+44 (0)1424 225228 sales@torrscientific.co.uk

Customers with demanding optical requirements often specify a high degree of flatness for their viewport windows. In most applications, the ideal window would have no effect on the light passing through it. TSL has engaged a company with proven expertise in the field of optics to help assess the relationship between the physical flatness of our windows and the effect this has on the transmitted light.

A plane sheet of light entering a window at normal incidence should ideally pass through the window and emerge still as a perfectly flat sheet of light and still parallel to the window. This flat emerging sheet of light represents the ideal wavefront. In reality, the window will be less than ideal and any deviation of the emerging wavefront from the ideal wavefront is referred to as transmitted wavefront error (TWE). It is typically represented as a height map and one quantity that is commonly reported is the maximum peak-to-valley distance of this map.

The deviation of transmitted light from its ideal shape can be measured and quantified as transmitted wavefront error (TWE). TWE is commonly reported as a height map relating this deviation to the optical area. The maximum peak-valley distance is commonly reported in fringes or wavelengths (λ or lambda). The power (a non-distorting lensing component) is frequently removed from the measurement, leaving the irregular TWE, containing those components of TWE that are more difficult to correct for than power.

Arbitrary colour scales

There are not yet enough results to make any general performance assessment, but the first few measurements are interesting. The total TWE is between 4 and 13 times smaller than the physical flatness deviation, while the irregular TWE is between 15 and 50 times smaller than the flatness deviation. Interestingly, the pattern of both the TWE and the irregular TWE bear no apparent resemblance to the pattern of the flatness deviations. The surface profile shape on the left (flatness 4½ λ at 633nm) is not apparent in the corresponding irregular TWE on the left (flatness ½ λ).

It might be that despite some inevitable distortion to the optic from the manufacturing process, there remains reasonable parallelism between the entry and exit faces, so that any deviation introduced at the air-optic interface is reversed at the optic-air interface, as illustrated in this highly stylized graphic, representing pairs of wavefronts with attached ray paths passing through a non-flat optic:

Pairs of wavefronts with attached ray paths passing through a non-flat optic

To see if TWE is many times smaller than the physical flatness in our other viewports, we will continue to acquire TWE and flatness data from across our viewport range and share the results and conclusions once a sufficient number of representative measurements have been made.