Narrowing the beam and focusing a laser through additional lenses?

[haveblue]

I have this 300mw+ CNI PGL-III-C. It has an initial beam diameter of roughly 2mm. Then of course, it has beam divergence. The upper image shows (exaggerated) the stock focusing lens and the diverging output beam. I want to replace this lens with a more convex one to converge the beam, and then add a second focusable convex lens in front of it that refocuses the beam to neither diverge nor converge, like in the lower image.



Where can I get optics to do this? If I know the angle Θ, I should be able to find the lenses I need, correct? I've converged the beam before with a lens. It seems that at its focus in the picture below, the beam is 0.04mm. Thus it should be very possible to rediverge the beam at the first lens's focal point.

(Each dash mark is a millimeter)

[danielbriggs]

Can't answer the full question; but if, for example, you are planning to half the beam diameter, your divergence will increase 2x.

I think your main problem will be alignment...

[mixedgas]

Can't answer the full question; but if, for example, you are planning to half the beam diameter, your divergence will increase 2x.

I think your main problem will be alignment...

2nd law of thermodynamics says NO, you can not go down in beam size. The concept is called Etendue if you want to google it. So If you make it 2x better in collimation you make it 2x bigger in diameter, has to do that you cannot gain energy for free, with no work done.

I do encourage you to play with the factory collimating lens though, often they are terrible.

This might help.

http://www.newport.com/OpticalAssist...mExpander.aspx

Steve

[danielbriggs]

2nd law of thermodynamics says NO, you can not go down in beam size. The concept is called Etendue if you want to google it. So If you make it 2x better in collimation you make it 2x bigger in diameter, has to do that you cannot gain energy for free, with no work done.

Thanks for that Steve, I'll have a detailed read of that later.
But why doesn't the rule work in reverse?
Surely no work is done on the beam?

[Laserman532]

man, when you figure it out...this would be very handy to use on laserscopes

[-bart-]

man, when you figure it out...this would be very handy to use on laserscopes

When he figures a way out, why use lasers in the first place, everybody could be using halogen lamps instead.

[142laser]

Steve is right; this is simply not possible. If you reduce the diameter by a factor of two the divergence will increase by the same factor. He is also right that another lend may work better than the factory model.

Steve...do you still need a job? We sure need a good engineer. Lines are becoming available after they fired about 1/2 the technical staff here. Phil, 142laser

[mixedgas]

Steve is right; this is simply not possible. If you reduce the diameter by a factor of two the divergence will increase by the same factor. He is also right that another lend may work better than the factory model.

Steve...do you still need a job? We sure need a good engineer. Lines are becoming available after they fired about 1/2 the technical staff here. Phil, 142laser

resume coming down soon.

Steve

[Laserman532]

resume coming down soon.

Steve

hey!!! not fair... will they let me telecomute from bend oregon

[mixedgas]

Thanks for that Steve, I'll have a detailed read of that later.
But why doesn't the rule work in reverse?
Surely no work is done on the beam?

Feynman would have said something sarcastic, like, "Do you know anyplace else in the Universe where God gives you a free lunch?"...



Ok, how to explain this???? I need a day to think. Its not intuitive to explain. Its called a transfer function.

The first part: Newtonian Physics. Kind of a B.S. answer, but it helps to think about it although not 100% correct.

The laser beam is never perfectly parallel, at some point there is a beam waist or crossover point. It may be in front of the laser, behind the laser, or inside the cavity. Any other point in the beam is a projected image of that spot. As the rays are not perfectly parallel to begin with, no matter what you do, you always are projecting a image of a diverging beam. So as you focus down to a tiny point, the angle increases because no matter what lens you use, you cannot get rid of the fact you are projecting rays that have crossed over and thus will always have a angle between them. You also have a issue that the crossover is not a perfectly thin , tiny spot, so there is a limit you cannot go beyond, The lens cant do anything about the fact that beams coming from near the crossover come from a different point along the beam. And the fact that the wavefronts from the laser are curved, not flat, as you move outwards from the center of the cavity mirror, the photons have a longer trip time and come out slightly delayed.

So you say use two perfectly flat mirrors? Won't work well, low gain, and very, very, hard to align. They will not stay stable, and will not confine many of the photons in the laser. In reality no mirror on earth is flat. Remember the curved, cavity mirrors help the beam feed back on itself.
-----

Next up, Non Newtonian Stuff:

To get to a perfectly parallel beam, the beam would have to be infinitely huge in diameter.

I stand corrected, its not thermodynamics as much as something called " Lagrange-Helmholtz" and Gaussian mathematics.

It also has to do with diffraction,

read: www.specialoptics.com/pdf/wp_laser_beam_expander_theory.pdf
---------------------------------
Download and install PSST!

from http://www.st-andrews.ac.uk/~psst/

And play with the cavity radius and spacing of each mirror.

try 200 mm radius concave mirrors at 50,99,100, 199, 200 and 201 mm apart from each other, and then do a flat and 200 mm at the same spacing. See how it can NEVER be parallel?
------------------------------------
Remember those math functions your teacher had you graph that would get very close to the axis of the graph, but never touch or cross it even at infinity? Aka asymptotic function? Guess what, this is one of those...

I do not have the understanding of Hermite Polynomials and Gaussian math to correctly explain this to you. God saw fit to install a recessive gene in half of the Roberts males that inhibits higher math, much to the dismay of my Uncles who were PhDs and thus highly unhappy with me. Its called dyscalculia , You can outgrow it to a point, but right now I'm slamming into my math limits.



The above answer has large amounts of factual flaws, but its the best I can do in a hurray.

Steve