




版權(quán)說(shuō)明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請(qǐng)進(jìn)行舉報(bào)或認(rèn)領(lǐng)
文檔簡(jiǎn)介
1、Advanced Imaging System Design Using Zemax | OpticStudio e!1 - 2 e!This five-day class is intended for users who need a thorough understanding of the advanced sequential analysis and optimization features of ZemaxIt covers all the major features needed for taking an imaging or afocal design from ini
2、tial specification to manufacture and testThe focus is on Using Zemax so we will only go into theory when its absolutely neededIn most cases we will simply state the relevant equations, and not derive themDetailed derivations are provided in all the classic texts1 - 3 Good Theoretical BooksGood theo
3、retical books include:Principles of Optics by Born & Wolf published by Pergamon PressSerious math!Aberrations of Optical Systems by Welford published by Adam Hilger Ltd.The Art & Science of Optical Design, by Robert Shannon, published by Cambridge University PressSome computer orientation, lots of e
4、xamplesWarren Smiths two books, Modern Optical Engineering and Modern Lens Design, published by McGraw-Hill1 - 4 Good Practical BooksGood practical books include:Lens Design, by Milton Laikin, published by Marcel DekkerPractical, experience, real-world, included with ZebasePractical Computer-Aided L
5、ens Design, by Gregory Hallock Smith, published by Willmann-BellExcellent use of computer techniques Uses Zemax for many examplesOptical System Design, by Bob Fischer and Biljana Tadic-Galeb, published by McGraw-Hill1 - 5 Table of Contents: Day 11-1: e1-12: The Optical Design Process1-26: Laser Beam
6、 Expander Design Example1-58: Sampling & Reference Points1-72: Merit Functions1-75: The Local Optimizer1-83: Default Merit Functions1-92: Operands1-105: Design Rules for Merit Functions1-117: Global Optimization1-123: Candidate Selection1-146: Design Refinement1-160: Thermal Effects in Optical Syste
7、ms1-183: Diffractive Optics1-198: Thermally Compensated Diffractive Lenses1 - 6 Table of Contents: Day 22-1: Diffraction, MTF, and Image Quality2-12: Fraunhofer Diffraction2-33: Computing MTF2-40: Double Gauss Design2-64: Huygens Diffraction2-76: Image Simulation2-103: Source Coherence and Imaging2-
8、135: Polarization Ray Tracing & Thin-Film Coatings2-182: Birefringence2-200: Ghosts1 - 7 Table of Contents: Day 33-1: Laser Beam Propagation3-4: Gaussian Beams3-16: An Example with Gaussian Beams3-39: Skew Gaussian Beams3-44: Physical Optics3-46: Overview of POP3-68: The POP Calculation3-91: Managin
9、g the POP Calculation3-104: POP Example A Spatial Filter3-136: Quantitative Beam Analysis3-161: Assumptions and Special Settings3-174: Fiber Coupling1 - 8 Table of Contents: Day 44-1: Multi-Mode Fiber Coupling Efficiency4-47: Single-Mode Fiber Coupling4-74: Design for Manufacture4-78: Error Sources
10、and Budgets4-87: Tolerancing Overview4-100: Tolerance Operands4-153: Compensators4-161: Tolerancing Procedure with Zemax4-182: Tolerancing a Singlet1 - 9 Table of Contents: Day 55-1: Tolerancing Irregularity5-23: User Defined Tolerances and Compound Groups5-30: Tolerancing in Double Pass5-73: Script
11、ed Tolerancing5-91: Tolerancing the Cooke Triplet5-107: Tolerancing a Fiber Coupler5-124: De-Sensitization5-132: Speeding Up Tolerancing5-153: Conclusion1 - 10 AdministrationStart & end times8:30 start each day4:30 finish each day (3:00 on Friday)Lunch & breaksCoffee breaks at about 10:00 and 3:00Lu
12、nch at about 12:00-1:30Please ask questions!These notes are copyright Radiant Zemax, LLC; please do not copy more than 2 pages without written permission1 - 11 Pre-RequisitesThis is an advanced class, so we expect attendees to already be familiar with both Zemax and optical design theoryIdeally you
13、will have taken Optical System Design Using Zemax prior to this class, or have equivalent knowledgeSome prior experience of practical lens design is also required to get full benefit of all the material we will coverNote: At this point the trainer will go through the various options to set up the Ze
