Lightbox - BLG • Vantage
Contents
1. unig urip umg untzz witgp uig wip wng wipe o wrigz uige wgp umg wip wigi wrigz unige uri L ual wig EBSD Based Strain Measurement Software User Guide for the CrossCourt 3 Strain Measurement Tool Contents 1 Before You Start 1 1 Software Specification 1 2 Computer Requirements 1 3 Installation Instructions 1 4 Notes for Users of CrossCourt 2 1 5 Overall Description of the Software 1 6 Practical Considerations when collecting EBSD patterns 2 Getting Started 2 1 Loading a Project 2 2 General Software Layout 3 Single Crystal Case 3 1 Project Page 3 2 Materials 3 3 Thresholding 3 4 How to Select and Display a Reference Pattern 3 5 Pattern Display 3 6 Regions of Interest 3 7 Filter Settings 3 8 Process Panel 3 9 Results Pages 3 10 Display Page 4 Polycrystalline Case 4 1 Finding Grains 4 2 Grain Tolerance Angle 4 3 Find References 5 Additional Functions 5 1 Right Click on Map context Menu 5 2 Right Click on Linescan Context Menu 5 3 Menu Items 5 4 Beam Shift Calibration 5 5 Load Folder BLG Productions Ltd 2010 sia 6 Matlab Data Structure for CC Analysis Section 1 Before You Start aes Wiel W 1 1 Software Specification The software produced by BLG productions and titled Cross Court 3 provides a means of measuring small distortions and rotations of electron backscatter diffraction patterns obtained from near perfect crysta2. as read from project file indexed as ypts xpts If no data exists for a particular data type the information is not included Euler angles are in degrees BLG Productions Ltd 2010 39 Project Details continued filters 1x4 array of doubles Contains filter settings high frequency cut off high frequency cut off width low frequency cut off If cut off width refloc Array of Strings Each array element is a string containing the name of the reference image followed by the position of the reference as a number from 0 to numims version String The version of Cross Court used voltage Double SEM voltage in KV date String Date of Processing projectname String Name of original Project File scanformat String Description of Scan format i e Rectangular or Linescan imagedir String Path of Directory containing EBSD images Results Results are only exported if they have been calculated Variable Name Type Description shiftsx shiftsy peakhgts corrshiftsx corrshiftsy Cell array of 3 Dimensional arrays of doubles Cell array indexed as refnum containing results of the cross correlations in a set of 3D arrays Each is indexed ypts xpts roinum dataph 3 dimensional array Indexed as ypts xpts refnum and containing of doubles geometric average of normalised cross correlation peak heights of all ROI dataerror 3 dimensional array In
3. ROI within two EBSD patterns one obtained from a reference point in the sample and the other from a test point Obviously if the crystal contains highly disoriented regions where the test point is rotated with respect to the reference point by an amount greater than the angle subtended by the ROI then the content of the ROI will be completely different in the reference and strained cases and hence cross correlation cannot be used to compare them Thus as the rotation between reference and test pattern increases the amount of overlapping pattern decreases as does the quality of the cross correlation This is illustrated in the two images above The amount of overlap between the two images is small thus the cross correlation quality suffers The maximum angular difference i e the GTA suggested by the software is about 1 4 of the total angular range across the ROI When the Find Grains button is clicked the Grain Tolerance Angle dialog is shown with this suggested value the user is free to alter the values if thought necessary If the find grains button was not activated then the set Angular tolerance window only become active once an ROI has been selected in the project display window and this in turn requires prior selection of an EBSD pattern This is done by clicking on one of the maps shown and clicking show in the reference list Clicking the Set Tolerance button will launch a dialog box in which a maximum sa
4. distinguished in the maps by a cross within the data point If the first selected reference cannot be used for the whole grain additional references will be selected It is still necessary to set the Boundary ROI and Filter settings as for the single crystal case Once the references are selected continue to determine the strain by selecting Process and progressing to Do XCF Once the Grain Tolerance Angle has been set the behaviour of the software changes slightly The data for each reference is processed separately References from the same grain are combined as described above once calculated After each grain reference is processed the Data Quality maps are displayed updated If the Use Batch Mode tick box on the Project panel is ticked the results for each reference are saved to disk upon calculation The results are not displayed but can be reloaded using the load BLG Project option After processing the Combine Results button appears on the Process Panel Clicking it transcribes all the data onto a single data set This is purely for display purposes only as comparisons between different references are not strictly valid though misorientation and gradient information is relevant Please note that when you change the data e g displaying in Degrees rather than Radian changing the pattern centre elastic constants etc you will need to re press the Combine Results button to update the combin
5. include as much of the pattern as possible yet avoid the edges of the phosphor which often contain micro scratches which would affect the determination of the displacement tensor BLG Productions Ltd 2010 17 The Animate button will cause the software to loop through all the EBSD patterns collected and is useful for visually assessing their quality This process can be stopped using the Abort button The Filter Settings button brings up the Filter Setting Dialog see Section 3 7 for more information Clicking the Add ROI button starts the process of adding Regions of Interest 3 6 Regions of Interest A Region of Interest ROI is an area of the pattern used to perform a single cross correlation calculation The size of the ROI is set using the drop down box at the top left The sizes are limited to multiples of 2 e g 128 256 etc so that the Fourier analysis can be fast There are 3 ways to set the Regions of Interest All of them start with clicking the Add RO button within the Pattern Display Tab 1 manually left click in the pattern Make sure that the corners of the resulting square are all within the pattern It is best to avoid scratched areas of the phosphor when doing this 2 automatically by selecting a number of ROI These are positioned within the boundary defined with the New Boundary button The ROI will be automatically resized if
6. measure of the distortion of a pattern with respect to the reference pattern This is done with the pattern displayed as set out next BLG Productions Ltd 2010 16 3 5 Pattern Display The EBSD patterns can be displayed in several ways Press the Pattern display tab at the top of the screen right hand panel It can be called from the Reference list by clicking the show button Alternatively right click on a particular map pixel and choose Show Pattern from the context menu A section of the page is shown below X 597 394 Use the buttons to perform stuff Right Clicking on the pattern shows a context menu with 2 options One allows you to save the pattern as a bitmap The second toggles the display of Regions of Interest The pixel position of the mouse is shown at the top of the page to the left of the file name of the image shown A set of buttons are shown below it ROI size New Boundary Animate Filter Settings and Add ROI Overlaid on the pattern are the pattern centre purple star and the current phosphor boundary red circle We Start with setting a new boundary A new boundary can be set with the New Boundary button Click it and then click on 3 points on the boundary The boundary is primarily used to control the region used in the automatic placing of Regions of interest ROI around the phosphor see below It should be drawn so as to
7. needed to determine the displacement tensor selecting BLG Productions Ltd 2010 18 more means that the tensor can be determined numerous times and from all the results an average value can be obtained Further statistics are also carried out and are used later to assess the quality of the reliability of the result It is also possible to select ROIs manually The movement of particular zone axes can be followed using this option Scratches on the phosphor can also be avoided by careful placing of the ROI 3 7 Filter Settings Click the Filter Settings button Select an ROI towards the centre of the EBSD pattern and adjust its size so that it covers is alae aaa aaa os ee anid 1 3 of the image width The High Frequency Cut Off Width 16 left hand panel will now change to show the Use Filtering filter settings Select an ROI again referably centred on a zone axis in the a Saint where many bands cross Both the Fast Fourier Transform FFT image and the back transform image appear in the left hand panel The displacement calculations using a cross correlation function operates on the FFT of the image so it is important to set the cut off values judiciously Low Frequency Cut Off E Low Frequency Cut Off Width Ea hea hea 4 gt Each band in the EBSD pattern appears as a spike centred at the centre of the FFT and directed at right angles to the original band The FFT diagram is a plot of Fourier comp
