Building a Geodynamics Vibration Monitoring System with NI LabVIEW and SCXI Hardware

Author(s):
Jerry R. Hall - Quantum Controls, Inc.

Industry:
Research

Products:
SCXI-1000, LabVIEW, SCXI-1141, SCXI-1140

The Challenge:
Acquiring, analyzing, and displaying in-field multi-channel geophysical vibration data that needs to be acquired in both software triggered and free-streaming modes with possible concurrent analysis during the latter.

The Solution:
Creating a portable PC-based system with National Instruments data acquisition and SCXI modules using LabVIEW for data acquisition and analysis.

In-field geophysical vibration data is used to characterize soil conditions for construction and other purposes. The data is acquired from accelerometers placed on the ground. The acquired data consists of the measured time response of the variously placed accelerometers as the vibrations from an applied shock radiate from the point of application, however, the operator may need to acquire and record ambient vibrations such as nearby traffic.

The new equipment consists of an NI SCXI 1000 rack paired with a SCXI-1140 simultaneous-sampling differential amplifier module and a SCXI-1141 low pass filter module connected to a portable PC via an NI AI-16E-4 DAQCard. This system allows the simultaneous reading of up to 16 channels of data. The cutoff frequency and gain of the SCXI-114x modules are software configurable allowing the operator flexibility in conditioning the incoming signals (the signals of interest are generally of low frequency and appropriately setting the cutoff frequency eliminates aliasing. A portable power source completes the system.

Software

The challenge of this system was to integrate the various display, analysis, and acquisition modes into one application. Fortunately the ease with which multiple threads can be programmed in LabVIEW made the task much simpler. The program is written around three primary concurrent while loops and an intermittent fourth.

The first loop is concerned only with monitoring and setting the data display parameters (color, skew, offset, gain, selected channels, etc.). These parameters are monitored for change and displayed for any of the various channels independent of whatever other tasks the program is performing. The second loop is for data display and analysis. New data for display is retrieved from a storage-VI and plotted according to the display parameters acquired by the first loop discussed above. When these parameters change, the change is noted and a new display performed. The data is re-plotted only when necessary (parameter change, new data, etc) in order to burden the CPU as little as possible. Cursor information is displayed from this loop adjusted for the display parameters chosen. Changing the display gain, offset, or skew does not affect the channel information returned to the operator. Finally, data from the storage-VI can be analyzed by Fourier transform, integration, differentiation, cross correlation, etc with the result being displayed in an additional data channel. In total, sixteen data channels and one analysis channel can be displayed a once.

The third loop processes tasks that, with one exception (see data acquisition modes discussed below), are not meant to happen concurrently. These include, among others, data storage and recall; setting, checking, and display of data acquisition parameters (live channels, channel gains, low-pass filter settings, scan rate, trigger value, etc); and data acquisition (three different modes). File recall and data acquisition both write their new data to a storage-VI where it is retrieved for display and analysis in the second loop as discussed above. Note: data acquisition and display operate as independently as possible of each other. All of the parameters, controls, and displays discussed above are accessed from a single screen. Appropriate use of control elements’ visibility helps reduce screen clutter. An example of the screen can be seen in figure 1 below.

Triggered Data Acquisition

The first mode of data acquisition is the triggered mode using the LabVIEW analog trigger. On receiving a signal that crosses the trigger threshold in the designated channel, a prescribed number of pre-trigger and post-trigger data are acquired and displayed. The operator can choose to average a number of triggered events.

Continuous Data Acquisition

The second mode of data acquisition acquires data continuously until stopped by the operator. A pseudo chart display is achieved by effectively appending the new data to a fixed length array while rotating and removing the oldest data off of the front. The updated array is then written to the storage-VI. This data acquisition mode uses the data acquisition occurrence method in LabVIEW (asynchronous acquisition where the software doesn’t sit idle while the data acquisition device collects the data), freeing time for the CPU to plot the incoming data.

Data Acquisition with Streaming to File plus Concurrent Analysis

The third mode of data acquisition is similar to the method discussed above plus it also streams the acquired data to a file at the same time. In addition it gives the operator the option of simultaneously viewing or analyzing older data by putting the data acquisition in the background to run unattended. In this mode the operator specifies the size and number of data files he wishes to acquire. The system then saves the data to a series of files until either the specified number of files are saved or the acquisition is stopped by the operator. This data acquisition mode is intended for long, perhaps unattended data runs.

To save memory storage space, binary data is streamed to file for storage. To save CPU time, when the data acquisition is being done in the background and the acquired data is not being plotted, the analog output of the data-acquisition-VI is disabled. When this data acquisition mode is being run, the operator can switch the data acquisition to and from background without losing the older data that is being analyzed.

Conclusion

The customer is using the system in the field and is very pleased with the increased functionality. He especially appreciates not having to page through a number of screens to view or set a parameter. The use of data acquisition occurrences frees up more time for the CPU intensive data plotting, resulting in smoother display updates. The use of concurrent while loops prevents the display from interfering with the data acquisition system and eases the programming.

For more information on this case study, contact:

Jason Swoboda
Optimation Technology, Inc.
2000 South Dairy Ashford, Suite. 150
Houston, TX 77077
Tel: 713 355-3900
E-mail: jason.swoboda@optimation.us

Browse All Case Studies »