Ever consider a R,G,B knife edged PCB combined to make white on a triangle shaped rod

[logsquared]

Im so lost on this thread

[mixedgas]

LOG, I think you and Swami are referring to RGB dichroic combining cube prisms such as the one shown combining the three beams in post #8. I'm referring to a pyramidal reflective optic.
That make sense?

Steve

[Laser57]

I think was considering the same kind of thing, a 120 degree pyramid mirror to join three beams together right at the smallest part of the tip/top, if they came in at a 45 degree angle to the mirror, is that what this is, or is it a special PBS of some type? This isn't a four sided cube we are seeing, right?

[flecom]

I had them made of precision optical glass in order to be precise enough and to avoid the offset of the mirror thickness. Its not as easy as it sounds, because the beams must be incredibly close to the tip for the beams to appear to converge. They never really converge except for a small region down range. That region is set by the feed angle. Commercial users do not use these for collimated beams, they have other uses in power combining optical systems.

.
.
.

Keep in mind, There is a ACTIVE US PATENT on the concept. You can make one for your own personal use per patent law, but any commercial sales will not be tolerated by the patent holder.



Steve

link to patent?

[logsquared]

Like this: (drawing isometric three sided pyramids in paint aint easy)


Or more like this: (this is drawn more as a 4 sided to illustrate the different entrance angle).

[buffo]

Laser57 and Logsquared are talking about an axicon, which is the project that Spec and MixedGas worked on (and which I referred to in my earlier post). The consensus at the end of the project was that it wasn't worth the effort.

The picture Swami posted is a trichro, like this one:



It's not the same as an axicon, although the end result is similar. (The trichro can't be used to combine multiple beams of a single color; the axicon can.)

Adam

[swamidog]

Laser57 and Logsquared are talking about an axicon, which is the project that Spec and MixedGas worked on (and which I referred to in my earlier post). The consensus at the end of the project was that it wasn't worth the effort.

The picture Swami posted is a trichro, like this one:



It's not the same as an axicon, although the end result is similar. (The trichro can't be used to combine multiple beams of a single color; the axicon can.)

Adam

i wonder what wavelengths those trichos are good for... they might make tiny single mode builds even more awesome.

[mixedgas]

Its not worth the effort for RGB or even RRRR. It was worth the effort for another task involving 5 or more beams and not involving free space beam effects or graphics.
A four sided glass unit costs around 200$ to have made, and you need quite a batch to get a dielectric coating run to be economical.

Steve

[Laser57]

How did that Trichcro work out, I can see it looks great in the photo but is it superior to a PBS cube in any way?

[buffo]

Swami and Laser57:

In the end, a Trichro is no better than a set of dichros, with the caveat that as a general rule the Trichro has a wider range of wavelengths that it will work with. This is because they were originally designed to function in reverse. That is, they took incoming white light and split them into red, green, and blue segments. They were used in high-end 3CCD video cameras. (And still are.)

One problem with a Trichro is that you still need to have your near and far field alignment set up correctly, but unlike a dichro there is no adjustment inside the Trichro. So everything has to be done externally. That means the beams need to be the same height, and parallel to the baseplate, and at a 90 degree angle to the input face, and they need to hit the same spot on the face for all three beams. This is a real bitch to get right. (Which is why you see the CD-r discs being used as shims in my picture above.)

The best way to align a Trichro is to use a double-bounce arrangement for each beam. The first bounce gets the height and position right, and the second one squares the beam to the face and makes it parallel to the baseplate. But that's 6 MM1's if you want to do RGB...

In the end a set of dichros is cheaper and easier.

Adam