8. removed is rotated around into the crystal axes system i e 100 010 amp 001 and its components are displayed e Gradients This page shows the partial gradients of the rotational components in radians per micron never degrees per micron Each rotational component is differentiated with respect to the X 6x and X2 dy axes Unfortunately it is not possible to differentiate with respect to X3 6z e GND s Crosscourt 3 allows the calculation of the numbers of Geometrically Necessary Dislocations GND necessary to create the rotation gradients measured above This number is a lower limit estimate because a CC3 is not able to produce the rotation gradients with respect to X3 6z and b there are generally more slips systems present in a crystal than there are degrees of freedom for the crystal to deform 9 gradients of which 6 are available As such a lowest energy solution is found Please note that this calculation is more complex and hence slower than others in the software when entering this page of results note the progress bar at the bottom left of the application no results will be shown until this is at one hundred per cent GND values can currently be calculated for FCC BCC and Hexagonal materials If you have an interest in looking at a GNDs in a crystal of a different symmetry please let BLG Productions know including a list of the slip systems found in that crystal BLG Productions Ltd 2010 24 sym
9. section 4 BLG Productions Ltd 2010 12 3 2 Materials Phases from Project File Elasticity Coefficients GPa v C11 cia C13 ei ells tell M Iron Carbide 249 152 152 0 0 D Atach ELSO C21 C22 C23 C24 C25 C2 Phase 152 249 152 0 0 D Name lt C31 Cie BERI C34 C35 C36 Vie 152 152 p249 fo 0 0 C41 C42 C43 C44 C45 C46 0 D 0 124 0 0 C51 c52 C53 C54 C55 C56 0 0 0 0 124 0 C61 C62 C63 Ch4 C65 CEG D 0 D 0 D 124 Crystal Symmetry Cubic v Material Name Nickel v It is necessary to ensure that the appropriate elastic constants are loaded into the software for the materials of interest If no name or conflicting data is loaded the Materials button will appear red and the cross correlation analysis can not be run fully until the correct data is loaded When the Materials button on the Project Page is clicked the above Materials Page is displayed on the right hand side of the application On the left of the Materials page is a list of the phases loaded from the project file Those phases with a ticked box next to them have automatically been associated with a set of Elasticity Constants ECs stored by CrossCourtt The stored list of ECs can be accessed by using the drop down box Material Name at the bottom right of the page If it is required to assign a stored set of ECs to a particular project phase first highlight the project phase in the left hand list Nex
10. shown Firstly the Standard deviation SD of the MAE secondly the ratio of the SD and the MAE and finally the back calculated angular error for each region of interest e Distortion Sample Axes This page shows the distortion matrix after being rotated into the sample frame of reference Axis X is horizontal X3 is vertical and X3 is normal to the sample surface Additionally the g component of the distortion tensor is calculated using the elastic constants and the fact that the traction stress normal to the sample surface is zero e Normal Strains This page shows the three components of the normal strains 11 22 33 The default settings are red positive tension and blue negative compression e Shear Strains This page shows the three components of the shear strains 12 23 31 The default settings are red positive tension and blue negative compression Note that as this is a symmetric tensor the additional components 21 32 13 are equal to 12 23 31 respectively e Rotations The distortion matrix for the sample axes can be split into the pure rotation and strain components by separating it into its asymmetric and symmetrical components This page shows the three rotations 12 23 31 The axes of rotation are X3 X and X2 respectively BLG Productions Ltd 2010 lt 23 Also calculated is the High Res Kernel Average Misorientation HRKAM The HRKAM is calculated by compar
11. Cross Court 3 User E11 Normal Strain Sample Axes 0 018 0 014 0 01 0 006 0 002 0 002 0 006 0 01 0 014 0 08 W12 Rotations Sample Axes 0 018 0 014 0 01 0 006 0 002 0 002 0 006 0 01 0 014 0 018 E12 Shear Strain Sample Axes 0 018 0 014 0 0 0 006 0 002 0 002 0 006 0 01 0 014 0 018 Mises Stress Sample Axes GPa 55 5 45 4 35 3 25 2 1 5 1 05 0 wig upy wig uniz unrigg unipe unrige uri up wizz wig up umg umg up unig L Manual E22 Normal Strain Sample Axes 0 018 0 014 0 0 0 006 0 002 0 002 0 006 0 01 0 014 0 018 W23 Rotations Sample Axes 0 035 0 025 0 015 0 005 0 005 0 015 0 025 0 035 E23 Shear Strain Sample Axes 0 005 0 004 0 003 0 002 0 001 0 001 0 002 0 003 0 004 0 005 Geometric Mean of XCF Pk Height 0 95 0 85 0 75 0 65 0 55 0 45 0 35 unig urip umg untzz witgp wari up wng wipe o wig wip wng wrigz unige wiigp wrigz uri L url wig unige E33 Normal Strain Sample Axes 0 0 0 008 0 006 0 004 0 002 0 0 002 0 004 0 006 0 008 0 01 W31 Rotations Sample Axes 0 035 0 025 0 015 0 005 0 005 0 015 0 025 0 035 E31 Shear Strain Sample Axes 0 005 0 004 0 003 0 002 0 001 0 001 0 002 0 003 0 004 0 005 Mean Angular Error radians 0 0026 0 0022 0 0018 0 0014 0 001 0 0006 0 0002
12. Only Horizontal available for the Hexagonal Scan Choosing one of these extracts a linescan going through the pixel that was clicked The Linescan is displayed on a new right hand tab called Custom Results e Use As New Reference This option is only available for use on single crystal results or when only a single grain in polycrystalline sample is displayed The shifts from the chosen data point are subtracted from the measured shifts for all other data points in that grain All strain and stress data is then recalculated This technique should be used sparingly as there will be a loss of accuracy any noise on the subtracted shift measurements is transferred to all data points e Export to Excel this option exports the values in the displayed map directly into Excel No supporting information is exported e Export to Matlab Similarly this option outputs the map data into Matlab the matrix saved will have the name userdata e Close This Closes the current map 5 2 Right Click on Linescan Context Menu l Choose Reference Manually Choose Manual Threshold Output As Bit E S A When a linescan is right clicked a subset of the menu on Edit Y Axis Limits the left is shown These options are mostly similar in Edit X Axis Limits functionality to those found on the Map context menus Remove Current Pixel Show Pattern Use 4s New Reference Export to Excel Close e Ch
13. Stresses and Strains A minimum of 4 ROIs are needed to calculate the distortion tensor Normally up to 20 ROIs are used to oversample and the best fit distortion tensor is calculated The pattern centre values and specimen to screen distance values read from the data collection project files are used to calculate the distortion tensor in the reference axes of the sample surface The lattice distortion matrix is then split to provide the Strain tensor Normal amp Shear strains and Rotation tensor rigid body rotations Given further information about the orientation of the crystal and the elastics constants of the materials the stress tensor is calculated BLG Productions Ltd 2010 S 1 6 Practical Considerations when collecting EBSD patterns 1 Collection of High Quality EBSD Patterns The most important ingredient for successful strain measurement is high quality high resolution EBSD patterns These should always be collected at the maximum camera resolution for best results Pattern quality can generally be increased by see below for more details e Use a camera with a detector containing at least 1000x1000 pixels and preferable with 12 bit gray scale resolution e Increasing the exposure time of the camera to the point where the brightest point in the image is almost saturated This would correspond to an intensity level of 4096 in a 12 bit camera Do not over expose Long exposure times however may lead to probl
14. a range of tools to assist with choosing a suitable reference The polycrystalline settings are input in the Project Page BS cross Court Strain Measurement E File Settings Qutput View About Project Sating About Project Display Materials Pattem Display Poe 11 0 0 IPF Map Kernel Average Misorientation Map Rad Image Quality Map Type Rectangular Map 81 rows 51 columns Scan Sie 81x51 ym Source Postion 0 511 0 74 0 547 Stage Tit 70 Materials No Threshold Set Fd Grains Find Refs Grain Tolerance Angle Gick on the map pixel containing the reference image PEPPER L PR EEE Confidence Index Map Fit Map 10 1 0 IPF Map 30um 35um 40um 454m 50um 55um 60um 65um Oym Tum 80m On loading the project file the default maps as described in Section 2 2 appear They do not include a grain map Itis necessary to create a grain map because the cross correlation method only works if the patterns from the strained region are close in orientation to the reference pattern In Crosscourt 3 we define close in orientation using a measure called the Grain Tolerance Angle or GTA BLG Productions Ltd 2010 26 4 2 Grain Tolerance Angle The Grain Tolerance Angle GTA is used to define a set of pixels that can be compared with a reference of a particular orientation Cross Court works by comparing an identically positioned
