I'm interested in directing as much light as possible from the LED into the SMA-terminated optical fibre. What sort of adaptor or bushing would I need to be able to connect the LED to the fibre? I am using Osram SFH 4725S for the NIR LED and a lab-grade patch cord from Ocean Optics.
http://www.osram-os.com/osram_os/en/products/product-catalog/infrared-emitters%2c-detectors-andsensors/infrared-emitters/high-power-emitter-gt500mw/emitters-with-940nm/sfh-4725s/index.jsp
http://oceanoptics.com/product/lab-grade-patch-cords/
Hi Evaline
There are LED adapter to sma from Thor Labs but only applicable for 5mm round lamp packaging. i have seen several research paper use clear epoxy to attach to sma optical probe but signal loss quite high but again thats depend on how you align the led to sma. based on provided LED datasheet it seems like it has quite large half angle which translate to easier alignment to sma. probably best setup is freespace where you can use focal lens to maximize output from led to sma.
Evaline,
What is the core size and numerical aperture (NA) of your patch cord? Your Ocean optics link lists core diameters between 8 microns and 1 mm, and does not seem to mention numerical aperture, which defines the accepance angle for light..
Is the LED mounted on a board, or will you solder directly to it? If you are driving it close to its rated current, you probably need to consider how to conduct heat away from the junction, though this may be less of a concern for pulsed operation.
I assume you are using fibre with a relatively large silica core. With a 0.39 NA polymer clad fibre, mounting the LED lens on the fibre with clear epoxy as Rehman suggests might work. For 0.2 - 0.24 NA fluorosilicate clad fibre, higher angle rays are not guided and with larger core fibres you may do better with a small air gap between the LED lens surface and the fibre.
Thorlabs offer SMA threaded bushes which may be of interest https://www.thorlabs.com/thorproduct.cfm?partnumber=SM05SMA
The Osram LED has a relatively large 1 mm x 1 mm active area. For smaller core fibres (e.g. less than 400 microns) you may get better results from an LED with smaller active area which can be operated at higher current density without thermal penalty.
For a one-off application, a pigtailed LED could be a simpler solution https://www.thorlabs.com/thorproduct.cfm?partnumber=M940F1
https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=69
https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=5206
Hi Evaline, based on my personal experience and having into account that:
- LED is not a punctual source,
-your patch nucleus may vary between 8-1000 microns
I could try a cheap solution by using some lens in order to focus the LED at the fiber optic patch input tip, while monitoring power of light from the other extreme (with some photodiode or power meter). Surely you´ll find an optimum light guiding condition for some x-y-z case (with z being the distance from Lens to input F.O. tip). It often takes time and patience, but finally you´ll arrive to a reasonable solution. If the power into fiber is lower than your requirements, maybe you ´ll need a LED coupled to Fiber (yes,It will be more expensive...)
Good luck!
I am applying free-space laser diodes. For coupling light into an optical fiber (SMF in my case), I am using optical lenses and either a fiber port or a fiber collimator. Thorlabs offers reflective collimators: https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=4093 So collimating the laser beam and using such a reflective collimator might work in your case. But I do not have any experiences with LEDs.
Unfortunately, the "etendue" of LED's prevents efficient coupling into fibers. A large portion of the light generated is contained in very high angle rays that simply cannot be coupled into the fiber. Optics wont help this case as conservation of irradiance says that as the apparent source size is reduced (demag) the NA is increased. Making the source spot small to match the fiber mode size increase the source NA and ends up not being guided in the fiber. Laser diodes have a much smaller NA in both axis and fiber designs have been tailored (or maybe both) to work with each other.
- Badges
- Science topic