Configuring: Source block

It is necessary to import one or more datasets in order to analyze the process data that clearly defines the problem being experienced. A source block will provide one such dataset. This could be a discrete or a continuous field dataset. The Configuration Wizard has different configuration options depending on the type of dataset.

Worked example: Source block

Steps to configure

  1. Source selection: Select the relevant source from the populated window.

  2. Source configuration: Edit the properties of the dataset. The definition of each property is given in the window below the property list.

Runnability check: The runnability error page will only be shown if there is an error, with a description of the error. If runnability is successful, the Configuration Wizard will simply continue to load the data.

  1. Field selection: Select the fields to be included in the dataset.

  2. Time window: Define the time window for data collection.

  3. Source timestamp: The Continuous Troubleshooter is ideally suited for handling time series data, where each data point is associated with a timestamp. When loading continuous data into the Troubleshooter, you need to assign a timestamp to the dataset.

  4. Source index: Indexes provide a reference to real-world events or batches produced. If a discrete dataset is chosen, the index of the dataset needs to be configured.

  5. Source quality: When loading data from an external data source, you need to set the associated data quality.

  6. Execute: On execute, the import of the fields will be seen, and the new dataset will populate the workspace.

Step 1: Source selection

Select the relevant data source from those listed. Both continuous and discrete data sources are available.

  • The source that you select determines whether the Source Index (discrete source) or Source Timestamp (continuous source) step will be displayed.

  • Default: No source is selected

  • Minimum configuration: A source must be selected.

Step 2: Source Configuration

Configuration of the Architect Source block inputs and outputs differs according to the Source chosen; refer to the correct Architect source block for specific details on each Source type:

  • Default: No file or source configured.

  • Minimum configuration: A file name/data source must be specified.

Source Runnability check:

This page is only displayed if you configured the block incorrectly. The error that will be displayed is the error that you will get when you check the runnability of the block in the Architect. Click on [<Previous] to fix the error. If the block is runnable this page is not displayed.

Step 3: Field selection

This page will show whenever the fields that you want to import exceeds the limit of available columns. See Limitations of a dataset. When you see this page, you need to select which fields you want to use. The rest of the fields will not be imported.

Step 4: Time window

This option is only available for sources that use time windows.

Using execution time means data will be requested over a defined duration relative to the execution time.

If the [use execution time] box has been checked, then the data window configured for a set date in the source configuration page will be overruled. Conversely, if the box is not checked, then data will be collected over the time window defined in the source configuration page.

The advantage of using a window relative to the execution time is that you don't need to configure the time window for data retrieval every time you want to execute the operation. The source block will provide data for the defined duration of the scheduled time window on every execute.

  • Execution time: This can be either:

    • the execution of the data recipe on loading the recipe into the Troubleshooters

    • the deployment time of the data recipe in the CSense Runtime Manager.

    • This execution time is deemed to be time at NOW, and the time window and offset period go back in time from NOW, reading the data from the historian over the defined time window.

  • Offset from execution time: This offset is a "buffer" time period, from the time of execute, back in time to the end of the time window.

  • Time window duration: Specify the duration of the time window required to provide data, in days, hours, minutes and seconds.

  • Default: The [use scheduled time] checkbox is not checked.

  • Minimum configuration: None.

Step 5: Source timestamp

Select if you want to use a timestamp or not. If you select to use a timestamp you need to specify if you want a master timestamp, or a timestamp field for each field. This will list a timestamp field after each field. This will list a timestamp field after each field.

  • This step is only available for continuous data.

  • If your file/ data source contain too many fields, the Create a timestamp for each field tick box will be disabled.

  • Timestamp alignment will be disabled if you select to create a timestamp for each field.

  • Selecting timestamp alignment will ensure that each row contains a unique timestamp.

  • Default: No file or source configured.

  • Minimum configuration: A file name/data source must be specified.

Step 6: Source index

Select if you want to use an index or not. If you select to use an index, you need to specify how this index will be created. The options are to either create one from an existing field, or to create a new index.

  • If you select to use an existing field as index, the one specified in the IO Editor step will be used.

  • The Create new index will create an index starting on 0.

  • Default: The No index tick box is selected.

  • Minimum configuration: Any one of the three radio buttons must be selected.

Step 7: Source quality

Select the way that the quality of each row of the fields will be shown. You can either choose to not show the quality, to set bad quality values to null or to create a quality column.

  • When you select to set bad quality to null, each row of a specific field that is assigned a bad quality will receive a null value.

  • When you select to create a quality column, you can either create a master column or an individual column for each field.

  • The create master quality option will list the worst quality of the entire row (across all the fields).

  • The create quality for each field will create a quality column for each field.

  • In this operation 0 refers to bad quality, and 192 refers to good quality.

  • Default: The No quality option is selected.

  • Minimum configuration: Any one of the three radio buttons must be selected.

Step 8: Execute

The selected source and the configurations completed within the source will be given in a new dataset.


Related topics:

  

CSense 2023- Last updated: June 24,2025