Field testing of red & white LEDs

Some interesting (I hope!) observations.

Recently received TLE-5 modules in red and white for MiniMag, also red Peak McKinley 7LED 123a and a brother in white. Took them and a snow white Peak 1xAA and an Inova T3 for a two-hour nightime run in the deep woods with the dogs.

Conditions: Overcast, some sky glow from general Atlanta lighting, misty with light ground fog. Woods border housing across stream in places, but they contribute little to general lighting. With night-adjusted eyes, could see to navigate somewhat uncertainly on trails, much more difficult off-trail.

I got the red lights for their value in preserving night vision. That was all I expected. I wanted to use them intermittently as needed and be able to see surroundings when I turned them off. The McKinley was much more of a bright spot - exactly as bright as the red TLE-5 MM if I blocked off the spill with my hand - but figured that too would have a use. The red TLE-5 MM was equipped with a frosted glass lens and that gave it very usable focussing properties - in contrast to the original which is far too ringy for me - and provided a wide diffused red light with adjustable hotspot.

I discovered some unexpected advantages to the red. Not what I expected at all. I figured that monochromatic light would limit my perceptions and its only value would be the preservation of photopic (rod) vision.

In fact, the red light emphasized contrast, gave more apparent texture and depth to the illuminated scene, and made identification of trails and hazards much more apparent.

There was a difference in accomodation period: after flashing a comparable white light for a few seconds, it would take about one minute to get back to where I was initially with night vision. Using the red flood for even extended periods, I seemed to return to normal vision in about 10 seconds. This suggests that the red accomodation was mostly due to iris adjustment (going to a lower f-stop as it were).

As I got more analytical about it, I noticed that under red light (a) dark colors that were not shiny turned entirely black, (b) shiny dark leaf surfaces were brighter, (c) high albedo objects like beech leaves and branches were very white. There was a very noticable difference between bare soil on trail surfaces and surrounding leaf mold and bark - but also the shadows appeared to have more depth. I did many comparisons, and the upshot was that the flatter, slightly blueish LED light at any level tended to make surfaces appear "flatter" due to diminished contrast, and even though monochromatic, the red light seemed to strongly emphasise the contrasts and show more detail and texture.

This phenomenon became more evident after some observations in my wife's closet. For example, consider a medium-grey cloth shoe on an oak floor. In daylight, moderately contrasty and distinguishable. White LED shone in the dark closet rendered them very similar in brightness, the color difference largely offset by similar reflectances. So much so that the shoe seemed to recede into the background of the floor. Incandescent halogen flashlight rendered it a little more distinct. Under red LED light, the shoe fairly seemed to jump out at me. Monochromatic but way different in reflectance properties. Also shadows around objects emphasized and less illuminated.

Another thing I noticed about the red LEDs was the rapid diminishment in throw with distance. Close in the red could seem almost blinding, but at 15 meters, rather dim. White light did not show nearly as much falloff. That makes me wonder if the eye's sensitivity to red is non-linear. In other words, perhaps half the lux of white light is perceived as about half as bright (barring adjustment factors) but perception of red might fall off more rapidly - perhaps a logarithmic relation. That might also contribute to the apparent enhanced contrast; if reflecting half as much light, could seem like one-quarter.

Another property was reaction with atmospheric haze. There is a related discussion about this elsewhere on this forum under "fog lights", but suffice it to say that Rayleigh scattering of bluer light doesn't appear to be a viable mechanism re fog. However, I did note that both white and red lights illuminated fog from my breath equally, color being the only difference, but at greater distances the atmospheric backscatter from white light was unquestionably more prominent that that from red. So red, regardless of what instruments might tell us, appeared to penetrate the foggy haze better (adjusted for falloff in throw).

In any case, with night-adjusted eyes, I could see to navigate at running speed better with the red than the white. This came as a surprise. Loss of color more than offset by higher contrast. In many places I shone a white light on the route, turned it off, let my eyes accomodate, then turned on the red - and the red always provided more certainty about terrain and the location of trails.

One other thing related to backscatter. The red can sometimes help you to see "deeper". Consider a speaker in my truck; it has a black metal grille and beyond that, a cone with a tweeter and various interesting details. I shine a white light on it and I see very little beyond the grille. The reflected light from the grille blinds me to everything else. I shine a red light on it and the grille seems to disappear, and all sorts of details in the interior become visible. Translate this to the woods - backscattered light from branches, mist etc. - and you see the wider implications.

Just my observations with my eyes - any comments or diverging opinions welcomed!

- Bill

woodfluter said:

Another thing I noticed about the red LEDs was the rapid diminishment in throw with distance. Close in the red could seem almost blinding, but at 15 meters, rather dim. White light did not show nearly as much falloff. That makes me wonder if the eye's sensitivity to red is non-linear. In other words, perhaps half the lux of white light is perceived as about half as bright (barring adjustment factors) but perception of red might fall off more rapidly - perhaps a logarithmic relation.

Click to expand...

First, keep in mind that ALL vision is roughly logarithmic...

You're right in your observations but, I believe, only partially correct in your analysis. As best I can tell, the reason red seems more contrasty and seems to fall off faster is exactly the reason you wanted to use it - the eye's most sensitive rod receptors, used for night vision, are not sensitive to red. If the light is bright enough, the color receptors in the eyes will perceive intensities similarly, but when the brightness crosses that boundary, dim red will drop out. Green/cyan is the peak of night vision sensitivity; a white LED has a peak in blue (the LED itself) and a somewhat broad second peak centered on yellow (the phosphor), both of which would be detectable to a decent degree by the eye's rod cells.

Shadows will appear deeper because the faint illumination that bounces off objects and would ordinarily fill in shadows is too dim in red.

Hey Weep -

I think I see what you are saying, and it makes sense. Peak rod sensitivity in the blue-green segment, none below 630 nm, so with white (or green or even yellow) light I am seeing combined benefit from cones plus rods at farther, dimmer reaches - but with red light, no added illumination for the rods to see, so looks like it dims faster (only because the rods aren't in play as far as the added red illumination is concerned). If I understand it correctly.

However, not sure it explains the closet observations. In daylight I thought the rods were depleted, or mostly so, and I was doing that in daytime, lighting up a dark closet.

But in general, seems like a very good explanation unless a better one emerges - thanks Weep!

- Bill