15. ap or graph and that all thresholds except Mean angular error see Section 3 9 reject lower values No project information is lost when thresholding so multiple attempts can be made The Clear button on the project page can be used to reset any project threshold Individual data points can be removed from the project by right clicking on the particular data point and selecting Remove Current Data Point from the context menu This does not set any threshold value The data can be reinstated by clicking the clear threshold button There is another opportunity to set a second threshold later on using the Quality Results from the Cross correlation process BLG Productions Ltd 2010 215 3 4 How to select and display a Reference Pattern Unless a polycrystal is being examined and the Find Grains button has been pressed a red warning is displayed to click on a data point in one of the displayed images that you wish to use as a reference point That is the EBSD pattern from that point is to be used as the reference pattern against which all other measurements of strain will be measured The point chosen should be one of zero strain If it is impossible to select a point by eye then one can use the point of best fit highest quality or highest confidence index The values of these parameters for any given data point are displayed at the top of the relevant map More than one point can be chosen in which case strai
16. come Type Rectangular Map visible here 100 rows _101 columns The labels just below this show the name of the project file that has been loaded the i Position 0 5345 0 659 0 5878 type of scan collected i e line rectangular or hexagonal scan and the number of rows Stage Tilt 70 and columns in the data set Below these are a number of buttons Each can be clicked to launch a dialogue box to No Threshold Set edit the values displayed If the button is a red colour then the values stored are not Find Refs valid and should be altered accordingly Grain Tolerance Angle The buttons are e Scan Size click this button to edit the spacing of the collection points in microns e Source Position this is in image units i e one unit is equivalent to the height of the image Oxford HKL images are trimmed to squares so image height image width e Stage Tilt this is the angle between the sample surface and the horizontal This value defaults to 70 when no information is present in the project file as is the case with EDAX TSL files See also Camera elevation Tilt in the Settings Menu e Materials this button is described in the following section e The remaining buttons Find Grains Find Refs and the Grain Tolerance Angle section are only applicable if a polycrystalline or multiphase crystal is being investigated See the polycrystalline case in
17. cy cut off indicator This controls the very long range frequency components which are responsible for blurring the EBSD image It also acts as a flat fielding filter The low frequency cut off width acts in a similar fashion to the high frequency width Try and set the low frequency cut off at a level where the overall pattern looks flattened with sharp lines delineating the edges of the bands Changing these setting should affect the noise level of the measurements The effects these settings have on the EBSD pattern can be followed by observing the lower EBSD image This image is that part of the EBSD pattern within the ROI after it has been processed using the FFT filter It is the back transform of the filtered FFT The buttons Cancel and Savel operate in the manner implied by their label BLG Productions Ltd 2010 20 Reset 3 8 Process Panel Process Panel Effective Camera Pixel Size 35um Do Not Combine Overlapping References Correct for Beam Position Effects Use Batch Mode Auto Export to Excel Auto Save to MatLab Do XCF Once at least one reference and one or more ROI have been selected and the elastic coefficients have been assigned the Process button on the Project Settings page can be clicked The Process button displays the Process Panel on the left hand side of the application The Project Page can be re shown using the Projec
18. d Time Left 26s Mili Each reference is processed in turn and the results are shown as each is finished If the Abort button is clicked the process is stopped and the user is prompted whether to display or discard the partial set of results calculated so far Please note it is unwise to restart the processing after aborting Instead we recommend saving the current project to Excel and then restarting the software and reloading the project If you do restart wait for the Display partial results dialog to appear before doing so BLG Productions Ltd 2010 ee 3 9 Results Pages X Shifts Y Shifts XCFPkHeight XCorrected YCorrected Phosphor Axes Quality Distortion Sample Axes Normal Strains Shear Strains Rotations Stresses GPa Principal Strains Crystal Axes Strains Gradients Mean Angular Error Radians Geometric Mean of XCF Pk Height After the results are calculated they are displayed on the right hand side of the application as shown above The two displayed by default are the Mean Angular Error and the Geometric Mean Cross Correlation Function XCF Peak Height These are defined below Results for each Reference pattern have a tab at the top of the screen For each reference there is a lower toolbar shown above detailed as follows e X Shifts Y Shifts These show the results from the cross correlation calculations There is a map or linescan panel for each ROL e XCF Pk Height This pa
19. d by an associated project file All the images should be in a single folder Selecting Load Folder starts an Open Folder Dialog Use the dialog controls to select the folder containing the image and Click OKI Scan_0N Type No Project 38 rows _1 columns Scan Size 38 x 1 ym About Project Display Materials Pattern Display Images Check boxes to select Reference Patterns Source Position 0 5 0 5 0 5 C Sean Stage Tilt 702 No Threshold Set Find Grains Grain Tolerance Angle C Sean C Scan C Scan C Scan C Scan C Scan C Scan C Scan C Scan C Scan Euler Angles 0 0 0 Click on the map pixel containing the reference image C Scan C Scan C Scan C Scan C Scan C Scan C Scan r c tif a ricb tif A r2cb tif r3cb tif r4cb tif rocb tif r6c0 tif r cb tif r cb tif r9cb tif r10c0 tif r11c0 tif r12cO tif r13cO tif r14cO tif r15cO tif r16c0 tif r1 cO tif v Vv Linescan order by number lt gt Map order by row column Date order by creation The Project panel is shown along side some new controls based around a list of image names Click once on an image name to select the image Double Click to choose as a reference The images can be rearranged using the radio buttons below the list Once an image is selected a Remove Pattern Butt
20. dexed as ypts xpts refnum and containing of doubles mean angular error distsem 4 Dimensional Indexed as ypts xpts 1 8 refnum containing the array of doubles elastic displacement gradient tensor as calculated in the phosphor frame of reference The 1 8 elements are in the following order A11 A33 A12 A13 A21 A22 A33 A23 A31 A32 distsamp 4 Dimensional Indexed as ypts xpts 1 9 refnum containing the array of doubles elastic displacement gradient tensor as calculated in the sample frame of reference The 1 9 elements are in the following order A11 A12 A13 A21 A22 A23 A31 A32 A33 datarots 4 Dimensional Indexed as ypts xpts 1 3 refnum containing the datashears array of doubles Normal Shear and Rotational components of datanorms distsamp The 1 3 elements are indexed as w12 w23 w31 e12 e23 e31 and e11 e22 e33 respectively BLG Productions Ltd 2010 AQ datastress 4 Dimensional array of doubles Indexed as ypts xpts 1 7 refnum containing the stresses calculated in the sample frame of reference The 1 7 elements are in the following order S11 S22 S33 S12 S23 S31 Mises With thanks to Ben Britton BLG Productions Ltd 2010 4
21. e CD Continue following the on screen instructions for installation of the Cross Court 3 software package and drivers for the security dongle The first time the dongle is inserted to any USB port the Found New Hardware wizard will start Follow the instructions and allow Windows to automatically find the drivers for the dongle 1 4 Notes for Users of CrossCourt 2 CrossCourt 3 CC3 is an expansion of the CrossCourt 2 strain measurement Tool The main improvements are as follows Importing of EBSD project files from the EDAX TSL and Oxford HKL systems allowing the software to utilise stored project information such as Euler Angles Phase Pixel position etc Such information means the software can process maps both rectangular and hexagonal scans as well as line scans and non spatial collections of data These data sets are plotted graphically in a manner suitable to the scan type Results can be output as bitmaps for use in reports or exported to Excel for further analysis Correctly formatted Excel datasets BLG Projects can be reloaded into CC3 Cross correlation is now performed across multiple processors if available using an improved engine to CrossCourt 2 This can speed up the processing of data sets by a factor of almost the number of processing cores i e 2 cores gt almost twice as fast 4 cores gt almost 4 times as fast etc CC3 is designed to allow processing from more than one reference point so that more
22. e the GTA Because they are in the same grain we can be sure that there will be a set of one or more points in the grain that have orientations within the GTA of the orientations of both references I e there is always overlap between the two references because of the definition of the grain used If there was no overlap the references would be in separate grains by definition Next consider a point O in the overlap region Point O has a measured shift with respect to A of say Soa and say of Sog with respect to reference B Hence the shift between A and B can be calculated as Sag Soa Sop Where there is more than one point in the overlap an average is taken Once the shifts between the references are calculated all the data measured w r t reference B can now be assigned to reference A This is clearly a linear approximation and it will become less accurate as the extent of the bend across the grain increases Note it is possible to override the automatic combination of references from the same grain by ticking the Do not combine overlapping References Checkbox on the Process Panel Grain Number Map It is normal when analysing polycrystals to use the find grains button and accept the displayed grain tolerance angle When this is done a grain map is added to the already displayed maps Each grain is assigned a number starting at zero The number for any grain is displayed above the map and changes as the mouse cursor is
