I was attempting to explain in my post why your reasoning is incorrect. For example, an aspheric lens light usually collects a ~30degree half angle when focused. With your reasoning, wouldn’t you expect the light collection to be good in aspheric lights? Well, it is not good because it only collects ~25% of the total light. Djozz’s measurements here support that; just 21% of the total light from the LED is in the beam when in focus mode.
But it is not incorrect.
Just look at the radiation diagram of an LED and than superimpose that on the picture in the OP, and compare that to the picture above.
The recoil set up will focus all of it, including the 65° a regular reflector simply spills,which happens to be about 66% of the emitted lumens (my guesstimate).
EasyB is 100% correct, I’m planning to add those functions to my reflector calculator soon.
Also with a parabolic reflector the outside of the mirror should not be creating a fuzzier image.
The ideal catch angle is 180 degrees for 100% light efficiency.
You can imagine it would collect the same 90 degrees as a smaller shallower reflector but if you keep extending the sides you will end up with a larger diameter reflector but 180 degree capture.
The problem is that the focal length gets shorter the more degrees of capture you try to cram into something tiny like 4 inches, and that is what will increase dispersion and “fuzzyness”
But if the focal length is too long, you will get a really light beam and low light efficiency.
Yesterday I was looking for other mirrors, I tested with one I had at home which didn’t focus very well (weird shaped spot) but it still collimated the light as good as a lens.
This will only get about 60-90 degrees from in front of the light, but I think it would work well
His conclusion is not correct.
What’s so hard to get from that radiation diagram?
Just look at the surface area of the circle.
Only about 33% is green i.e. reflected with a regular reflector, whereas a recoil set up practically uses all of it, and when limited to 120° it still is almost 90%.
That’s nearly 3 times more light focussed / collimated for the recoil compared to the regular set up.
Why do you think i’m so enthusiastic about a recoil thrower?
LEDs are practically made for it.
It’s clear you have not considered what I wrote in my post above (regarding how the radiation diagram does not tell the whole story) and so this discussion is just us repeating ourselves.
You use terms i’m unfamiliar with, so it’s kind of wasted on me.
However, there’s not much more to the radiation diagram than what it shows us.
(edit) ehrmm… there is… sorry…(/edit)
The radiation diagram is not of a circle, it is of a sphere.
The way the intensity is measure is with a fixed distance, so at 0 degrees it is Xcm in front of the led, at 45 degrees it is at Xcm diagonally from the LED, at 90 degrees it is at Xcm to the side of the LED.
I think you can imagine the circumference of the circle at different heights along a sphere, like this
But since the radius increases to the power of 1/2 and the intensity decreases by cosine then between –45 and 45 degrees there is over 50% light output.
So you’re both right, the intensity graph doesn’t tell the whole story and a recoil reflector is still definitely more efficient than forward reflectors or lenses.
The radiation diagram is not of a circle, it is of a sphere.
The way the intensity is measure is with a fixed distance, so at 0 degrees it is Xcm in front of the led, at 45 degrees it is at Xcm diagonally from the LED, at 90 degrees it is at Xcm to the side of the LED.
I think you can imagine the circumference of the circle at different heights along a sphere, like this
But since the radius increases to the power of 1/2 and the intensity decreases by cosine then between –45 and 45 degrees there is over 50% light output.
So you’re both right, the intensity graph doesn’t tell the whole story and a recoil reflector is still definitely more efficient than forward reflectors or lenses.
Okay, thanks for explaining that, especially the picture showing it is (of course! silly me…) a sphere.
This also makes sense as to why i felt that we should collimate at least 120°
And afterall, had i been right, a regular reflector would be quite silly for an LED, and there would be a LOT of silly lights out there.
