To share the users initial design process of a LED Reflector

The initial design, the ID design has been fixed, the diameter D is a reflective cup delimited. Range according to the requirements of lighting, a meter that is outside the flood zone is the largest diameter of one meter lighting range, determine the depth-diameter ratio is about 1:1 or so. Then the depth of H has also been broadly estimated.


Consider the central spot of the requirement is as small as possible, which is reflected light exit point to relatively small, so sure is generally parabolic reflector, what CPC and the like were all not so curve equation to determine, according to D, H the value of the focal length can be calculated parabolic led to see the size of f., Cree XP-E, En, focal plane size of the LED can meet the needs of place.


For the above three parameters are interrelated, while the depth of H is given a broad range of requirements, then the calculated f values ​​can fluctuate within the range of about 1mm. Taking into account the requirements of this design is to try to make the optical angle is small. So although the shorter focal length of the light source seem more inclusive point of view, a convergence of capabilities better. But the big focus will bring a large dish size, means that the chip led farther from the reflector means that the chip's physical size relative to the whole system more similar to a luminous point, the angle was smaller. So try to focus big time, calculation and focus of LED chip edge relative to the maximum opening angle of the dish, about 1.5 degrees, the whole angle is about 3 degrees out. Should not be too great central spot. Anyway, if I completely parallel to the light, the center spot is not a black Mody. Or to an angle. The idea of ​​matter in the end not to do that.


Cree's chip actually connected to the battery pack, bare lamp against the wall looked at reading the specifications do not know, XP-E of the chip edge is really a yellow light, ah! More yellow than the XR. Is indeed almost a cheaper dollar. Easily turn cast out of a laboratory reflector cup, and sure enough lighting to see the area between the wall of a rhubarb circle. Hard to read, and headache is the boss also seems very concerned about the yellow circle. Han asked to do reflective glass is not specific, that the do not like the yellow circle. According to information before the Internet frenzy search, starting a cup of orange peel. Do not know the process, do not know how to control the scope of orange peel light Cup, anyway, is the light scattered.


Installed a Tracepro, simulated test out the above analysis about what Orange do not know how to simulation, is clearly not diffuse white surface properties of a class. Want to customize one, the definition of A, B, G scattering parameter, so that most of the light or in accordance with the law of reflection direction, but something more recent, and my heart is not static and impatient, a direct smooth cup of simulation, estimated to range up to flood poor some. Who does not familiar with their software, cramming information and examples see the forum to learn. Outcome of the simulation, illumination map is nothing like, worry about. There Kenken manual, set count points, a little understood. And then went straight to the lab, with the best that these machines simulate two million light, and then set the value of the minimum illumination, a little bit like this. Look at the software simulation of the range of half-intensity angle, and computing in scope than that or too big. But decided to do first to do the sample again, the simulation still can not see many problems, such as the yellow circle that I have not simulated, seemingly Tracepro can not directly show the distribution of spectral effects. And so have the time to carefully set the wavelength points, divided into different colors look.