Best way to combine Beams for Power?

Ok guys, I have an idea that may sound crazy but I have been musing over it for quite a while. While I am planning on being a physics major, most of the detailed formulas are quite beyond my level right now and so I cannot seem to figure a solution out. Since yall have more experience with light and optics than me, I now turn to you.

What I want to do is to combine many powerful Laser Beams into one laser beam to maximize the power. Much like the build where you focus a bunch of the 1W projector diodes into a focal point, but here you wouldn't result in a spray of 1W beams diverging from the many Watt focal point. Instead there would be a many Watt beam that would be slightly less focused than the one with all of the diodes going in on a focus point and spreading out behind it. I have thought about prisms, I have thought about mirrors, hell, I even thought about magic (combine sesame!).

The only thing I could think of would be a many faceted prism that would have the light from every beam hitting a different face of the prism where the face and the angle of the beam were each designed to refract and focus towards the same spot at the end of the prism.

So you would have unique entry points, all designed so that the beam would be refracted to the same spot on the opposite end of the prism which would be a unified exit point. It would need very exact angles and the prism would be difficult to construct, granted, but if you could, would this work?

If not, do you have any other ideas? :bowdown:

Tada!

Why settle for just knife edge combining or just PBS cube combining when you can do both! :evil:

Gryphon Looks Awesome and Damn Expensive!!

Gryphon said:

Tada!

Why settle for just knife edge combining or just PBS cube combining when you can do both! :evil:

Click to expand...

Philip over on PL made that a couple years back, still impresses
every time I see it.

Alright, I took one look at that and was like :wtf: But then I actually took the time to figure it out and that is such an ingenious design....I hadn't thought of doing that before....only problem is it wouldn't be anywhere near portable....

Correct me if I am wrong, but do the lasers right before the end look like a concise square of many beams, put through some sort of concave lens or something with another lens at the focal point?

If so I have an idea....a GRAND idea!

If not, what does it look like at the end?

wannaburn said:

Philip over on PL made that a couple years back, still impresses
every time I see it.

Click to expand...

So He has 23 reds combining into 2 beams or does it go from 23 to 2 to 1?

It's a little hard to make out the final output.

23 to 2 to 1.
Near the tail end of the right green arrow is a black square, thats the top of a PBS (Polarizing Beam Splitter) cube. The 2 beams become 1 in that PBS cube, entering from left & north sides, the single beam exits the right side of the cube, bounces off two mirrors & heads left through the lenses (green arrows).

(PBS can split or combine)

wannaburnstuff said:

What I want to do is to combine many powerful Laser Beams into one laser beam to maximize the power.

If not, do you have any other ideas? :bowdown:

Click to expand...

Combining lasers is easy. But how do you intend to deal with the phase problem?

LASER OPTICS: Sixteen laser beams combined with 88% efficiency - Laser Focus World

Now I'll give you one better. Mind you this is a PhD level topic. Combine all the lasers at one point. But pulse them in such a way that when they combine they emit a continuous beam. The pulsed output of each beam can be much higher then if they were all CW.

zerocover said:

Now I'll give you one better. Mind you this is a PhD level topic. Combine all the lasers at one point. But pulse them in such a way that when they combine they emit a continuous beam. The pulsed output of each beam can be much higher then if they were all CW.

Click to expand...

I was thinking that SAME EXACT thought at work today!!!! Glad to hear another person propose it who is more experienced than I!

Gryphon said:

Tada!

Why settle for just knife edge combining or just PBS cube combining when you can do both! :evil:

Click to expand...

That is just plain unholy. :bowdown:

zerocover said:

..... Combine all the lasers at one point. But pulse them in such a way that when they combine they emit a continuous beam. The pulsed output of each beam can be much higher then if they were all CW.

Click to expand...

Yes, but the result will be the same.

Trying to explain: say that you have 10 diodes of 1W each CW, ok ?

Then you can combine them with knife-edge in CW mode, and have 10W output (ok, ok, less than 10W due to the various loss, but this is a theoric example ).

