I often get asked what are the decision points and paths to take because of those decision points when looking at cross sections to flip, so I thought I would start to write that up for everyone’s benefit
Step 0 - How Many Pages / Cross Sections are we working with?
-
The number of cross sections to convert can be a major factor in determining which method to adopt depending on the answers to the Decision Tree points outlined below. If you only have 2 or 3 pages of sections and on each page there are only 6 cross sections, then you can “fix” the missing data manually and quickly because you only have a few pages to work with. If however you have 100 pages of 10 sections per page, the work to manually “fix” the missing data is higher so you may choose an alternative method based on that decision point.
-
The elevation range of each cross section can also be an influence - if the range is fairly constant because the road is fairly flat, then it can be quite quick to copy elevation text from one section to another, however if the elevation is varying a lot, then it is likely that no two sections will be the same so copying elevation text from section o section may not be such a viable option.
-
For example, if you have no text to work with, you can create some text items on one section for offset labels or elevation labels manually, and then you can copy the text to all the other sections on a page using e.g. Smart Copy command and that can be an efficient process in some scenarios. This would apply where you have no text to work with at all, or where the text is written as a single line of text for e.g. offset labels vs discrete text items for each offset label.
Step 1 - Cross Sections come in 3 forms
-
CAD Sections - 2D Drawing of Cross Sections - typically a large sheet of sections in plan view and will be supplied in DWG, DXF or DGN format
-
PDF Sections - 2D Drawings in Sheets of Cross Sections - supplied in PDF files
-
3D Sections - can come in LandXML or CSV or DWG / DGN formats (already in 3D)
-
If you get CAD Sections - Go To Plan A
-
If you get PDF Sections - Review the PDF Criteria to determine the correct path to take
-
If you get 3D Sections - You have nothing more to do and you can jump straight to Modeling
Step 2 - PDF Sections can be in two main formats
-
Raster PDFs - these are image only i.e. the images are pixelated and they have no linework or text
-
Vector PDFs - these contain 2D drawing linework and text that can be extracted and used
-
If you get Raster PDFs you have no option other than to digitize the cross sections - Plan D
-
If you get Vector PDFs - Review Step 3 to determine the correct path to take
Step 3 - Vector PDFs can be in two main formats
-
Layered PDFs - These contain the original CAD Layers from the Engineer. They are less often provided than unlayered PDFs, but when you get them they are easier to work with than unlayered PDFs.
-
Unlayered PDFs - These contain much of the CAD data from the Engineer but in an unlayered format. The lines do however often have lineweights or linestyles (dash patterns) that can be used to somewhat separate the data types in the PDF (Existing, Proposed, Subgrade, Grids, Other)
-
Neither of these options provide a decision tree option - Layered is better than unlayered, however these do not really affect the Plan to follow. It is good to know what you are working with however.
PDF Data Content
Vector PDFs can contain the following data elements
- Text objects - Note CAD systems use two types of text object - True Type Fonts and Stroke Fonts. It appears that true Type Font based text typically comes through the PDF as Text objects whereas Stroke Font text comes through as Polyline linework. The Text objects are useful. The polyline text is not at all useful at this time.
- Fill Objects (typically drawn as triangles) - note CAD Hatch Patterns are typically drawn as lines in PDF files, fills tend to be the shaded areas of a plan set. The fills are often found in text white outs also. Fills may or may not have a fill boundary - fill boundaries if present can be useful. Fills themselves are normally not at all useful in the extracted vector PDF data.
- Dashed Lines - during import of Vector PDF data, Dashed lines can be joined into solid lines automatically (making them easier to work with, however they are placed on TBC layers that are named using Joined Dashed Lines and the Lineweight and Dash Patterns of the source lines - this allows different types of line in the source data to be separated to a certain degree during the import process.
- Solid Lines - during import of Vector PDF data, Solid lines are imported and placed on TBC layers that are named using Solid Lines and the lineweight of the source lines - this allows different types of line in the source data to be separated to a certain degree during the import process.
