Fiber launching these 445nm diodes

[ogoun]

Hi,

Anyone have any opinions on how well these diodes would work coupled (via something like a 405nm glass collimator) into a 473nm SMA fiber launch?


Cheers,

Pete

[mixedgas]

Pete pushed 100% recyclable electrons along to say:

"Anyone have any opinions on how well these diodes would work coupled (via something like a 405nm glass collimator) into a 473nm SMA fiber launch?"

You need to specify the fiber diameter and fiber NA you want to use. 600 micron fiber, sure, no problem. However, to get significant amounts of this light into a 62/125 common communications fiber is going to involve circularizing optics.

For a bit more of a challenge read up on optical "Etendue"

Steve

[ogoun]

The SMA fiber launch has its own lensing optics to focus a typically collimated laser output beam into a single mode fiber cable, via an SMA fiber connector.

These are ex lab instruments, and were used with small 473nm DPSS lasers and a spectrophotometer.

I am wondering whether the combination of the 445nm diode, the aixiz 405 3 element glass collimator and the fiber launch will result in a useable percentage (~50%) of emitted light to be coupled into the fiber...

I realise that there are many unknowns here.. I am mainly just musing out loud.


Pete

[Laserman532]

The SMA fiber launch has its own lensing optics to focus a typically collimated laser output beam into a single mode fiber cable, via an SMA fiber connector.

These are ex lab instruments, and were used with small 473nm DPSS lasers and a spectrophotometer.

I am wondering whether the combination of the 445nm diode, the aixiz 405 3 element glass collimator and the fiber launch will result in a useable percentage (~50%) of emitted light to be coupled into the fiber...

I realise that there are many unknowns here.. I am mainly just musing out loud.


Pete

without achieving the proper beam characteristics to enter the NA at the power level of a watt or more you might burn the cladding which could cause you to burn the fiber tip.

Get the beam perfectly round first, then launch into a 10x or more objective lens to couple into at least 100 micron should be doable.

[ogoun]

I might be missing the point here, or maybe not expressing myself well. The parts I am talking about (launcher, sma fiber connector, fiber patch cable) are all existing commercial products which I already have here in the lab.

The optics in the launcher (similar to a collimator) already launch into the fiber connector (already attached to the single mode fiber as a patch lead, with sma fiber connectors on each end). There is no alignment necessary between the launcher's female SMA connection, and the fiber's male SMA connector.

The existing 473nm dpss laser has a fairly big diameter collimated beam output, which is just fired into the fiber launcher directly. There are no focusing optics except for the launcher itself. The launcher's angular position (xy tilt) is manually aligned with grub screws to maximise beam coupling into the launcher.

In terms of the 473nm output beam dimensions (before the launcher, of course), imagine the output beam from a hene tube, except blue.

In terms of the 445nm setup I am proposing, since the launcher will be in close proximity to the collimator output, I was hoping that the resultant beam shape would fit within the "light gathering area" (NA) of the launcher.

I think I'll just give it a go and see what happens. The fiber connector end is fairly well protected against damage due to misaligned beam energy heating it up (the flat "launch" end of the connector is all stainless steel, except for the actual fiber part), however I am a little worried about the potential for backreflection returning to the diode, with the obvious issues that could cause.

Pete

[mpoulton]

The existing 473nm dpss laser has a fairly big diameter collimated beam output, which is just fired into the fiber launcher directly. There are no focusing optics except for the launcher itself.
(snip)
In terms of the 473nm output beam dimensions (before the launcher, of course), imagine the output beam from a hene tube, except blue.
Pete

Any 473nm DPSS lab laser will have much higher beam quality (diameter*divergence product, roughly) than these diodes. The optics in the fiber launch take that nice tight beam and focus it into a spot on the end of the fiber. In order for it to work, that spot must be smaller than the end of the fiber core and the angle of the converging rays must be within the acceptance angle of the fiber (determined by its NA). This is relatively easy with a nice clean DPSS beam.

It is not so easy with a diode beam. In order to successfully use this fiber launch with a diode, you need to make the diode's beam look as much like a perfect laser beam as possible - small, low divergence, and symmetrical. If you don't get it right, the light will over-fill the fiber aperture and may melt it. That's going to require some additional optics. If you are trying to launch into a small single-mode fiber, it may not even be possible using your hardware. Fiber launching diodes is hard.

[ogoun]

Bugger.
Oh well, I have an aixiz 3 element glass 405nm collimator on its way, so I'll see what I get from pairing this with the 445nm diode, and make sure I keep the diode power output low... Hopefully I'll get an acceptable amount of output from the other end of the fiber...

Pete.

[Admin]

I am extremely interested to know of your fiber launch efforts...

The way I see it launching these into fiber is the holy grail to 10W output. Expense aside, this could get very interesting. From a purely primal perspective - I just want to see something like that that doesn't involve water,208 3ph, and 500 pounds of hardware.

If you already have the tools at your disposal... have a look at bowtie/panda fiber. Another thing to look out for... try to avoid commodity communications fiber @1550, from what limited information I have gathered they don't look like they would play well.

As mentioned above a function of getting light in and getting it to stick involves "non trivial" beam control and measurements for 62/125 fiber. Hence why the collimated "stripe" produced by these won't play well optically. A circularized beam is critical to get all the angles right and not dump "heat" into the cladding.

That aside: larger diameter fiber would be a cool "cheat" and with a long enough piece of fiber you could get a cleaner but larger beam out of the end coupler.

[mpoulton]

I am extremely interested to know of your fiber launch efforts...

The way I see it launching these into fiber is the holy grail to 10W output.

I am also working on fiber coupling these, using ball and half-ball lenses. I'm shooting for 200 micron .37NA fiber first, and hoping for 100 micron 0.22NA eventually. Now I have a question: What's the best way to combine multiple diodes in a fiber launched system? Would a fiber bundle (say 10 fibers 200 microns each with one diode per fiber, together in an SMA connector) provide sufficient brightness to collimate well, or would the resulting beam be unusable? Would it be more useful to try using cylindrical lenses or prisms to couple an entire array of multiple diodes into a single fiber, as is sometimes done with IR bars? This is a problem I've never solved before in real life.

[andyf97]

I am also working on fiber coupling these, using ball and half-ball lenses. I'm shooting for 200 micron .37NA fiber first, and hoping for 100 micron 0.22NA eventually. Now I have a question: What's the best way to combine multiple diodes in a fiber launched system? Would a fiber bundle (say 10 fibers 200 microns each with one diode per fiber, together in an SMA connector) provide sufficient brightness to collimate well, or would the resulting beam be unusable? Would it be more useful to try using cylindrical lenses or prisms to couple an entire array of multiple diodes into a single fiber, as is sometimes done with IR bars? This is a problem I've never solved before in real life.

With 100um-200um fibers you will have a huge beam or massive divergence, you need to get them into a multi-mode fiber like 25um - 50um for anything usable. You would have a nice round beam but that's all.