A new design approach to innovative spectrometers.Casestudy:TROPOLITE.TNO,Stieltjesweg 1,Delft,Netherlands;ABSTRACTDesigning a novel optical system is a nested iterative process.The optimization loop,from a starting point tofinal system is already mostly automated.However this loop is part of a wider loop which is not.This widerloop starts with an optical specification and ends with a manufacturability assessment.When designing a new spectrometer with emphasis on weight and cost,numerous iterations between theoptical-and mechanical designer are inevitable.The optical designer must then be able to reliably produceoptical designs based on new input gained from multidisciplinary studies.This paper presents a procedure that can automatically generate new starting points based on any kind ofinput or new constraint that might arise.These starting points can then be handed over to a generic optimizationroutine to make the design tasks extremely efficient.The optical designer job is then not to design optical systems,but to meta-design a procedure that produces optical systems paving the way for system level optimization.We present here this procedure and its application to the design of TROPOLITE a lightweight push broomimaging spectrometer.Keywords:optical design,imaging spectrometer,starting point,yybar diagram,plate diagrams,optimization1.TROPOLITE1.1 ScopeA push broom imaging spectrometer for earth observation contains six functional optical elements as shownin figure 1.Its measurement principle is described in figure 2.In this study we focus on investigating all-reflective solutions for the ensemble collimator,imager for a UVIS channel (320-495nm).It is common touse all-reflective telescope designs,!however the ensemble collimator,imager is commonly realized via severalchannels of catadioptrics systems.Advantages of all-reflective solutions are:detectorimager/gratingtelescopeslitcollimatorFigure 1.Functional layout of a push broom spectrometer in spectral plane.Optical elements are represented as lensesfor sake of clarity.The slit is rectangular such as imaging a line on earth (the longer slit dimension is out-of plane inthe figure coordinate system).The image on detector is two dimensional with one dimension corresponding to samplinglocations on earth (out-of-plane in the figure)and the other dimension corresponding to sampling locations in spectrum.Optical Modelling and Design Ill,edited by Frank Wyrowski,John T.Sheridan,Jani Tervo,Youri Meuret.Proc.of SPIE Vol.9131.91310L.2014 SPIECCCc0de:0277-786XW14/s18·doi:10.1117/12.2051596Proc.of SPIE Vol.9131 91310L-1swathflightdirection-7 km2600kmFigure 2.Measurement principle of a push-broom imaging spectrometer.1Design can be copied unchanged for other channels (with the exception of the grating).Fewer optical elements due to the insensitivity to chromatic aberrations.More design freedom regarding stray light baffling due to the off-axis nature of such a design.This can lead to a lighter and cheaper instrument.1.2 RequirementsTable 1.Driving requirements.F#spectral16fieldslitspectral0.1mmfield24.0mmspot size (RMS)0.013mmco-registration0.003mmMax Volume1.5LThe co-registration is defined as the peak to valley straightness of the image at detector plane of a point onearth.The volume requirement is the most stringent and was the motivation to explore new methods capableof finding compatible starting points.2.NEW DESIGN APPROACH2.1 Theoretical backgroundOur new design approach makes use of well known techniques such as y-g diagrams,2 Seidel aberration theoryand plate diagrams that we revive by combining them together,and with modern computing technology.The link between y-g diagrams and Seidel aberrations is not new4 but we will also show that unsurprisinglyy-g and plate diagrams are also connected.2.1.1 y-y diagramA y-y diagram is a paraxial representation of an optical system.It consists of the projection of an off-axis ray,marginal in the sagittal plane ("skew"ray see figure 3)on the image plane as shown in figure 4.y-y diagrams have many interesting properties.The most important ones for our following discussion are:Proc.of SPIE Vol.9131 91310L-2Downloaded From:http://proceedings.spiedigitallibrary.org/on 07/11/2014 Terms of Use:http://spiedlorg/termsFigure 3.The famous Cook achromatic triplet,seen in perspective with only sagittal rays drawn for two field points.Thetwo "skew"rays are indicated.Figure 4.This is the same system as in figure 3,seen from the back and rotated around its optical axis (z).The projectionon image plane of one of these "skew"rays is a y-diagram.Each change of direction (indicated by a vertical mark)corresponds to an optical surface.We choose one of the rays out of the two in order to fulfill the convention that y-diagrams propagate in the clockwise direction.Proc.of SPIE Vol.9131 91310L-3Downloaded From:http://proceedings.spiedigitallibrary.org/on 07/11/2014 Terms of Use:http://spiedlorg/terms