Now, you decide to pulse them at 10% of the duty cycle, and this means you can drive them at 10W for 100mS each second, and combine them sequentially pulsed, so the beam appear continuous ..... but still you have 10W output, at the end ..... and a bigger complication at the start, due to the fact that you need pulsed drivers and precise sequencing circuits, where for CW mode, you just need current drivers .....

wannaburnstuff said:

The only thing I could think of would be a many faceted prism that would have the light from every beam hitting a different face of the prism where the face and the angle of the beam were each designed to refract and focus towards the same spot at the end of the prism.

So you would have unique entry points, all designed so that the beam would be refracted to the same spot on the opposite end of the prism which would be a unified exit point. It would need very exact angles and the prism would be difficult to construct, granted, but if you could, would this work?

If not, do you have any other ideas? :bowdown:

Click to expand...

It will not work. If you trying to think of ways to combine beams, envision how your setup would work in 'reverse': if you shine a beam into the exit face of the prism, would it split into several discrete beams coming out each face?

It obviously would not: from the entry face the beam goes straight through the prism (if at a normal angle) and out of whatever (single) face is on the other end. If you were to shine the beam in at an angle, it would get deflected to a certain angle, and still come out of one single face, as one single beam.

Any beam combiner MUST also be a beam splitter if light enters where it would normally exit. For PBS cubes this holds true for natural polarized light, and for knife edging it also holds true if the beam is the same diameter/shape than the one exiting.


One important note: a beam splitter is not neccesarily usable as a combiner. The prime example is a half-silvered mirror that splits a beam in 2 just fine, but the reverse process does not work, because it would create 2 output beams.

As for the pulsed laser idea: This could work if you had a spinning mirror, for example hexagon shaped. You could position a number of lasers along the sweep that would be created if you where to shine a beam at the spinning mirror, and fire those lasers sequentially.

This is only useful if you have lasers that can do powerful pulses, but cannot sustain the power level continously. For diode lasers this is pretty much useless, but you could combine the output of a number of pulsed-YAG lasers into a beam of their combined average power.

zerocover said:

Combining lasers is easy. But how do you intend to deal with the phase problem?

Click to expand...

:thinking: Phase problem is referring to destructive interference?
Unless I am mistaken the beams are not mutually coherent, they don't share the same relative phase. Also depending on the wavelength, diodes will vary +/- 10nm. Multiple diodes won't actually be quite the same frequency. Destructive interference doesn't seem to be a problem in these setups.

Uhm, being sincere, there is one way for combine multiple laser diodes in one, but first, it's not practic, and second, it work good for "burning focusing point" (welding, cutting, etc) but, probably, NOT for obtain a straight beam ..... optic fiber.

If you shine multiple beams in an optic fiber (like in the multiple emitters IR arrays fibercoupled, where all the emitters are "combined" in the output connector in a "clover" hope it's the right word, pattern), at the other end of the fiber you get a round cone of laser emission, that can be focused precisely in a round spot ..... but this, first, require some minimum lenght of fiber (before you ask, yes, i already tried using small thin rods in front of LDs, but it does not work ), so the bouncing of the light in all the fiber and the transmittance effect of the fiber core "omogeneize" (LOL) the various beams reflection angles in a single emission cone ..... and second, also after this, a cone of light is not the better thing for obtain a straight beam (uhm, maybe with a custom expander ..... never tried it, anyway)

HaloBlu said:

:thinking: Phase problem is referring to destructive interference?
Unless I am mistaken the beams are not mutually coherent, they don't share the same relative phase. Also depending on the wavelength, diodes will vary +/- 10nm. Multiple diodes won't actually be quite the same frequency. Destructive interference doesn't seem to be a problem in these setups.

Click to expand...

Indeed, there will be differences in wavelength using multiple lasers that are not linked by some mechanism, and the total beam will not be coherent at all.

If this is a problem or not depends on the intended application. I suppose that for example, obtaining a powerful beam for a laser projection, doesnt require the light to be monochromatic, coherent or consitently polarized.

Things like knife edging and PBS combination don't actually combine the beams, but allow several beams to overlap to some degree. It would still be theoretically possible to tell which photon in the combined beam came from which laser diode.


If you actually had a set of lasers that were mutually coherent, combining them using interferiometrics would be possible. In practice it would still be a daunting task, as you'd have to position the lasers with accuracy smaller than the wavelength. In astronomy such feats are achieved in telescopes, but that's a bit beyond budget for most hobbyists