I need to use a small (size of a flashlight) ultra narrow projection light to cast shadows or shape the beam (like a gobo). The beam needs to be well collimated and even with a very narrow beam angle, about 5 degrees or less, so that at a distance of 10yards the projected beam measures about 10inches in diameter. I prefer to buy something that is already available but I am willing to modify or build one if needed.
For example I have an inexpensive flashlight with CREE XML LED that almost works. The problem is that when focused the beam angle is around 8 degrees, I need something with half the beam angle. In addition the only way I can shape the beam and project a shape is by unscrewing the aspheric lens and placing something directly on top of the LED which might not be great for the life of the LED or the element that is close to the LED.
I’ve been looking on the internet for tight beam lights and there are different technologies available, like TIR, recoil reflector, waiven TLR. Unfortunately I don’t own a flashlight with any of that technology so I can’t do any test and see which one works best for casting shadows.
I found some technical information on TIR and there are some that produce a 3.6degree angle but I don’t know if they would wok for casting shadows. I’ve also seen spot lights with claimed 1 degree beam angle but they don’t mention what technology they use. Any ideas what would work for this application?
Generally, you get a smaller spot with a longer focal length lens, a smaller LED or by dedoming the LED. You can tell the focal length of a zoomy by how far the head slides. The usual small LED is the Cree XP-E or its older version the XR-E. There are smaller LEDs, mostly Osrams. The Black Cat 1 x AAA light has one. Dedoming is more or less for experts.
The heat normally is mostly taken away by the metal star and pill behind the LED, so putting something in front of it won’t hurt it. It will be easier to form a shadow if you use a large LED such as an XM-L2 and a long enough focal length to get the spot size you want. You probably should move the lens forward, or the LED backward, maybe by putting an o-ring behind the bezel ring, so the shadow shape is in focus in stead of the LED.
Not all glass lenses have the optical quality you need, but I think nearly all plastic zoomy lenses do.
Show us pictures!
Thank you very much for the reply and explaining some things. Placing a complex shadow shape right on the LED would be tricky because of the small size of the LED. I would have to move the shadow shape slightly away from the LED to increase the size and also move the aspheric lens farther away to focus on the shape instead of the LED. This means that the LED will not be at the focal point of the lens and the beam would not be as focused and collimated.
I found this paper by waiven where the explain how they show how it can be used for a gobo projector.
The slide with the shape is placed right after aperture. I am not sure what the field lens and objective lens are and how to calculate them. Is this perhaps a more efficient way to project shapes?
Also, the TIR or reflectors are not suited for this application?
To use a TIR or silvered reflector effectively, you would need to turn the LED backwards, place it in front of the reflector and find some way to cool it such as an aluminum or copper bar or filling the space with liquid. So a lens seems much more practical, unless you want to use a large reflector that is separate from the flashlight part. Reflectors have the advantage of not having chromatic aberration.
The aperture and shadow mask are focused by the projection lens. The “field lens” seems to avoid loss in light that would otherwise result from moving the projection lens forward, without significantly affecting the projection of the image. As a flashlight, the additional converging lens and the LED being out of focus decrease the throw, so the spot won’t be as bright as that of the flashlight from which it is derived.
Regarding the field lens approach, what type of lens should the field lens be, a double convex? Also both the field and objective lenses should have the same focal distance or the focal lengths are different?
To get those things fully optimized you would have to find or write a computer program that optimized a figure of merit within constraints. But I think a lot of configurations would work. The diagram shows the field lens with shorter focal length, because it is closer to the LED than the objective lens is to the shadow mask and because it forms an image of the LED at the objective lens. Forming an image of the LED at the objective lens may allow the objective lens to intercept maximum light for its size, or nearly so, but that may not be essential. It might be better to work on other parts first.
It looks like it will not fit in the optics space of the same flashlight that the main lens is from, but it might fit if you use a lens that is shorter focal length than that of the host. For example, a UniqueFire UF-t20 lens and a Yezl t9 host. (or, more reliably, an UltraFire in the same style as the Yezl)
An other point. The diagram shows the shadow mask much larger than the LED which, with the same focal length lens, would make the spot much larger than that of a flashlight using the same objective lens. So if the proportions resembled those of the diagram, you would need a much longer focal length objective lens.
I’ve been thinking about this for a while, unfortunately I am not good in optics theory so I still have some questions. Here is another diagram I came up with:
The field lens is focused on the LED so the light emitted by the LED passes through the lens and leaves almost parallel or slightly expanding if the objective lens is larger. The gobo is placed right after the field lens. The objective lens is focused on the gobo/field lens so any light (which is almost parallel) reaching the objective lens will have to converge towards the focal point on the other side.
If light is converging after the objective lens how can I keep the beam angle really tight (less than 5 degrees)?
Or is the schematic wrong and more lenses are needed for this task?
If you want the shadow to focus on a distant object, focus it so the rays from a point on the mask are essentially parallel as they leave the objective lens. Moving the objective lens to the right slightly in the diagram above will do that.
I like the field lens being behind the mask, it doesn’t effect the focus of the mask at all that way.
I don’t know if you need the field lens. Maybe leaving the dome on the LED is enough convergence. The smallest spot for a given objective lens is with no field lens, the LED dedomed and the LED immediately behind the mask. If you need the mask to be bigger than the LED, then you need a longer focal length, and preferably larger, objective lens and a configuration more like that above.
Here is a proof of concept.
I used a Yezl t9 that I has a UniqueFire UF-T2 lens in. The t9 had a longer focal length than the t20, so there is a lot of spare zoom space. I cut a piece of window screen from a repair kit, bent it and glued it to the pill. It is bent so it is flat in the middle and about a centimeter in front of the LED. Then I just slid the head until it focused. To get a smaller spot, I would bend it so the screen is closer to the LED so that the light was almost zoomed in when the screen was in focus. The magnification of the screen would be the same, but the spot would be smaller and brighter.
I lean towards placing something directly on top of the LED like you did, if possible dedome the LED and have the mask almost touching the LED. This way I can get the aspheric lens to focus on the image without loosing too much of the narrow spot light function that I need. I think this situation would be more efficient than using a field lens, what do you think?
