// Showcase

Tolerance Analysis of a Fibre Coupling

A collimated beam leaves a device, a small lens focuses it, an optical fibre picks it up.

01Starting point

A collimated beam leaves a device, a small lens focuses it, an optical fibre picks it up. Sounds harmless. But the fibre core is only 100 µm across, and the interesting question is not whether it works - it is how far out of true the whole arrangement may sit before the light stops arriving. In six directions at once: lateral, axial and tilt, for the beam, the lens and the fibre. Without that answer no mechanics can be designed, because five micrometres and forty micrometres are simply not the same part.

02Approach

We had an optical tolerance analysis computed as a non-sequential ray trace, with a source from 440 to 640 nm and a 100 µm aperture standing in for the fibre core. Only one figure counts: the fraction of source power that actually enters the fibre. First 18 catalogue lenses from twelve makers in the 3 to 5 mm focal length band were traced and judged on focus quality; then the chosen candidate was systematically shifted, tilted and defocused. The report arrived in two versions eight days apart, and the second one added exactly one chapter: the same analysis again, with a colour-corrected lens pair.

03Result

That added chapter is what changed the project. With the single lens, efficiency falls from the very first micrometre - a slope on which every tolerance figure is a negotiation. With the achromat, nothing at all happens out to 40 µm of lateral offset, and then it collapses within ten micrometres. A discussion turns into a specification you can inspect against. Just as valuable is the ranking of the six degrees of freedom, which is nothing like the intuitive one. The fibre may be tilted by 5° at no cost. ±100 µm of axial position is free. The beam may wander sideways by tenths of a millimetre. But it may only tilt by 0.5° - at 0.7° there is 11 % left, and from 0.8° nothing. The tightest budget in the system therefore sits exactly where the connector can do nothing about it: in the collimator upstream.

What's inside

A traced lens survey covering 18 candidates with focal spots from 1 µm to 700 µm, a full six-direction tolerance analysis for two optical variants, and the ray-trace models behind it. Plus an honest statement of what the study deliberately does not contain: lens price clarifications were explicitly excluded rather than answered badly in passing.

And what happened next

The numbers from this analysis became the specification for everything that followed - for the search for an affordable single-use ferrule as much as for the question of which part finally absorbs the remaining micrometres. And the report leaves one caveat visible instead of hiding it: the two optical variants were traced at slightly different collimated beam diameters, and a smaller beam does give a smaller spot. That does not change the direction of the result, but it does affect how precisely the plateau widths can be quoted.

All showcases

Get in touch

We are ready for new challenges

You bring the challenge. We develop the solution. From the first idea through concept and development to the validated machine, we guide your project.