14、max user interface on his/her machine for maximum visibility for the attendees. Feel free to do likewise with the installation on your machine.Advanced Imaging System Design Using Zemax | OpticStudioThe Optical Design Process1 - 13 Optical Design ProcessBroadly speaking, you can structure the optica
15、l design process into five steps:Specification and test planningConcept selectionDesign, optimization, analysisTolerancingAssembly, test, and post-production analysisIn a well-managed (or lucky!) project you should only have to go back one stepFor example, after optimizing and analyzing a candidate
16、design (step 3) you might have to reject it and generate a new candidate (step 2)But it would be a bad thing to have to go from step 5 to step 1!1 - 14 SpecificationThis has the least Zemax-specific content (and in fact should have nothing to do with Zemax), but as the optical designer you should be
17、 thinking about how the specification is implemented as both a merit function for optimization and as a tolerancing targetSpecifications fall into two main categoriesThe optical performance the design must achieveConstraints the design must meet, typically length, weight and costIt is vital that you
18、 understand how the optical system will be tested, as this will drive your choice of merit function more than anything else 1 - 15 Default Merit FunctionsIf the system will be tested on an interferometer, or an MTF test rig, then usually the RMS Wavefront Error, relative to the centroid, will be an
19、important part of your merit functionIf the system will be tested with a CCD array, then RMS Spot Size, relative to the centroid, is more relevantFor afocal systems it is vital to understand if the test method will use the afocal output directly, or if the output will be brought to focus by an auxil
20、iary optical system, or both1 - 16 Cost, Quantity, and AdjustabilityThe target cost and quantities are important drivers of the design process, as many options can be eliminatedGenerally, the lower the per-item cost, and the higher the quantities, the fewer adjustments are available in production, s
21、o the design must be built and just workThink of fixed-focus mobile phone lenses for example, 95% of lenses should have 65% MTF, no individual lens should be less than 60%” is a different specification to “MTF must be greater than 60%”1 - 18 Translate the SpecificationYou should be able to translate
22、 the specification, and the test methods, directly into Zemax optimization operandsUltimately the specification es a merit function, and the test method es a tolerance criterion, merit function or tolerance scriptSo before you even start the design you should know what the merit function is going to
23、 be!1 - 19 Concept SelectionAt this stage, you are roughing-out what the design might look likeTypical activities:First order analysis using paraxial surfacesReview of established literatureZebase is a very useful resource for any kind of imaging system designGlobal SearchWell cover all of these in
24、detail in the classAt this stage we should not be too rigorous about applying the final system specificationWe are looking for candidate designs to perform detailed analysis and optimization on1 - 20 Candidate DesignsWhatever method of candidate design selection you use, remember that we are roughin
25、g-out the designYou will probably only use one wavelength at this stage, as color correction is strongly dependent on glass choice, which is influenced by design shape, and were trying to establish the design shape at this stage!Typically only use default merit functions with keep it real constraint
26、s, and only the most important final system constraintsF/#, total length, etc. The most common Zemax tool at this stage is Global Search1 - 21 Global SearchBefore starting, you must specify the number of optical surfaces to useThis may be defined in the specification, and cost constraints usually im
