where the human eye is most sensitive at low levels of light, I presume the retinal rods. Vision with your rods is not in color, so at the limit of seeing, a green light would not appear to be green at all, and a green light of the same radiated power as a blue light would 'throw' farther at the lower limits of human light perception. This is why the Lumen measurement favors green.

At higher light intensities, the eye is more sensitive to light in the yellow part of the spectrum. Most of us use our lights such that we can see well enough to see in color, so the lumen standard may not accurately reflect what we see as brightness under typical lighting levels.

I had a U2 that was quite green, and had a throw of 2100 lux, which was pretty bright as these light go (I think Lux measurements use the same weighting as Lumens). When Surefire 'fixed' my greenie U2, they put a head on it with a much whiter LED. To my eyes, the whiter one appeared brighter and far more pleasing than the greener one, even though it measured only 1600 Lux in throw. So I agree to an extent with ViReN, but I'm not sure myself if his yellowish vs. bluish theory would hold for everyone - but it certainly held for me in the greenish vs. bluish comparison.

So the Lumens/Lux measurements are an imperfect thing when dealing with human perception and LEDs of different tints. It might be much more useful to we flashaholics to have a Lumens/Lux standard that was weighted to more accurately reflect brightness at the levels we typically expect from our flashlights. Don't bet on seeing it, though!

Re: Open Challenge to Lumens Testing..............

/ubbthreads/images/graemlins/smile.gif lets see what future holds /ubbthreads/images/graemlins/smile.gif

We should see More Just Light Ratings.. instead of Just Lumens or Lux ....

Another Good Idea is to use Lux Reading at 1 Meter (Without use of Optic) as a Standard... (i.e. not even the 5 MM LED's Optic) .... Emitter Placed Perpendicular to the Lux Meter's Sensor

Re: Open Challenge to Lumens Testing..............

Remember that typical Lux meter sensors are not perfectly corrected to the human eye response. Thats why you see correction factors for different light sources, since their power is different parts of the spectrum. On some of these meters, there is up to a 40% spread in the correction factors for different light sources.

A solar cell is even worse, as it does not even come close match the human eye response.

A typical solar cell has the response of what is listed as silicon in this chart:




You notice how the silicon sensor (solar cell) and even the filtered sensor have alot more response than the human eye in the 600-700nm range. This is where an incandescent bulb is way over weighted for it's output (compared to the human eye), since there is alot more spectral power in the 600nm and longer wavelengths.

Here is the spectal plot for an incandescent bulb:



A better instrument for getting more exact measurements of actual lux is a spectroradiometer:
http://kmpi.konicaminolta.us/eprise/main/kmpi/content/ISD/ISD_product_pages/CS_1000?mDetail=Introduction

It actually measures the spectral power at each wavelength and produces a much more accurate result for Lux. I've seen side by side tests where some Lux meters can be a factor of 50% off, depending on the spectral output of the light source.

If you get a little further into color science, you will learn that there are actually different response curves for the human eye, depending on if the lit object covers a 2 degree or 10 degree area.

A person does have to be careful with simple sensors when measuring light output, as I have seen folks use a CdS (photoresistor) for a few runtime charts here on CPF. The problem with this is that the sensor has a logrithmic type response, and it make the runtime plot look extra flat.

Re: Open Challenge to Lumens Testing..............

[ QUOTE ]
A better instrument for getting more exact measurements of actual lux is a spectroradiometer:
http://kmpi.konicaminolta.us/eprise/main/kmpi/content/ISD/ISD_product_pages/CS_1000?mDetail=Introduction

It actually measures the spectral power at each wavelength and produces a much more accurate result for Lux. I've seen side by side tests where some Lux meters can be a factor of 50% off, depending on the spectral output of the light source.

[/ QUOTE ]

Wow.. Thankfully such instruments do exists /ubbthreads/images/graemlins/smile.gif ... does some thing like Spectro-Radio-IntegratoMeter Exists ?? I mean an Integrating Sphere with an Ability to Record Intencity over a Range of Wavelengths ... /ubbthreads/images/graemlins/wink.gif that would be a True Lumen Measurement /ubbthreads/images/graemlins/smile.gif

But Even then, It wont be sufficient because again.. that meter can be fooled like lets say we have a light which emits very high intencities of Green or Blue or Infra Red Region.... that will make the True Lumens Reading Higher after Integration of All the intencities in All the Direcitons throughout the wavelength spectrum.

NewBie, You are a Walking Database when it comes to QA, QC & QS in feilds of Engineering & Light Sciences ... I think you might have some ideas on what could make a Best Means of Measuring Light Intencity & Amount ... that will approximate Human Eye's Perception....

Or is it still the like the old way....

"Wine is Best .. when tested by Testing Experts" (donno what exact proverb is) ... but it meant ... that We require Special TasteBuds in our tounge to Appriciate the taste of Wine /ubbthreads/images/graemlins/smile.gif

So, We need "Experianced" Eyes ???

Re: Open Challenge to Lumens Testing..............

Spheres do have different types of senor for different types of light source. It's important to use the right measurement equipment and know what you're doing.

Al

Re: Open Challenge to Lumens Testing..............

[ QUOTE ]
Size15's said:
Spheres do have different types of senor for different types of light source. It's important to use the right measurement equipment and know what you're doing.

Al

[/ QUOTE ]

kool... do they measure Intencity over a Range of Wavelengths or just the Peak Wavelength of light?

Re: Open Challenge to Lumens Testing..............

Here Viren:

http://www.labsphere.com/products.asp?parent_id=67&catId=96


Here is a good read:
http://www.labsphere.com/uploadDocs/A%20Guide%20to%20Integrating%20Sphere%20Radiometry%20and%20Photometry_6.pdf

Re: Open Challenge to Lumens Testing..............

I have a royal blue P4K Luxeon3 in a NuwaiQ3 that is a scream.
It's lumen rating does not reflect how strong it is in real life imho.

Re: Open Challenge to Lumens Testing..............

That sounds pretty interesting - I'm getting ready to mod a couple of Q3's with DD and (I thought) TWOJ's but I haven't bought the LEDs yet. Any way to show us some comparison beamshots with non-blue Q3's?

Re: Open Challenge to Lumens Testing..............

Great Read NewBie...

Thanks for the information...