23. e this value is forced to 0 Examine the Y shifts They should appear as in the figure Raw Y Shift ROI 0 If as shown there was a shift in the Y direction of more than 1 pixel then in this case the beam rose up the specimen surface as it moved across it The specimen should be rotated about the beam axis to correct for this and the data retaken until the Y shifts are below 1 pixel for the entire scan When this is achieved then examine the X shifts BLG Productions Ltd 2010 35 Raw X Shift ROI 0 The total pattern shift was 30 pixels in 9 beam steps In the case shown the actual distance moved by the beam was 900 microns The Pattern therefore moves 30 900 0 033 camera pixels per micron Or to put it another way each pixel in the camera sees an area of 30 3 um x 30 3 um The effective camera pixel size in this case is 30 3 um Use the Effective Camera Pixel Size button on the Process Panel to input this value into the software The value is remembered between different software sessions The result is independent of the camera position the pattern centre the scan size or the step size Recalibration is only necessary if the camera set up is changed i e if the distance from the camera to the phosphor screen or the camera focus on the phosphor screen is changed BLG Productions Ltd 2010 36 5 5 Load Folder The Load Folder menu item gives the user the ability to process a group of images that are not connecte
24. ed data display If the results for a single grain are to be studied they can be selected by clicking on the tab labelled with its reference pattern If you have navigated away from the Process and you wish to view combined grains again then scroll to the very right hand side of the top menu where you will find the Combined Results button again BLG Productions Ltd 2010 29 Section 5 Additional Functions 7 5 1 Right Click on Map context Menu Output to Bitmap Copy to ClipBoard Edit Colour Scale Remove Current Data Point Show Pattern Extract LineScan gt Use As New Reference Export to Excel Export to matlab Close When a map is right clicked a subset of the menu shown on the left is displayed Each option is considered in turn below e Output to Bitmap Click this to save a bitmap of the map to file e Copy to ClipBoard the image of the map is saved to the Clipboard e Edit Colour Scale Clicking this displays the Set Limits dialog as shown Use the dialog to select a range of values and click OK Ticking the Use Automatic Limits checkbox uses the maximum and minimum values in the dataset to define Maximum Value 0 01703 the colour range Minimum Value The Minimum Value Radio buttons can be ve Max Value used to set the minimum to zero or the ve maximum value Manually entering the minimum value will automatically select the Custom radio button I
25. ems associated with specimen drift e Increasing the electron beam current this may reduce spatial resolution because of the increased electron probe size e Tilting the Sample to 75 for stronger EBSD pattern contrast However this will increase the electron probe footprint on the specimen surface and so add uncertainty as to where the pattern originates e Having a bright and scratch free phosphor e Selecting an appropriate electron beam voltage Higher voltage improves pattern contrast but increases the sample depth of beam penetration 2 Avoid Scratches on the Phosphor Any scratches or other blemishes on the phosphor represent areas of static content when comparing images Static content may cause a true shift between strained and reference images to be smaller than it really is or even result in a zero shift measurement Ideally the phosphor screen should be in perfect condition The normal EBSD procedure of subtracting the background image of the phosphor from all recorded images of the EBSD pattern can have several advantages in this respect It not only boosts the contrast of the patterns but may also remove phosphor defects However it is not always successful and good background images may be difficult to obtain especially if the sample is a single crystal In certain circumstances e g in the case of measurements near an edge of a sample where electrons emitted from both of the surfaces at the edge contribute to the EBSD
26. f as shown the range straddles zero a colour range of the maximum absolute value E g in the example shown a scale of 4 55 will be used This is so that zero is always green The Use Log Scale checkbox is self explanatory but the Minimum Log Value is not This value is the number below which data is coloured green i e equated to zero As the log of zero is Infinity this allows data that contains both positive and negative values to be plotted essentially on different colour scales For presentation purposes this value makes a huge difference to how the data is displayed too small a value and the data will appear mostly red and blue in the standard HSV colour scheme too large a value and low BLG Productions Ltd 2010 30 values will not be displayed at all If the colour scale is entirely positive this value is not used The Load Lasti button loads the last colour scale selected it can be used to apply the same colour scale to a series of maps e Remove Current Pixel This is used to remove the data point that was clicked on from the dataset It will not be used when calculating strain values e Show Pattern This option displays the EBSD pattern associated with the clicked data point The display changes automatically to the Pattern Display page e Extract LineScan This opens into a choice of Horizontal or Vertical linescans in the case of a Rectangular Scan
27. fe tolerance will be suggested This is the angle subtended by 1 4 of the ROI width as seen from the electron source 4 3 A Note about Grain Definitions Crosscourt 3 0 used the GTA to define a grain as the collection of data points that have orientations within the GTA of the reference orientation This collection was not necessarily contiguous and had little in common with more traditional grain definitions BLG Productions Ltd 2010 227 A more standard grain definition is one of a contiguous set of data points bound by a step change in orientation of more than a certain angle That is to say every point within the grain will have an orientation within that angle of its immediate neighbours This is the grain definition now used by Crosscourt 3 1 the angle in question is the GTA The change that has allowed this is the use of more than one reference point per grain In the new system the problem then becomes how do we relate the references to one another in a single grain Imagine the situation where a grain is bent into a gradual curve such that the orientation at one end of the grain is disorientated from the other end of the grain by more than the GTA In the old system the 2 ends of the grain would be 2 different grains because at least 2 references would be needed Consider 2 references A amp B in the same grain as defined for CC3 1 that are separated by a disorientation that is more than the GTA but not more than twic
28. from these phase descriptions the attached ECs are never used BLG Productions Ltd 2010 14 3 3 Thresholding Next in the Project Panel is a panel initially marked No Threshold Set No value need be input here However maps or graphs containing information from the project file are displayed on the right hand side of the page as shown above Thresholds are a useful device for removing low quality data from the dataset For example the map shown below shows the Confidence Index map for an indent experiment The area underneath the indent gives low quality EBSD patterns hence the low Confidence Index and such data will produce erroneous strain values A project threshold level can be set to ignore all data with a value below it by left clicking depress the left button of the mouse on the colour scale of a map or the vertical Y axis of a graph In the case of line scan the threshold value can be entered manually using the right click menu Section 5 2 Image Quality Map Image Quality Map 2um 6um 10pm 10pm 14um 14um 18um 18um 22um 22um 26um 26pm 30pm Opm ND Dp N Y N 3 3 i 3 3 wiog aA 2 P R R 3 3 B 3 a 3 wig wip wig Before Thresholding After Thresholding at IQ 200 Once set the details of the threshold are displayed within the panel and a CLEAR button becomes visible there to allow the threshold to be cancelled Note that thresholds can be set using any m
