Multiple laser diodes prior to collimation?

[Alaskan]

If you had several laser diodes mounted on one heat sink together and then collimated all of them with one lens, can you get a single beam out of the lens this way? If not, can the mounting surface for the diodes be curved in a way to help make the beam tighter? I'm guessing things get a bit hay wire when attempting something like this, otherwise I would see projects using this method instead of combining diode outputs with mirrors. Still, the idea interests me.

 

[Pi R Squared]

Only 3 ways to combine beams as far as I know. http://laserpointerforums.com/f51/reference-guide-how-combine-lasers-77449.html

Alan

 

[Alaskan]

Thank you, I've never seen that thread before and helpful, but I still wonder what the product is when having several diodes aimed into one lens, maybe the problem is the wavelength troughs don't line up right and you get plus and minus distortions and end up with a very out of phase product, collimated, but out of phase. I'd sure like to see what happens but was hoping someone has done it, maybe save me the trouble of building one to see.

 

[Pi R Squared]

The beams would be a small distance apart, I think you would get multiple dots, better to properly knife edge them creating one larger beam and then use a lens, as is a common practice. You do get a single beam that can be focused, its just a larger beam.

Alan

 

[ARG]

If the beams don't enter the collimation lens at the same angle they will not exit at the same angle. You'll have to overlap the beams somehow before the collimation.

 

[upaa27]

You can use a bigger lense and focus the light into a fiber cable. Also provide a connection between the diodes and the cable so that there is minimal power loss. I would use an ytterbium doped cable for max output in the ir range but I do not know what wavelength the input will be. Don't forget to slap on a focusing optic at the end of the cable!

Can you tell what diodes you are using?

 

[Alaskan]

I have some interest in doing this with IR, but also with blue laser diodes too, if some method will work. I was wondering if the light from several raw uncollimated diodes each falling upon the same area of a collimating lens (all producing overlapping wide spots on a single PCX lens) will roughly combine into a single collimated beam, once going through that lens. Somewhat distorted or not with interference patterns or not, my goal is to make a long throw flashlight with minimal divergence yet a big fat beam of a few inches in diameter for as long a distance as I can keep it like that.

If I can get a powerful enough beam by combining the raw output of several laser diodes then I will work on making a beam expander to both widen the beam while reducing the divergence at the same time, if combining power together like this is technically possible while continuing to maintain a high level of coherence.

Can something like this work? Anyone?

I haven't selected diodes for this yet, but I do have a large number of M140 blue laser diodes (over 30) which will likely be the ones used. Damn, I worked 45 minutes on this post trying to word this right, my momma always said I was her special boy, I guess she was right because I feel so low IQ'd on this.

 

[upaa27]

That should definitely work but all the loss would make it more efficient to just use a single diode and unfocus the lense a bit.

Lol all I do is research lasers now -_-

I need to go out more often.

 

[Alaskan]

What will cause so much loss? Cancellation from out of phase energy?

 

[upaa27]

I don't really know what cancellation from phase energy is but basically if you are firing a bunch of unfocused diodes into a lense 20%+ of each diode's energy is going to hit the lens holder or module and not go through.

 

[Alaskan]

I meant if uncollimated light from several laser diodes are all aimed at one spot, some will be out of phase from one another and cancel out. i.e., if a wave of light is 180 degrees out of phase from the light (at the same frequency) from another diode, they cancel and no net gain. I expect an amount of that will happen because each diode is slightly different and the wave fronts won't meet at the same exact phase.

Regarding the loss, I think I understand what you are saying but from my perspective it doesn't matter, you still loose the same percentage whether 1 diode or 10 diodes. 20 percent of 1 watt, or 20 percent of 10 watts is still 20 percent either way you look at it, you can't avoid it.

 

[upaa27]

Very true. Have you looked into 1/2 waveplates? They might be able to do the trick. Also there will be little to no loss if you are only using a lens. Kind of like a standard laser just defocus the lense

 

[Alaskan]

I guess you mean defocus the lens to make a wider beam, but I also want the divergence to be as low as possible. What is the relationship of a defocused beam to divergence?

I am just now looking into (googling) 1/2 wave plates and trying to understand the application such could have for this project, I'm not following on that yet.

 

[Cyparagon]

Would you mind clearing things up by sketching your idea?

 

[Alaskan]

My drawing skills are poor but I found a photograph which is somewhat representative of what I am trying to do, but I can draw one for you if this doesn't explain things well enough:

Looking at these three spot lights, see them as three laser diodes without lenses all aimed at one spot and that spot is the flat side of a PCX lens, wouldn't the light from all three combined make a brighter output and would that output have a limited divergence the same as if one diode? Before you start laughing, I realize if there was this much of an angle between each of the sources the output of the lens would have three widely separated focal spots of light, because of this I would put the diodes as close together as possible, all in parallel radiating into the flat side of a PCX lens with minimal angle between them.

Doing so, would this work to increase the power of the light if using several laser diodes as the source instead of one, maintaining a single beam with tight divergence regardless of the interference patterns which might be produced from having separate coherent light sources? I'm thinking it would but I cannot find an example of something built this way yet, not for use as a flashlight. Hope you guru's aren't laughing too hard anyway - I work as an engineer with RF, although light is also electromagnetic, it's a different ball game.

 

[upaa27]

You would have very intense divergence but it would work