These are quite varied from PDF file to PDF file, and this is likely the highest priority decision point to determine a) what you have to work with and b) the methodology to adopt to flip the sections to 3D. The variations include
- “Real Text Objects” - Text objects import into TBC as Text Objects, however in a fairly high percentage of PDF files, Text can be drawn as polylines that makes the text unusable in the flipping process. If you have Text for Grid Labels, Elevation Labels and Station Labels you can take “Plan A” approach to flipping sections. If you do not have text to work with then you have to go to Plan B.
- If the PDF Sections do contain “Real Text Objects”, sometimes the Grid Labels for Offsets are written as a single string of characters, and sometimes on one side of the 0 offset line. If the Offset Labels do not contain discrete labels for each offset you may need to go to Plan C.
Type of Cross Section
TBC supports 3 different styles / drawing formats for cross sections that are commonly used by CAD systems that generate the cross sections. The formats are as follows
- No Grid - in these sections, the axes of the section include a single horizontal grid line that provides an elevation datum (it is typically labeled with its elevation value) and a single vertical grid line at the 0 offset location (center line) (it is typically labeled). Each section will have a Station Label that states where the cross section belongs on the road / alignment. Each section will display Existing, Design, Subgrade and Other information. The sections will typically be scaled in Horizontal and Vertical - in CAD sections, the horizontal scale is typically 1:1 and the vertical can be exaggerated. In CAD sections, the horizontal is typically drawn at full scale (ground distances). In PDF sheet sections, the horizontal and vertical will typically be scaled - remember in TBC Sheet View, the units are in inches, so if the sections horizontal scale is 1:20 (1" = 20’) then this is a scale of 240 (20’ = 240") in the sheet view.
- Single Grid - In these sections, each section will typically be drawn within a box and there will typically be major and or minor grid lines that fill the box for both offsets and elevations. The offsets and elevations will be labeled and there will typically be a station label for each section. Each section will display Existing, Design, Subgrade and Other information. The sections will typically be scaled in Horizontal and Vertical - in CAD sections, the horizontal scale is typically 1:1 and the vertical can be exaggerated. In CAD sections, the horizontal is typically drawn at full scale (ground distances). In PDF sheet sections, the horizontal and vertical will typically be scaled - remember in TBC Sheet View, the units are in inches, so if the sections horizontal scale is 1:20 (1" = 20’) then this is a scale of 240 (20’ = 240") in the sheet view. TBC can determine the scale automatically from the labeled grid lines if the labels are real text objects. Without Real Text Objects, you either have to provide them (create them manually and copy to each section) or you have to replace them using Plan B below (in this method you convert the Single Grid sections to No Grid Sections by retaining one elevation grid line and the 0 offset grid line. You elevate the 0 offset Grid lines at elevation 0 and each elevation grid line you elevate those to the true elevation value of the grid line. TBC reads these two elevated lines to place at zero offset and at the correct elevation, and uses a user provided scale to scale the sections, and a user provided station label (created at each section using the RPS Increment Text command) to place at the correct location and orientation along the alignment.
- Multiple - In these sections, the drawing sheets have a column (one or more discrete columns) that are gridded with offset and elevation grid lines and labels, and that have more than one section drawn within the column. Each section shares common offset labels but has its own elevation and station labels. In some cases there are double column grids that are not separated by a vertical gap - in these scenarios you need to break / clip the elevation grid lines between two offset grid lines to create a vertical gap so that TBC can interpret the sections as the Multiple style. You can use the RPS Crop Crossing command to achieve this.
CAD cross sections are often delivered in the No Grid drawing format
PDF cross sections typically come in the Single Grid, Multiple Grid or Double Multiple Grid formats.
The key elements to review are
Grid styles
In most cases, grid lines are drawn as solid lines however sometimes
- The grid lines are drawn as dotted lines - these come through in the PDF as a series of dots that are not interpreted by the TBC dashed lines to solid lines conversion process - typically you convert these sections to No Grid sections and use the Plan B approach.
- The grid lines are drawn as dashed lines - these do get converted into solid lines, however each grid line is then made up of multiple segments - typically you run either Project Cleanup or RPS Convert to Linestring to filter out all of the line vertices making the grid lines single segment lines prior to processing the sections.
Text / No Text
TBC requires that the Station Labels and Grid Text Labels are on separate layers and that there is no other data on those layers whatsoever. The Offset and Elevation Grid Labels go on one layer and the Station Labels on a second layer.