29. ge shows the normalised Peak Height of the Cross Correlation function XCF between the reference pattern and the pattern from a particular point in the map or linescan Again a map or linescan is displayed for each ROI The better the correlation between the two ROI the bigger the XCF peak is The values are normalised so that the Reference pixel has a peak height of one Sometimes because the brightness of a pattern may be more than that of the reference pattern the normalised value may be slightly more than one e X Corrected Y Corrected These are the shifts after correction for the beam offset Figures will appear here only if the Correct for beam shift effects toggle was ticked on If they are displayed all of the following results are based on these shifts If they are not displayed the results are based on the raw shifts e Phosphor Axes This page shows the 8 components of the distortion matrix that can be calculated directly They are presented with respect to a frame of reference in the Phosphor screen Axis X is horizontal X2 is vertical and X3 is normal to the screen Initially all the components are shown with the same colour scale range These can be reset by right clicking on the map and choosing the Edit Colour Scale context menu option This page is only shown when 4 or more ROI have been used BLG Productions Ltd 2010 22 e Quality Mean Angular Error and Geometric Mean of XCF Pk Height Th
30. ible It may not be known precisely if the selected region for the reference pattern is actually strain free In this case although the precision of the technique remains the strain values measured will not be accurate All relative measures will have an accuracy and precision at the stated 2 parts in 10000 In the mapping facility used in CrossCourt3 it is possible to set the reference pattern to be used both before and after the strain tensor and rotation tensor have been calculated The before calculation reference patterns can be input manually or automatically using different criteria For manual input the user may wish to choose a point far away from a known point of strain e g an indentation point crack or grain boundary triple point in a polycrystalline material The selection can be on the basis of the quality of the EBSD pattern for example how sharp it is or how well a simulated pattern overlay the actual pattern BLG Productions Ltd 2010 SFe For automated reference pattern detection these same criteria can be used in which case the software will search through data provided in the project files of the EBSD pattern indexing routines of TSL or HKL and search for the best pattern s as reported using their internal measures Often the reference patterns found by these routines are not at the centre of grains for example where the strain might be thought to be least but closer to a grain boundary In this case the user has the opti
31. imen to film distance i e the standard calibration data for interpreting EBSD patterns This information is stored in the project files for the datasets but must be manually entered in the case where no project file exists The calibration routines in TSL OIM DC and Channel 5 software package can be used for this The accuracy of the measurements depends on the accuracy of the calibration data provided The calibration parameters are required to enable conversion of the directly measured shifts of the EBSD pattern within an ROI which are in units of pixels to an angular measure made with respect to the EBSD pattern centre This calibration is essential for calculation of the strain tensor 7 Pattern Centre Position Specifically the value of Y is important If the sample is too low with respect to the phosphor screen then the spread of the spray of back scattered electrons will be centred low down the phosphor screen which means that the signal at the top of the screen will be weak and the results from ROI centred there will be very noisy and will degrade the overall performance of the calculations Try and adjust the height of the stage so that the spray of electrons is centred on the centre of the phosphor screen 8 Choice of the Reference Pattern The reference EBSD pattern should be taken from a region known to be strain free All other patterns will be compared to this pattern so that it should be of as high a quality as poss
32. ined as used in Matlab Project Details Variable Name Type Description numrows integer Number of rows in data set numcols integer Number of columns in data set numphase integer number of phases in project file numims integer Total number of images in project imhig integer Height of EBSD image in pixels imwid integer Width of EBSD image in pixels pixelsize double Effective size of camera pixels in microns stagetilt double This is the effective tilt angle between the phosphor and the sample surface xstep double X direction Step size in microns ystep double Y direction Step size in microns roisize integer Size of the Regions of Interest in pixels numroi integer Number of ROI used in analysis numref Integer Number of References used pcfrac 1x3 array of centre as a decimal fraction doubles X Y 2 stiffnessvalues 3dimensional array Contains stiffness values for each of doubles phase indexed as 6 6 0hasenumber phasetxt Array of strings Contains phase names roiloc 2 dimensional array of doubles Array contains ROI centres indexed as roinumber 2 with each line of the form roixloc roiyloc confidence_index_map euler_phi_map euler_phil_map euler_phi2_map fit_map grain_number_map image_quality_map kernel_average_misorientation map_rad xpos_map ypos_map phase_index_map 2 dimensional array of doubles Arrays containing Project Data angle
33. ing a scan and how to create a ctf file containing the project information Orientations etc The Load Excel Project button can be used to reload datasets that have been previously saved to Excel using the Output Save As option as detailed later The Load BLG Project button is used to load blgp projects which is a proprietary format used by Crosscourt to save and reload data These options are also available from the File Load Project menu item on the main menu bar In addition is the Filel Load Folder menu item This allows the user to select a folder containing only EBSD patterns These may have been recorded as a separate line scan or as a set of patterns that have no defined relationship with each other this is discussed later in Section 5 5 BLG Productions Ltd 2010 10 2 2 General Software Layout Once the project file has been successfully loaded the Project Page is shown on the left hand side of the application On the right hand side of the application there are four tabs About Project Display Materials Pattern display Initially the Project Display page is shown containing all maps and any line scans gleaned from the project files The illustration below is for data obtained from an EDAX_TSL project file The maps shown are IPF Kernel Average Misorientation KAM Image Quality and Confidence Index r s gt BB cross Court Strain Measurement Versio
34. ing the rotation of a given data point with those of its immediate neighbours This gives very similar information to the standard KAM but with a precision of better than 1 100 of a degree The units of the rotation maps can be switched using the Settings menu e Stresses This page shows the Stresses in the sample axes as calculated using the Elastic Coefficients and the crystal orientation The Mises Stress is also calculated e Principal Strains For any combination of Shear and Normal strains there exists a frame of reference in which the strain matrix reduces to purely Normal strains These are called the Principal strains and the frame of reference is called the Principal Axes CrossCourt calculates these and displays them However at this moment the maps are not perfect For a given data point 3 principal strains and axes are calculated but it is not trivial to line these up with the principal strains axes of the neighbouring data points Currently CrossCourt orders the strains so that the first principal strain is the most tensile and the third is the most compressive This works well some of the time but is not a perfect solution Often the 2 weakest strains are mixed up Similarly the maps of the Principal axes are not ideally plotted A given axis may be in any direction but for plotting purposes these are reduced to the unit IPF triangle e Crystal Axes Strain In this set of results the symmetrical strain matrix rotations
35. is is the initial page of the results that is displayed when the calculations are finished The Mean Angular Error map shows the back calculated errors for the distortion matrix shown in the Phosphor Axes page i e the mean angular difference between the measured shifts and the back calculated shifts from the distortion matrix If only 4 ROI are used this calculation is null If more than 4 ROI are used the best least squares result is shown in the Phosphor Axes page and the Mean Angular Errors become meaningful The Geometric Mean of XCF Pk Height map is calculated by multiplying all the normalised XCF Peak heights for the ROI The resulting number is a good indicator of the quality of the data for each point Both of these maps can be used to set a threshold for the Results For example if some pixels are exhibiting a large Mean angular error then it is possible the patterns taken at these points were poor or there was some artefact spoiling the results Clicking on the scale bar at the values of the error will remove all data above or below this from the calculations as appropriate Accordingly the data will be re calculated without them This is done as described above for the project maps Any threshold set only applies to the results data for the current reference More details can be found in the Tutorials The Settings Menu has a Show Extended Quality Maps option When this is checked a series of extra maps are