27、ply that fewer is betterGenerally we start with as few surfaces as possible, and increase the number of surfaces only if good candidate designs cannot be foundThe initial design should be at least roughly in focus and have roughly the correct focal length, because Zemax uses the starting focal lengt
28、h as a scaling parameterParallel plates with an F/# solve on the last surface is all you needGlobal search will rarely find the global optimum by itselfGlobal Search concentrates on finding new, promising design forms rather than converging exactly on the best possible solution for each form1 - 22 D
29、esign and OptimizationThis is the stage where the detailed work is doneThe most appropriate Zemax tools are:Damped-Least Squares (local) optimizationHammer optimizationWhatever analysis features are appropriate1 - 23 TolerancingTolerancing means we account for finite manufacturing accuracy in the ma
30、nufacture of the components and their assembly into a systemZemax provides many tools for this, from simple perturbations to complete, scripted tolerancing of the entire assembly processThis is the least creative, most hard-work part of the processAnd is consequently one of the most neglected!But it
31、 is essential for any kind of volume production1 - 24 Post-ProductionOnce a prototype has been built, or the lens is in production, how close is it to its design goals?A good review process that signs off each stage of the process before work starts on the next stage is one of the most important QA
32、processes you can haveInvolving the likely manufacturer(s) and assembly staff as early in the process as you can is also a good way to avoid unexpected problems1 - 25 Course OverviewThis course will cover all aspects of this procedure, but will focus on steps 2, 3 and 4 as these are the most Using Z
33、emax sectionsWe will do several examples all the way through, and others where we focus on a specific part of the processPlease ask questions as we go along!Advanced Imaging System Design Using Zemax | OpticStudioLaser Beam Expander Design Exercise1 - 27 Beam ExpanderOur goal is to design an optical
34、 system that expands a 5 mm diameter He:Ne red line beam to 25 mmIt should be as short as possibleWell agree that a 250 mm length is OKThere must be 5 mm clearance on either side of the expander, so we actually only have 240 mm to work withIt should be as good as possibleIt should produce an output
35、wavefront flat to l/10, and will be rejected if worse than l/5It should be as cheap as possibleOnly a few will be made, and they will be mounted on an optical bench, so positions are easily adjustableDesign using catalog optics. Lets say we have a preference for using CVI Melles Griot lenses, but we
36、 could use others if necessary (We will be using Sigma Koki and Thorlabs, due to recent updates to the CVI-MG catalog).1 - 28 Test PlanThe expander will be tested on an interferometerZemax features to use: OPD plot, InterferogramThe magnification will be measured using an aperture that passes 1/e2 o
37、f the output beam. There should be a ratio of 1:5 between these apertures on the input and output beams.Zemax feature: Encircled Energy1 - 29 Candidate DesignsBeam expanders fall into two main typesGalilean, which does not have an intermediate focusKeplerian, which does have an intermediate focus1 -
38、 30 Candidate DesignsThe Galilean version requires lower-powered optics than the Keplerian for a given overall lengthSince length is restricted in our spec, well choose the Galilean design as lower power introduces less aberrationAlso, intermediate foci are not a good thing in a laser beam expander
39、because if it is used with high powered lasers a spark can be caused at the focal pointNot a likely problem with He-Ne lasers, but it is still better to avoid intermediate foci just in case the expander is ever used with say a Ruby laser 1 - 31 Candidate DesignsSince we want to use catalog optics to
