Hi, welcome to Thorlabs. My name is Bill. Today I'll be talking about aligning fiber
collimators to create some free space between single mode fibers. For today's demonstration, I have a red benchtop
fiber-coupled laser, identical fiber collimators in matched kinematic mounts, some single mode fiber in the yellow patch cables, some multimode fiber in the orange patch cable,
and a power sensor, and a power meter. Now, before we begin, I'd like to talk a little
bit about collimation packages. In theory, you could take a point source of light, place it at the back focal point of a lens, and create a perfectly collimated beam that will maintain that same beam size all the way out to infinity. But, when you get into the lab, you'll find that there's some physical limitations as to how far you can maintain a similar beam size. When we talk about collimators, we generally
talk about an output beam diameter, and that output beam diameter is really a beam waist. It's the minimum beam size that will result
from that collimator. So when the light is output from the collimating lens, it will converge slightly to that beam waist. It'll diverge at the same rate, until it hits a point where it will start
to diverge relatively quickly. And so, one way to characterize how far you
can maintain a similar beam size is to calculate the Rayleigh range. And what the Rayleigh range will tell you
is how far it'll take for the light to expand an additional 41% of that beam waist. Now, not all collimators are made the same. Some will place the beam waist directly at
the output housing, and so the maximum distance that you can maintain
a similar beam size would be one Rayleigh range. Other collimators push the beam waist out beyond the housing, and so the maximum distance that you can maintain
a similar beam size would be two Rayleigh ranges. So, depending upon your application, you may have to weigh the tradeoffs between a certain beam size and the distance that you can maintain that similar beam size. Now, for this demonstration, I've set up a
random amount of free space between my two fiber collimators, and I'll be showing the adjustments to the optomech in order to couple as much power from free space into my second fiber. So, the first thing I do is try to minimize
the number of degrees of freedom. So, you'll notice that I've taken my post
holders and I've threaded them directly into the table, and this is to align my fiber collimators with a row of tapped holes in the table. Next, I've taken identical post holders, posts,
kinematic mounts, and adapters, and I've taken my posts and put them all the
way down to the bottom of the post holders, and this was to try to get the fiber collimators
to be exactly at the same height. So in the end, all I'll be doing is adjusting
the pitch and the yaw adjusters on the two kinematic mounts in order to maximize my coupling. Now, minimizing the degrees of freedom will
ease your alignment, but it also improves the stability, as there's
much less things to move and drift over time. So to begin, I'm going to move the ruler out of the path, and I'm going to use the first kinematic mount
in order to align the beam with the center of the fiber collimator. Once I feel that I'm pretty close, I'll take
my power sensor, and I'll bring that into the path, and this will give me a reference as to how much power is incident upon my second fiber collimator. So once I have that recorded, I'm going to
attach an adapter to the power sensor so that I can couple my fiber directly. Now, you might be wondering why I'm using multimode
fiber when we want to use single mode fiber in the end, and that's another step that you can take in order to ease your alignment. It'll be much faster to couple the light into the larger core. And once you have it, you should be relatively
close to the alignment for the single mode fiber. So now that I'm fiber coupled, I can see that
I have some power, and so I'll use the two adjusters in order
to maximize the power. So I'm going to travel through a relatively large travel range in order to ensure that I don't accidentally try to peak up on a local maximum. So once I think I have a maximum with one
adjuster, then I'll move over to the other, and I'll go back and forth in order to maximize. And once I think I have something that's pretty
much at the maximum, then I'll switch over to my single mode fiber. So, I'll block my beam before I disconnect
my fiber connectors. Now you might notice that I have an APC connector
on the single mode patch cable, and I'll be connecting that end to my power sensor. And that's because there is a high Fresnel
reflection between the power sensor and the free space, and so that should prevent any back reflections from going
back into my system and causing instabilities. Another thing that's important is always to
clean your connectors. You don't want to waste a whole lot of time trying to align something when it was really just a dirty connector that was preventing you from getting the maximum coupling. So once I'm connected, open that up, and you
might notice that the boot of the connector is actually glowing. And some people will actually use that as an indicator
to help them align their fiber, and so what you're actually seeing is that some light has coupled into the cladding and is leaking out of the fiber and through the patch cord. And so I'll use my adjusters and look to maximize the power. And here, because we started with the multimode
fiber, we're already up into the milliwatt range relatively quickly. And now, once I have maximized the power with
just this kinematic mount, now is when the real work begins. And so, what I will do is, I'm going to actually
misalign and then compensate on the opposite side. Misalign again. And then compensate. Misalign. And compensate. And you can see that we're actually slowly
walking the power up. Now, there's actually two ways to approach this. You can use one side to misalign and the other
to maximize the power. Or you can actually misalign both sides. So, I'll go past the maximum on one side,
and I'll go past the maximum again. Past the maximum. And you can keep stepping this way until you
don't pass a maximum point twice. OK, so now I'm seeing that I'm not hitting
the same maximum that I was before, so I'll actually come back, past the maximum,
and see if I can get back to that largest number that I saw previously. So I come back the other way. There we go. And now that I'm getting what I feel is more
or less the maximum on that one axis, now I'll shift to the other. So here again, I will misalign, and I'll see
if there's a direction that gives me more power, and if there is, then I'll use that direction as my adjustment. OK, so now I've gone past that maximum point,
so I'll come back. And then you can iteratively go back and forth
between the two axes until you're able to couple as much power as you can, or at least enough power for your application. Hopefully this helps you out in your lab someday, and if you have any questions, feel free to contact tech support.