36. lline material This is achieved by comparing an EBSD pattern from the target area with one taken from unstrained material within the same crystal These rotations and distortions can be measured to a precision of 2 parts in 10000 dependant on the quality of the EBSD pattern In the program it is assumed that the measured distortions result from elastic strain in the volume giving rise to the patterns or to a rigid body rotation at zero strain or to a combination of the two The program includes analysis to extract components of the strain tensor and rigid body rotations The precision of these measurements is also in the 2 parts in 10000 regime The analysis includes use of the fact that the surface examined is in plane stress and the assumption that the strain beneath it is constant within the sample volume The results obtained must be used with full recognition of this latter assumption It should be noted that any artefacts in the recorded pattern arising from scratches on the imaging phosphor defects in the camera uneven sample surface or poor sample preparation can also give rise to apparent distortions and rotations and care must be taken to avoid them See appendix 1 for more details Further if there are significant changes in the contrast of the diffraction pattern caused amongst other factors by sample topography specimen charging shadowing or large atomic number differences between reference area and target area then there may be some los
37. metry and we will endeavour to add them Maps of the total GND density totals of Edge amp Screw GND densities and maps of the GND numbers for individual slip systems are produced 3 10 Display Page The data display page is selectable once the XCF calculations have finished using a button which appears just below the main form menu Once the page is selected use the mouse cursor to hover over any of the data points in the maps or linescans The data display page will show a synopsis of the information about that data point The top display shows the unit cell including the measured distortion and the bottom display shows the principal strains and their axes The distortion multiplier control adjusts how much the distortion is magnified when applied to the wire frame unit cell BLG Productions Ltd 2010 25 4 1 Finding Grains In the polycrystalline case we have to face the problem that we cannot use cross correlation to measure the differences between patterns from radically different orientations because the ROI will contain completely different information to each other As such each Reference pixel is only valid for comparison with pixels of similar see Section 4 2 orientation A second problem occurs in the polycrystalline case finding an area that can be assumed to be strain free is now much more difficult CrossCourt 3 uses a measure called the Grain Tolerance Angle to define a grain and contains
38. moved into another grain In addition the Settings menu contains a Minimum Grain Size option This number is the minimum number of data points in a grain necessary for that grain to be included The default value is two The map displayed on the left has the minimum number of points set to 20 BLG Productions Ltd 2010 28 4 3 Find References It is possible to select the reference pattern to be used for each grain manually Alternatively the Find Reference button can be clicked and they will be found automatically A dialogue window opens asking for the criterion to be used in selecting the reference pattern The options are Kernel Average Misorientation Image Quality Confidence Index and pattern Fit These parameters are determined in the data collection software used We currently recommend using the Kernel Average Misorientation See the tutorials for an expanded discussion of this It is important to note that the reference pattern selected will probably not have zero strain It is only the best pattern observed with the criteria selected All strain values measured with respect to it will therefore be relative to it If the reference grain is in tension for example the true strain for all other measurements will be the measured strain plus the strain of the reference pattern All measured rotations are with respect to the actual orientation of the reference grain The reference patterns will then be found automatically and
39. n Previously all files had a simple blg extension Old blg files can be renamed by changing the extension BLG Productions Ltd 2010 4 1 5 Overall Description of the Software The CrossCourt software package uses cross correlation based techniques to measure the relative shifts between EBSD patterns caused by small lattice rotations or elastic strains Comparisons are made between a reference pattern representing a point of zero strain and a number of patterns taken from the area of interest There are 3 stages of the analysis 1 Calibration and Preparation In some respects this is the most important section Good quality data leads to good quality results and it is worth spending time optimising the pattern collection The user is referred to the separate manuals issued by EDAX_TSL and OXFORD_HKL for details as to the procedures adopted to carry out the procedure 2 Measurement of Pattern Shifts Once the EBSD patterns are transferred from the data collection software used i e OIM_DC or Channel 5 and loaded into CrossCourt3 the software measures the relative shifts of small selected Regions of Interest ROI between each pattern and another selected as a reference pattern The ROIs are selected either automatically or manually and the shifts are measured in pixels The results from this are displayed graphically and can also be exported to a BLG Project File based on an Excel worksheet 3 Calculation of Relative
40. n 3 indent75 0se Sa File Settings Qutput View About Project Settings S ae Paten Det 1 0 0 IPF Map Kernel Average Misorientation Map Rad F Source Postion 0 5345 0 659 0 5878 Stage Tit 702 Find Grains Find Refs Load Completed Succesfully p R gR B H On the left hand side is the Project Panel see Section 3 1 On the right hand side project maps of the following types are displayed Cross Court Title HKL Oxford Title TSL EDAX Title Kernel Average Misorientation n a n a Image Quality Band Contrast IQ Confidence Index MAD CI Fit Band Slope Fit Other IPF Maps are also displayed and it is possible to display maps of the Euler angles using the Settings menu Also use the Settings menu to switch between degrees and radians Right Clicking on the maps graphs will show a context sensitive menu the functionality of which is detailed in Section 5 1 amp 5 2 This manual initially considers the case of a single crystal section 3 and later describes the extra controls used with poly crystals section 4 BLG Productions Ltd 2010 1 3 1 Project Page This Page initially displays information about the Project that has been loaded At the top is a Project Settings button that can be used to re open this page As indent 5 osc analysis progresses more buttons be
41. n maps will be drawn using each reference point in turn It should be noted however that following calculation and display of the maps new maps can be drawn using a new reference selected from any one of the maps The new maps are calculated by calculating the strain tensor of each point by subtracting from the old values the values of the selected new reference point as calculated using the previous reference point For the maps including hexagonal maps which are not shown left clicking on a map data point will not only select the data point as a reference but also mark it with a cross In the case of the linescan graphs left clicking on a point on the curve will choose that point as a reference Alternatively right click on the graph and choose the Choose Reference Manually and enter the number of the desired reference point Reference List Scan2_r14c8 tif Once a reference has been selected the Reference List is displayed on the project page Clicking the Show button will display the selected EBSD pattern on the Pattern Display Remove tab see Section 3 5 on the right hand side of the application Clear All The Clear Alli button will remove all references from the list whilst the Remove button will only remove the highlighted reference After the reference pattern s have been selected it is still necessary to set up the Regions of Interest ROIs to be used in the
42. ns These linked settings change the angular units used by the KAM and other rotation maps The units of the Gradient maps are always in Radians micron Settings Show Extended Quality Maps Displays extra maps on the quality tab as discussed in Section 3 9 Output Export As BLG Project to Excel This option is only enabled after a project has been loaded Selecting this starts an instance of Excel and outputs the current data set to it Maps of more than 255 columns require Excel 2007 or later All project data is saved as well as all the shift data for all the references and the calculated strains and stresses Combined amp Custom Results are not output they can be output to Excel by right clicking on the individual map linescans Output Output As Bitmap This option creates a bitmap containing an array of all the currently displayed maps linescan panels The user is prompted for a filename Output Export as Matlab File This option outputs the data to an mat file to allow further processing in Matlab Data is BLG Productions Ltd 2010 33 exported in MAT File 5 format This format is very useful when very large amounts of data are collected as Excel files can become overly large Note Hexagonal Scans can not be exported to Mat File format e Output Export BLG Project This outputs the current data in a blg file format e View This menu allows the user to choose a particular right hand tab from the list for dis
