// Showcase
For a wavefront measurement with 0.5 nm sensitivity, practically nothing can move during exposure – not even because of the control loop itself.
For a wavefront measurement with 0.5 nm sensitivity, practically nothing can move during exposure – not even because of the control loop itself. A six-axis positioner holding a position isn't sitting still: actuators, control loops, gain – and that's exactly where it can become its own source of disturbance. "The axes are stable enough" was the original claim on the table. Except: how do you actually verify that?
Pretty pragmatically: build an interferometer, look at the fringe pattern, measure the contrast. In four configurations – measuring head close, measuring head far, on the positioner, on the component plate. We didn't estimate the contrast, we calculated it using a min/max analysis on a 2000 × 10-pixel strip. Plus six measurement series with the control loop switched off and six with it switched on – reproducible and comparable.
A number instead of a guess: 29.8% average fringe contrast with the control loop off, 21.5% with it on – a loss of 8.3 percentage points, or 28% relative, and that in every single one of the six series. The reassuring part: the spread within a series stays practically the same in both cases, at around 0.84%. The control loop doesn't make the measurement chaotic, it just makes it predictably worse – and something predictable can be worked around without touching the mechanics.
A test procedure for acceptance testing in four configurations, a solid contrast figure instead of a gut feeling, and derived from that a small but effective tweak to the control software: the interferometric exposure is now timed so it doesn't coincide with an active control loop cycle. Pure software, no extra hardware – and only possible because we knew the effect was systematic, not random.
A claim that was hard to verify became a repeatable test procedure with a number in the acceptance report. And because the interferometer is built entirely from the measurement bench's own components, the machine got a built-in self-test as a bonus – one that can be repeated any time, for example after a service.
Get in touch
You bring the challenge. We develop the solution. From the first idea through concept and development to the validated machine, we guide your project.