40、 reduce costs, there are two basic options:Plano-curved lensesAchromatic doubletsIt may sound odd to use achromatics in a laser lens, but the extra surface is very useful for controlling spherical aberrationSince we have a lowest possible cost driver (when do we not?) well start off initially with p
41、lano-curved lenses and only switch to the more expensive achromatic lens types if we cant get the optical performance we needNote the first plano-curved lens must not have its plano side towards the laser, as this will reflect light straight back into the laser cavity1 - 32 Our Candidate DesignA lit
42、tle back-of-the-envelope consideration gives us our starting point:A Galilean arrangement (negative and positive lenses)Plano-curved lensesFirst lens has its curved face towards the laser Now to enter it into Zemax!Here are the initial requirements:5 mm dia input, 25 mm outputCorrected for best OPD
43、at 0.6328 (He-Ne)PTV wavefront 0.1 waves or less, 0.2 waves is a fail5 mm clearance on either side of the expander, total 250 mm length or shorter (including the clearance)1 - 33 SetupUse:Entrance Pupil Diameter = 5 mmAfocal ModeAdd a 1 mm margin to all semi-diameters to ensure there is some dark gl
44、ass for mounting against1 - 34 SetupSet the wavelength to 0.6328 microns and then enter surfaces like so Note I have made both radii of the second lens variableWe dont yet know which is the best to make planoWell get that in a second pass1 - 35 Initial DesignYou should have this:The y-stretch is use
45、ful because this optical system is much longer than it is high!1 - 36 Initial ValuesIt is a good idea to use values like +1, 2, 5, 10, 20, 50, 100, etc. (plus INFINITY for plano surfaces) as starting values for radii and thicknessThis gives a starting point that is a reasonable place to start1 - 37
46、Merit FunctionSince we want wavefront flatness, optimize for RMS wavefront errorAdd reasonable boundary conditions on glass and air thicknesssWell come back to rings and arms later: just enter as shown for now1 - 38 ConstraintsSince we want the output beam to be 5x the input beam diameter, use a sin
47、gle ray at the edge of the ray bundle and target its y-coordinate to be 12.5 on the image surfaceREAY (marginal ray) target = 12.5, weight = 1 We also want to total length to be 2But theres no operand for shape factor!So the next page shows an example of how to write one in your merit function1 - 10
48、4 Complex OperandNote use of zero weightsZemax computes these operands but they dont affect the value of the merit functionAdvanced Imaging System Design Using Zemax | OpticStudioDesign Rules for Merit Functions1 - 106 Can the Singlet be Improved?Reopen the sample file SamplesShort coursesc_singlet_
49、end.zmxOn axis, the limiting aberration is spherical. Can we get a better design by eliminating the spherical?Try adding an SPHA constraint. SPHA computes the third-order spherical aberration in waves.SPHA starts at 3 waves. Optimize!1 - 107 SPHA OptimizationSPHA drops to 1.08 waves. The MF drops, b
50、ut RMS spot size goes way up! It increased from 25 to 35 m across the FOV to 28 to 85 m!1 - 108 Why Cant We Fix the Spherical?Which variable controls spherical?Stop position is irrelevant, spherical appears on-axisBack focus is irrelevant: focus and spherical are independentLens thickness is a weak
51、variable and at its limit anywayLeaves just the front radius, which must compensate on and off axis aberrationsRequirement for minimum RMS spot and minimum spherical contradict, as focus and spherical balance for minimum spot sizeSome variables are ineffective at reducing certain aberrations, and so
52、 reduces effective degrees of freedomAberration reduction and spot-size reduction often conflictAberration balancing is key1 - 109 Moral of the StoryZemax always finds a lower merit function, but the merit function must represent what you really care about!Lower merit functions are only “better” by