43. of the 3 Load Project buttons as shown when the software is first started File Load Folder This option allows the user to select a folder containing images that are not necessarily part of a project All images in the folder are loaded See the section Load images from Folder for more information File Close Current Tab This option closes the currently visible results page on the right All data associated with the results is discarded Settings Minimum Grain Size This number is the minimum number of data points in a grain necessary for that grain to be included The default value is two Settings Camera Elevation Angle This value is used when correcting the pattern shifts for position offsets Camera tilt is the elevation of the camera tube from the horizontal ve when the camera is pointing upwards ve for tilts when the camera is facing down It is used in combination with the specimen tilt to allow the correct rotation of reference axes from the screen to the specimen Settings Change Colour Scale Allows the user to change the style of look up table LUT used to display the map data Settings Show Euler Maps Toggles the display of the Euler maps on the project maps panel Settings Show IPF Maps Toggles the display of IPF maps on the project maps panel These maps are with respect to the cubic axes of the sample not w r t the directions in the crystal system Settings Use Degrees amp Settings Use Radia
44. ollecting a wide solid angle of the diffraction pattern 12 Data Scan Collection For the collection of the actual data set to be analysed we recommend using an automated line scan or area scan generated by your EBSD collection software It is first necessary to initialise the recording of patterns at each point in the video page When you select the format of the image always select 12 bit tiff images if possible In the data collection page select line scan for line scan images or square grid option if recording an area image If available collecting and averaging more than frame at each point in the data set makes a large difference to the quality of the data collected This improves the signal to noise ratio in the patterns collected and leads to less noise in the strains calculated 13 A note about Euler angle settings Each of the EBSD manufacturers uses a different default frame of reference for the Euler Angles Crosscourt automatically rotates these into the frame of reference used by Crosscourt However if the Euler data is collected in a non standard frame of reference Crosscourt will apply the default correction anyway which leads to the mixing up of the normal strains This is because the Euler angles are only used in the rotation of the BLG Productions Ltd 2010 8 Elasticity coefficients which are used to set the traction free condition i e S33 the stress normal to the surface is zero so that all 9 tensor component
45. on becomes visible use it to remove the currently selected image from the list A Selected image can also be moved up and down the list using the arrow buttons The behaviour of the Project Panel is also slightly different to the situation when a project is loaded The Source Position must be entered and the Materials set Also Euler angles for the images must be entered Cross Court 3 assumes all images have the same Euler Angles Click the Euler Angles BLG Productions Ltd 2010 37 button to manually enter the Orientation using the Euler Angle Dialog Box ES Enter Euler Angle phit deo 0 4 Euler Data Source HKL PHI deg 0 TSL phi2 deg 0 If the Euler angles are from EDAX_TSL software check the TSL Euler Data Source radio button This is necessary because Cross Court 3 uses a different reference system to EDAX_TSL Correcting this is done by the software if the TSL option is selected Essentially 90 is added to the value of phil Once a reference image is selected the software can be used as normal The data is treated as a linescan with a default step size of lum This can be altered using the project panel controls Section 3 1 BLG Productions Ltd 2010 38 6 Matlab Data Structure for CC Analysis Introduction The following file lists each of the variables used in CC analysis as output in the Matlab format The indexing is outl
46. on to override the automated choice After strain measurement has been performed strain maps are presented Inspection of these maps may then reveal a more likely location for the reference pattern For example at the centre of a region that shows no strain gradients over several microns for all strain components Alternative procedures are currently under investigation 9 Correction for beam movement When the electron beam of the SEM is moved across the sample there is an inherent and identical shift of the EBSD pattern across the phosphor screen If large enough generally more than 3 microns the software will interpret this as a rigid body rotation For this reason the reference pattern should be recorded from an area as close as possible to the area of interest If beam movements of more than 3 microns are made in a line scan or a map for example then a beam movement correction procedure must be applied The calibration method to facilitate this procedure is described at the end of this manual However it is best performed at the outset immediately after familiarity with the software has been achieved 10 Keep the Experimental Conditions the Same for All Patterns The patterns collected must all be of the same image size and must all be collected under the same microscope conditions 11 Specimen to Screen Distance The best experimental arrangement for taking measurements is to have the phosphor screen close to the sample c
47. onent i e its frequency versus the intensity of that component i e the brightness on the screen The lower frequency components are towards the centre of the figure and the higher ones at the limits of the spikes The higher frequency components carry information with respect to the sharp changes in the pattern whilst the low frequency components carry information as regards long range contrast changes The spikes only extend part way across the FFT It is important to set the red ring shown in the figure so that it is just outside the longest spike seen All Fourier components beyond the ring will be ignored in the calculations as they convey only noise in the EBSD pattern The frequency of the cut off is displayed in the list at BLG Productions Ltd 2010 19 the top of the panel It is headed High Frequency cut off The precise value can be altered using the tumbler controls Below this setting is High frequency cut off width This designates the range of frequencies beyond the red ring over which the contribution of that frequency component to the FFT is gradually reduced to zero The expansion of the cut off width can be seen in the figure as an expansion of the FFT image outside the red ring It is important to set this value to at least a quarter of the high frequency cut off value It reduces artefacts or ringing in the FFT The low frequency cut off is set by using the tumbler control adjacent to the Low frequen
48. oose Manual Reference This option brings up a dialog box designed to allow the user to enter a reference by number where 0 is the first point on the linescan 1 is the second etc etc This option is made available because it can be difficult to choose a reference by left clicking near the axes BLG Productions Ltd 2010 31 e Choose Manual Threshold Similarly this option allows the manual entry of a threshold value using a special dialog box e Output to Bitmap Click this to save a bitmap of the linescan to file Edit Y Axis Limits This option displays the Set limits Dialog as shown in the previous section Any range can be entered as long as the maximum value is greater than the minimum These limits are applied to the Y value axis of the linescan e Edit X Axis Limits This option is identical to the above but changes the data points displayed on the X axis instead e Remove Current Pixel This is used to remove the data point that was clicked on from the dataset e Show Pattern This option displays the EBSD pattern associated with the clicked data point e Export to Excel this option exports the values in the displayed linescan directly into Excel No supporting information is exported e Close This closes the current linescan BLG Productions Ltd 2010 232 5 3 Menu Items File Load Project This option replicates the functions
49. pattern there may be such a change in pattern contrast that erroneous strain measurements can result 3 Avoid Over Saturation of the Pattern If the patterns contain areas of over saturated pixels especially the background pattern used for background subtraction these will represent static features in the pattern and can cause an incorrect measurement of zero shift just as is the case for phosphor blemishes 4 Good Vacuum amp Clean Samples Significant contamination of the sample by the generation of a thin film of carbon deposit on the surface drastically reduces both the pattern quality and the ability to BLG Productions Ltd 2010 6 make precise and accurate strain measurements Using clean samples and anti contamination devices such as a liquid nitrogen cold finger is recommended 5 Number of Pixels in the Pattern Obviously if a pattern is taken with the highest resolution that the camera can provide then more precise and accurate shift measurements are obtained However patterns taken at a higher resolution take proportionally longer to expose which means that issues such as sample drift and beam contamination become more apparent Some compromise between quality and speed may become necessary Such compromises can be tolerated in cases where the focus of the experiment is not on elastic strain but on rotations caused by plastic strain 6 Calibration It is necessary to know the EBSD pattern centre and spec
