DIY fiber coupling?

[Gryphon]

Yeah knife edging and beam expanding are no doubt the most cost effective and arguably simplest means of projecting a condensed far field spot.

I would be curious to see how something like this does with a collimator and beam expander over just a collimated diode and beam expander.
https://www.ebay.com/itm/2257229491...XYvr0DnltM0dFw%3D%3D|ampid:PL_CLK|clp:3458402

On a side not I retried my rig with a NUBM07. 900mW prefiber with 160mW post fiber. Nice circular dot that would need a beam expander for long range throw. It still focuses down to a ~1" dot at 10'

 

[RedCowboy]

My NUBM07 with a 3E lens would make a bar 5/8 of an inch wide ( on the wide side ) @ 15 feet and my NUBM44's would make a bar 7/8 of an inch wide @ 15 feet with a 3E. Now the same NUBM44 with a G2 lens makes a bar about 1 7/8 of an inch wide ( on the wide side ) @ 15 feet. That figures as the FL is 2.39mm vs. 4.5-5mm

That said a 3E isn't a fair way to compare because of clipping and the raw divergence, for instance the raw divergence of a NUBM44 is so bad that it will clip on the edges of the copper module without a lens before it can get out of the module. This is why comparing divergence with a G2 is best. ( Lens barrel opening dia. vs. FL ) No clipping.

That unit says it has 100um fiber, so compare that to a single mode emitter @ 3-7mm x 1mm or a MM emitter @ 30-300mm x 1mm ...... by the math the fiber has a lot more surface area. So a big round spot vs a long bar/line.

 

[Gryphon]

I ordered one of these in the 470nm 3.4W variety.

MultiMode 450nm/465nm/470nm Fiber Coupled Laser Diode Fiber Core Diamete 100um | eBay

Find many great new & used options and get the best deals for MultiMode 450nm/465nm/470nm Fiber Coupled Laser Diode Fiber Core Diamete 100um at the best online prices at eBay! Free shipping for many products!

www.ebay.com

As well as one of Shmackitup's 3X beam expanders to see if its possible to get a good round dot at range

 

[RedCowboy]

3.4W of 470nm, very nice, I am interested to know how well it projects a spot as well, the 3XBE is bound to help when used after a coli lens.

 

[Gryphon]

It took a month to arrive but its finally here. I haven't done any tests to see how much power I can squeeze out of it but I can tell you that at 1.2A it makes 950mW at the end of the fiber with no lens.

The fiber glows when it's powered on, sure that's technically power loss but it looks cool!

Testing with a G2

Now a G2 with a 3X beam expander

This is the stock lens that came with the cheap fiber collimator. The ebay ad says its a FL10

Now the stock collimator lens and the 3X beam expander

The spot is perfectly round and looks amazing, but even with the FL10 lens and 3X beam expander it still cant be focused down to infinity. It gets it close but the beam is still expanding at a decent rate.

 

[atomd]

Guys, remember conservation of etendue. You just can't squeeze some things without increasing divergence.

 

[kecked]

If you see 160mw with 900mw input and the fiberis not 500’ long you have not coupled to the fiber well and likely the NA isn’t matched. A fiber has a tight entrance angle range in which light can enter. If the cone of light isn’t matched you can’t put the light in the pipe. The NA or numerical aperature determines this. Also Lau hinging I to single mode fiber is hard and not required for your application. Try a larger core multi mode fiber.

 

[RedCowboy]

Well you do have a round beam, so if the goal is a small spot at distance, then you could use a larger beam expander, even a reverse telescope.

 

[farbe2]

..still expanding at a decent rate…

Thats to be expected because the fiber is likely huge to get the MM wide emitter launched into it without having to deal with expanding the slow axis and correcting astigmatism.

What atomd is trying to tell you guys:
That won’t work!

Yes you can couple some power in a big enough Fiber and it will even make your output beam nice and round but!!the beam parameters after your fiber will be at best the same as before.
However as the fiber is round and has no way of keeping the good axis and the bad axis of the diode separate, you will get a round beam with (best case) same divergence as the worst axis of your diode.

Other words: You end up with a circle that would be the same diameter as your bar shaped beam was wide.
(At the same starting diameter)
So you „loose“ your good axis.



You can however use a fiber that’s way too small to couple the diode in.
This would give you a somewhat smaller diameter of your circle but you will get heavy power losses (like getting 150mW out with 900mW of input..).

And you would still need to beamshape your slow axis with cylindrical lenses if you want to have good coupling efficiency.

 

[atomd]

Thats to be expected because the fiber is likely huge to get the MM wide emitter launched into it without having to deal with expanding the slow axis and correcting astigmatism.

What atomd is trying to tell you guys:
That won’t work!

Yes you can couple some power in a big enough Fiber and it will even make your output beam nice and round but!!the beam parameters after your fiber will be at best the same as before.
However as the fiber is round and has no way of keeping the good axis and the bad axis of the diode separate, you will get a round beam with (best case) same divergence as the worst axis of your diode.

Other words: You end up with a circle that would be the same diameter as your bar shaped beam was wide.
(At the same starting diameter)
So you „loose“ your good axis.



You can however use a fiber that’s way too small to couple the diode in.
This would give you a somewhat smaller diameter of your circle but you will get heavy power losses (like getting 150mW out with 900mW of input..).

And you would still need to beamshape your slow axis with cylindrical lenses if you want to have good coupling efficiency.

You're almost right. PM fibers should allow maintaining poor beam quality in just one axis if coupled correctly

 

[farbe2]

You're almost right. PM fibers should allow maintaining poor beam quality in just one axis if coupled correctly

As far as I know, this isn’t going to work.

We are talking multimode fibers here. Multimode PM fibers aren’t readily available.

Even then: PM fibers only work with purely monochromatic light. Phase differences of different wavelengths will result in coupling and energy exchange.

Multimode diodes are quite non monochromatic so PM fibers shouldn’t work.

 

[Boontwoont]

high launch powers or high brightness coupling of blue wavelengths typical of the GaN consumer electronics focused modules that we all know and love is apparently enough of a nontrivial technology that at least 1 company has staked their entire business model on *that* as their secret sauce.

that is how Nuburu's technology works. They do not fabricate diodes. They purchase diodes off the shelf from Osram (why Osram and not Nichia.... the company in which the technology was invented I dont know*)

Their "Secret sauce" has been explicitly and speficially described as being associated with their ability to launch high powers into a fiber, and they apparently have a patent related to this technology.

I suppose they most likely have a unique fiber end cap or something.