53、the criteria they define!Optimizing for specific aberrations is not usually desirableZemax uses physically significant merit functionsBuild a merit function that describes how your lens is used or tested1 - 110 Residual AberrationOnce a design is well optimized (see global optimization section, comi
54、ng up), what were left with is the residual aberration of the designCannot improve design by adding more constraints in the merit function“Toothpaste tube syndrome”In order to improve design further, need toLoosen constraints (not add!)Add further degrees of freedomExtra surface, asphere, glass choi
55、ce1 - 111 Building Your Merit FunctionImportant decisionsWhat optical performance criteria is appropriate?RMS spot sizeRMS wavefront errorMTFEncircled EnergyWhat are the constraints?Min/Max length, edge thickness, center thicknessMax weightMax cost (# of elements, aspheres, material)What other prope
56、rties should the system have?EFL, F/#, FOV, Vignetting, etcCan I use a solve?1 - 112 SolvesSolves are built-in spreadsheet functions that set a value based on some calculationThe most common ones are pickups, F/#, and marginal ray height solvesIf a solve exists to compute a constraint, it is usually
57、 more efficient to fix a value with the solve than to let it be variable and control it with targets in the merit functionWe would always use a pickup solve to design an equi-vex lens instead of requiring the two curvatures to be of equal and opposite sign!Constraints can be your friend, as they lim
58、it the range of possible designs you need to consider 1 - 113 Determining the GoalsPerformance criteria:RMS spot sizeBest for systems with more than 2 waves of aberrationFastestRMS wavefront errorBest for systems with less than 2 waves of aberrationNearly as fast as spot sizeMTF, encircled energyGiv
59、es results similar to wavefront errorSlowerBest used when the design is maturing or close to finishedPTV spot, wavefrontMinimizes circle of least confusion1 - 114 Overall WeightWhen switching between different default merit function types, such as RMS spot radius and RMS wavefront error, the numeric
60、al magnitude of the default merit function can change dramatically This may make it tedious to manually adjust the weights of operands that are not part of the default merit function The “Overall Weight” is simply a factor that scales all the weights in the default merit function Under most circumst
溫馨提示
- 1. 本站所有資源如無(wú)特殊說(shuō)明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請(qǐng)下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請(qǐng)聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁(yè)內(nèi)容里面會(huì)有圖紙預(yù)覽,若沒(méi)有圖紙預(yù)覽就沒(méi)有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫(kù)網(wǎng)僅提供信息存儲(chǔ)空間,僅對(duì)用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理,對(duì)用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對(duì)任何下載內(nèi)容負(fù)責(zé)。
- 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請(qǐng)與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時(shí)也不承擔(dān)用戶因使用這些下載資源對(duì)自己和他人造成任何形式的傷害或損失。
最新文檔
- 2025房屋所有權(quán)擔(dān)保借款合同
- 2025年合同終止與解除勞動(dòng)合同的賠償標(biāo)準(zhǔn)一致嗎
- 2024年文化產(chǎn)業(yè)管理考試的新發(fā)展試題及答案
- 江蘇省連云港市本年度(2025)小學(xué)一年級(jí)數(shù)學(xué)部編版質(zhì)量測(cè)試((上下)學(xué)期)試卷及答案
- 提升專利申請(qǐng)質(zhì)量的2024年考試知識(shí)試題及答案
- 分析2024文化產(chǎn)業(yè)管理證書(shū)考試的試題及答案趨勢(shì)
- 文化產(chǎn)業(yè)管理證書(shū)備考試題及答案
- 激光工藝參數(shù)優(yōu)化試題及答案
- 2025鄉(xiāng)村全科執(zhí)業(yè)考試成功故事試題及答案
- 吉兆衛(wèi)生管理核心試題及答案
- GB/T 26354-2025旅游信息咨詢服務(wù)
- 2025年中國(guó)工業(yè)X射線檢測(cè)設(shè)備行業(yè)市場(chǎng)集中度、企業(yè)競(jìng)爭(zhēng)格局分析報(bào)告-智研咨詢發(fā)布
- 職工維權(quán)知識(shí)培訓(xùn)課件
- 《制作七巧板》教學(xué)設(shè)計(jì)-2024-2025學(xué)年五年級(jí)上冊(cè)勞動(dòng)浙教版
- 2024銀行春招招聘解析試題及答案
- 四川達(dá)州歷年中考作文題與審題指導(dǎo)(2004-2024)
- 第15課《青春之光》課件-2024-2025學(xué)年統(tǒng)編版語(yǔ)文七年級(jí)下冊(cè)
- 獨(dú)立儲(chǔ)能電站項(xiàng)目運(yùn)維管理方案
- 青島市2025年高三語(yǔ)文一模作文題目解析及范文:成見(jiàn)與主見(jiàn)
- (二模)晉中市2025年高三高考二模 語(yǔ)文試卷(含A+B卷答案詳解)
- 2020年1月浙江省普通高校招生選考科目考試政治試題及答案
評(píng)論
0/150
提交評(píng)論