Re: Making the sun's rays parallel

So you want to make light rays from the Sun into a parallel bundle? This is a bit more complicated than it might seem at first glance, due to the appreciable angular size (half a degree) of the Sun. Let me try to explain the problem.

One often deals with light rays emanating from a single point source, like those shown in Figure 1.

If one observes these rays at some small distance from the source, the rays are not parallel. However, they do follow a simple rule: the rays above the middle are tilted upwards, and the rays below the middle are tilted downwards. The further away from the middle of the bundle, the larger the tilt. You can design a lens or a mirror to correct these rays so that they all become parallel; all you have to do is adjust the surface of the lens (or mirror) so that its properties at some distance above or below the optical axis make the proper correction to the light rays at that position.

However, if you are dealing with a light source of appreciable extent, things are more complicated. Look at Figure 2, which shows two point sources some distance apart; you might consider these to be the middle and edge of the solar disk.

Each source emits its own bundle of rays, which diverge separately as they move towards the right. Note that some of the black rays ARE parallel to some of the blue rays ... but other black rays ARE NOT parallel to some of the blue rays. That's the problem. There's no simple rule which connects the position of an arbitrary light ray to its tilt.

What can you do about this? Well, you can simply things by selecting only a small subset of all the light rays from the Sun. Look at Figure 3, which there's an opaque wall with a small hole. If you pay attention only to the light rays which pass through the hole, then once again there is a simple relationship between the location of a particular light ray and its tilt.

You can see that the blue ray is tilted relative to the black ray, and it is also above it. If there were another point source below the black one -- say, a red source at the opposite edge of the solar disk -- then the only rays from that source which would go through the hole would be both below the black rays, and tilted the other way. You can easily design a lens or mirror to take just the rays which pass through the hole and make them all parallel.

However, as Figure 4 shows, the larger you make the hole, the harder it becomes to make all the light rays parallel.

You can see here, with a large hole in the wall, that some light rays from the blue source which pass through the hole are parallel to the black light rays, but other light rays from the blue source are NOT parallel to the black rays. I think you could figure out a way to design a lens or mirror which would make all the blue rays which passed through the hole parallel to each other ... or one which made some of the blue rays and some of the black rays parallel ... but I don't think you could make all the blue rays and all the black rays which pass through this hole parallel.

In summary, I think that if you want to have a very parallel set of light rays, you may need to include a small pinhole in your optical system. On the other hand, if you want a lot of light rays, you'll need a big hole, or no hole at all, in which case the resulting rays won't be very parallel. I think you'll need to decide exactly what is most important to you -- parallel rays, or lots of rays -- and then make your system accordingly.

Of course, if all you care about is LOTS of rays, and only a very approximate degree of parallel-ness (ugh, is that a word?), you can use no mirrors and no lenses; the rays from the Sun are less than a degree away from a parallel beam naturally.