If you have no real text objects then they either need to be created manually or you need to adopt Plan B (convert sections to No Grid type and elevate the Offset and Elevation grid line with their appropriate values).
Plan A - PDF has usable text
Where the sections are Single Grid or Multiple Grid style and they have real text objects for Station, Offset and Elevation labels, you can use this method to convert the sections.
- Place the Station Labels on their own layer - e.g. CADXS - Labels - Station
- Place the Offset and Elevation Labels on their own shared layer - e.g. CADXS - Labels - Grids
- Place the Grid Lines on their own layer - e.g. CADXS - Grid Lines - note that you only need the labeled grid lines i.e. the Major Grid Lines. Each grid line needs to be a single segment line. All tick marks need to be excluded. The grid lines need to be orthogonal i.e. they cannot be rotated from E-W or N-S. There cannot be any duplicate lines. There cannot be a box or partial box around their perimeter. There cannot be any other data on this layer.
The section content for Existing, Proposed, Subgrade, Other can be separated onto individual layers however you want it to be separated. The degree of data separation will depend on what you are going to use the data for once it is converted into 3D.
Common data uses in 3D include
- Visual checks only - no real data separation is required in this case, you can keep the data prep process to an absolute minimum and just flip the sections as they are.
- 3D measurements to verify design offsets, elevations and slopes - no real data separation is required, you can keep the data prep process to an absolute minimum and just flip the cross sections as they are.
- Building 3D models of Existing and Design and Subgrade - in this case, it will be easier yo work with the data in the Plan or Sheet view to split the data up into logical layers prior to flipping the data into 3D. The more granular the layering the easier it will be to select data and make 3D models once you get the data up in 3D. Note that you may / may not need to create linear features from the sections once you have the data in 3D - if you do, you can explode data at this point before you convert it to 3D. If you need extra data in the sections, it will likely be easier to draw it in while the sections are 2D prior to flipping. If you do not want ROW lines, Guard Rails, Barriers etc. in your surfaces, you should typically move them to their own layers before you flip them to 3D. Sometimes to create a good subgrade model, you may need to close out gaps in the sections prior to flipping. You can also use tools like Track Cross Section to find e.g. the Finished Grade Surface in all sections or the Base of Subgrade in all cross sections and then flip those lines in addition to all of the other cross section data. Use commands like RPS Relayer, RPS Geometric Selection, RPS Explode Lines to separate out the data as needed. Note that with experience, you may find that some things are easier to select and relayer in the 3D view, because you can use tools like the 3D view rotations, 3D vertical exaggeration, limit box etc. to make it easier to select objects in the 3D view - that will come with experience.
Plan B - PDF has no usable text
Without Real Text Objects, you either have to provide them (create them manually and copy to each section) or you have to replace them using the following method
- Convert the Single Grid sections to No Grid Sections by retaining one elevation grid line and the 0 offset grid line and discarding all other grid lines.
- Elevate the 0 offset Grid lines at elevation 0
- Elevate the elevation grid lines to their true elevation value i.e. the 100’ elevation grid line - set its elevation to 100’.
You can use any of the following commands to set the elevation values
- Set Line Elevation command
- Change Elevation command
- Properties Pane elevation field
- RPS Adjust Linestring Elevation command
- Smart Edit - Join Lines (which allows Elevate and Relayer at the same time)
TBC reads these two elevated lines to place at zero offset and at the correct elevation, and uses a user provided scale to scale the sections
All cross sections will require a station label, you can use the RPS Increment Text command to place Station Labels on each cross section. Typically place the Station Labels at the base or top of the 0 offset grid line using end point snap and a text style that has the appropriate Center Top or Center Bottom justification.
Plan C - PDF has partially usable text (offset grid labels are not discrete text items).
If you have only a few pages of sections to handle, it can be quicker to
- Create a set of Offset labels and simply copy them from one section to another using the RPS Smart Copy command.
- Create Station Labels using the RPS Increment Text command
- Create a set of Elevation labels and simply copy them from one section to another using RPS Smart Copy command.
Plan D - PDF is a Raster PDF - Digitize the cross sections
If you have a Raster PDF then the only way to recreate the data is to use the TBC Digitize Cross Sections command.