50. play e About This option shows a dialog containing information about the version number etc of CrossCourt BLG Productions Ltd 2010 34 5 4 Beam Shift Calibration Introduce into the SEM a clean single crystal of silicon or germanium for example and set up the microscope specimen stage and beam conditions for obtaining high quality EBSD patterns Set the EBSD camera to high resolution mode and adjust exposure time accordingly Adjust the specimen with respect to the beam x traverse so that the x traverse trajectory lies exactly parallel to the X axis of the phosphor This can be difficult to do However with patience using the following procedure the correct condition can be reached Observe the sample at 1000 x magnification Set up a line scan of length 100 microns and step size 5 microns Carry out the scan recording the pattern at each step You will have to turn on the record images function of the data collection system and set type to tiff Save to file Open Crosscourt3 and load the file Carry out the software set up as described above Set the reference at the first point and manually select a single ROI and position it at the pattern centre Do XCF The data will be a line scan of shifts The output will first show the Mean angular error and the Geometric mean height The range of the mean angular error excluding the first data point should be of the order 0 0002 radians The exclusion of the first pixel is becaus
51. s can be calculated If the wrong frame of reference is used the stress is set to zero in a different plane and the normal strains are not calculated correctly BLG Productions Ltd 2010 9 2 1 Loading a Project When the software is started the welcome screen is displayed One the left hand side is a panel containing 3 buttons as shown The Load TSL Project button opens a dialog box allowing the user to search for an EDAX TSL project Load TSL Project file with the name project_name osc where project_name is the name of the project CC3 will be Load HKL Project expecting the file to be alongside a folder that is also called project_name This folder should contain the 5 EBSD pattern image files ideally as Tiff files This is the standard EDAX TSL way of saving the datasets Please refer to your EDAX TSL documentation to find Load BLG Project out how to save the EBSD patterns collected during a scan The Load HKL Project button opens a dialog box allowing the user to search for an Oxford project file with the name project_name ctf where project_name is the name of the project CC3 will be expecting the file to be alongside a file called project_name crc as well as a folder that is called project_nameImages This folder should contain the EBSD pattern image files usually as JPEG files Please refer to your Oxford HKL documentation to find out how to save the EBSD patterns collected dur
52. s of sensitivity 1 2 Computer Requirements e Display Size at least 1280 x 1024 pixels e Windows XP Vista or Windows 7 32 or 64 bit and a copy of Microsoft Excel 2003 or later e CD Drive e Multiple CPU cores recommended e RAM of at least 2 gigabytes It is recommended that a hard drive of at least 80 Gb be installed and a computer speed of greater than 1 8 GHz be used For maps of more than 255 columns Excel 2007 or higher is needed to save the projects in BLG format BLG Productions Ltd 2010 sZ For analysis of large datasets of polycrystalline materials with more than 10 grains it is recommended that a 64bit version of windows be installed For very large datasets a 64 bit version of Excel is also necessary Please note that the language settings on both the Operating System and Excel must match this includes the number format 1 3 Installation Instructions CAUTION Do not insert the dongle until after the software has been installed Insert the CD The setup program should start to run automatically If it does not run the file setup exe that is found in the root directory of the CD The computer should ideally be connected to the internet If the installation program fails to detect the necessary Microsoft NET runtime libraries it will download the appropriate files from the internet and install them However if an internet connection is not available a copy of this dotnetfx exe can be found on th
53. t Settings button at the top of the panel The Effective Camera Pixel Size XXum button sets the effective size of the camera pixel This is used to correct for beam position effects Instructions on how to measure the pixel size can be found in Section 5 4 The Correct for Beam Position Effects checkbox is used to enable the beam position offset correction This can be switched on or off at any time and the Results will be updated accordingly See Section 5 4 for more details The Do Not Combine Overlapping references tick box is concern with polycrystalline analysis and hence detailed in that section The Use Batch Mode saves the project and a results file to blgx format no results are displayed This is useful for processing very big datasets with lots of references Results for each reference are saved as the analysis progresses The Auto Export to Excel checkbox is used for exporting the results to Excel When checked the results are exported as soon as the last reference is processed Similarly the Auto Save to Matlab checkbox can be used to save the generated data to a mat file see Section 6 Finally and most importantly the Do XCH button is used to start the cross correlation processing Once clicked it is replaced with an Abort button and a progress bar as shown below Processing Reference 1 of 1 10 Complete Estimate
54. t use the material name drop down box to select the desired set of ECs The Attach ECs to Phase button will now associate the material phase with the chosen ECs The Name radio buttons can be used to update the phase name with the ECs name or vice versa If the project phase is a commonly used one it is recommended to update the ECs material name so that association is automatic in future To enter a new set of ECs firstly select the desired Crystal Symmetry from the drop down box This will allow the input of only those ECs that are independent and will allow propagation of dependent ECs across the matrix BLG Productions Ltd 2010 13 Next manually type in the values of the independent ECs Once a value has been changed a new button will appear labelled Attach ECs As New Material Click this once all ECs have been entered correctly Enter the name of the new material into the dialog that appears The new material will appear at the end of the materials list Use the methodology described above to assign the new material to a project phase e Once all phases are assigned ECs the Commit button should no longer be red and must be clicked to return the display to the Project Display correctly Often spare phase descriptions are included in the project file though those phases are not present in the scan If this is the case attach ECs for any materials to the project phase as there are no patterns
55. than one grain can be analysed at a time in polycrystalline and poly phase materials CC3 can analyse the orientation information in the project file and decide whether cross correlation is possible between 2 data points The angular difference between the two points must be less than half that covered in a region of interest From this grains or sub grains are distinguished and treated separately The reference patterns used in each grain can be selected automatically using image quality best fit or confidence index or manually BLG Productions Ltd 2010 3 All measured shift data is held in active memory so that subsequent changes made to pattern centre position elastic coefficient values scan parameters reference pattern etc will be reflected in the results without having to re examine the EBSD patterns Notes for Users of Crosscourt 3 0 There are 2 major changes between Crosscourt 3 0 and Crosscourt 3 1 as far as the user is concerned The most important of these is that the definition of a Grain has changed to become more in line with the standard EBSD Grain definition see the Polycrystalline section for more information The second change is an internal change of data format which allows CrossCourt 3 1 to manage far larger data sets than Crosscourt 3 0 Notes for users of Crosscourt 3 1 The naming of the BLG format has changed Project files are now have the extension blgp and Results files now have a blgr extensio
56. they are too big to fit within the boundary 3 finally previously saved ROI lists can be reloaded using the Load button Once at least one ROI has been chosen the ROI list appears on the Project page below the Reference list as shown Regions of Interest The accompanying buttons are Manual ROI 0 459 482 size 256 IRemove removes the highlighted ROI Save List Clear removes all ROI for the list Save List save the ROI to a text file with a ROD file name extension 3 Also shown in this panel is the Processj button This is visible at all times but will not allow progress to the process page until various criteria are satisfied namely there is there is at least one item in the reference list an ROI has been selected If less than 4 ROI have been selected or the Material selection has not been completed the Material button is red then processing will stop after the pattern shifts have been measured Before proceeding to Process the images the Filter settings must be appropriate for the images section 3 7 Once set and before proceeding to execute the analysis the final step is to add all the regions of interest to be used to measure the pattern distortion Click on the add ROI button again to show the options It is normal to use the AUTO option and use the toggle to select more than 10 ROIs Although only 4 ROIs are
Download Pdf Manuals
Related Search