The Data collector I/O is used when performing streaming measurements. During streaming measurements, data arrives in consecutive chunks with a size equal to the instrument's record length. To form a continuous stream of data, these chunks must be appended to each other. The data collector does this job. It will fill its data with the arriving chunks of data, see the picture below.
When a Data collector is used in combination with other I/Os which process the data, there are two ways to connect them to each other, each with its own advantage and disadvantage:
The first method is preferred in most situations. When the original data is important, it is always possible to create a second Data collector to collect the original, unprocessed data.
When a graph contains a Data collector, it is possible, while the Data collector is being filled, to zoom in on the latest data and have the graph automatically keep the latest data visible. Choose the appropriate zoom factor in the graph and right-click the time base axis and select Follow source from the popup menu. This will show a sub menu with the possible sources to follow and an option not to follow a source.
When streaming data needs to be stored on disk, use a Disk writer sink which can store streaming data directly. Do not store streaming data that is collected in a Data collector I/O using a Disk writer. Each time the Data collector receives new data, this will be signaled to the Disk writer, which will then add the complete Data collector content to the file, not just the new data.
When for the property When full the option Save to disk and clear is selected, each time the Data collector is full, a .TPO file containing the Data collector is saved to disk. The Data collector is then cleared and starts collecting again. The file name for the .TPO files starts with the date and time of starting the measurement, followed by a serial number that increments each time the Data collector is full. Additionally an index file is stored keeping track of all saved .TPO files that belong to a certain measurement. The name of the index file contains the date and time of starting the measurement. The location where the files are stored can be set in the application settings. When the setup contains multiple Data collector I/Os, all saving to disk, they will all be included in the same index file. When the measurement is stopped, the final, incomplete chuck is saved as well.
When the measurement is started directly from a data logging Quick Setup and more than one chunk was saved, a dialog will appear when the measurmeent is stopped, asking to load and show all saved data.
When the measurement was setup manually without using a Quick Setup, the index file can be opened in the Multi Channel oscilloscope software, using File → Load Data collector index file. This will load a block of two, three or four of the saved Data collector .TPO files, depending on the data length of the saved Data collectors. The loaded Data collectors are shown together in a graph. When the original setup contained multiple Data collector I/Os, the .TPO files with matching times will be loaded and displayed together. Additionally, a special toolbar is shown that allows to navigate through the available .TPO files that belong to the specific measurement.
Clicking the indicator label on the toolbar allows to jump to a specific part of the measurement. A dialog will be shown in which the number of the required part can be entered.
It is also possible to manually load all .TPO files into the Multi Channel oscilloscope software. This will create Data collectors with the saved data. When the Data collectors are then all shown in one graph, the various parts will be displayed after each other, giving an overview of the total measurement. Note that depending on the amount of Data collectors and their data size, this can be very memory consuming.
To control the behavior of the Data collector I/O, several properties are available. These can be accessed through a popup menu which is shown when the I/O is right clicked.
The Data size property sets the size of the output block (the "record length") of the Data collector. Several default values are available, as well as a user defined setting. The size of the collected data can be set to a maximum of 50 million samples.
The Fill mode property sets how the data collector is filled with data.
The When full property determines what action the Data collector performs when the output data block is full.
The Data collector I/O can be manually cleared with the Clear action.