Sensor startup to cut curved image sensor costs

May 03, 2021 // By Peter Clarke
Sensor startup to cut curved image sensor costs
Startup Silina in Paris, France, has created a curved image sensor from conventionally produced planar sensors, supporting multi-die manufacturing and automation.

The use of curved imaging sensors can improve image quality and light transmission, bulk and cost by simplifying the lens systems used to correct for planar focal plane sensors. However, up until now it is not clear that the advantage has been worth the additional cost in sensor manufacturing, particularly for consumer applications. Michael Bailly and Wilfried Jahn point out that this is mainly because the manufacturing processes developed so far by the various players are manual rather than automated and for single chips at a time. Silina's process supports multi-die manufacturing and the process can be automated.

This follows French company Curve-One SAS founded in 2018 and research at Sony in 2014.

However, whereas Curve-one plans to produce curved sensors for specialized applications, such as scientific instrumentation, Silina is adopting a different business model. Silina plans to provide a direct "curving" service for small volume, niche customers and to license its curving process to high volume manufacturers.

Silina is offering to work with established manufacturers to produce curved versions of existing sensors for improved imaging. The company offers a curving process that it claims is applicable to various types of sensors, including CMOS, charge-coupled devices (CCDs), and front- and back-side illuminated sensors, to various spectral bandwidths, from ultraviolet (UV) to visible and infrared. The curving process works at various scales, including single-chip, multi-chip and wafer-level. And various shapes can be obtained, including spherical, aspherical, freeform, and custom shapes on demand.


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.