Testing

In the beginning of the project, fall semester, we began by researching signal amplifier and conditioners, digital oscilloscopes, foil and semi-conducting strain gauges, data acquisition and analysis software, data conversion and data storage options.  After doing this we had a better idea of what equipment would best accomplish the needs of the project and made a recommendation for each of the previous instruments.  At the beginning of the semester a trip was made to view the actual testing apparatus and watch test being preformed at Eglin Air Force Base .It was at this point we then found out that not all of the equipment needed to be updated and that Eglin Air Force Base would continue to use their own signal amplifier and conditioner and oscilloscope and that the main focus of our project would now be develop a prototype to test the capabilities of LabView, our recommended data analysis and acquisition software, and to verify that we could actually receive a signal from the foil strain gauges mounted to the prototype.  In order to this materials were purchased and a prototype was constructed as shown in figure 1. 

Figure 1

The prototype was constructed using a steel metal bar, modeled as the incident bar, secured to a wood platform using teflon bearings.  Four strain gauges were then mounted to the bar using epoxy and cured over the course of two days. A wood and styrofoam mixture was used as a momentum stop. Once the curing process was over the leads of the gauges were then soldered using standard electrical copper wire.  It was at this point that two of the gauges become disconnected from the bar.  To compensate for this problem two separate Wheatstone bridges where made on two different breadboard, each using three 330 ohm resistors and a ~350 ohm strain gauge  to complete the circuit as seen in figure 2.  

Figure 2

In order to test the compressive and tensile waves within the incident bar the circuits were excited by a 12V power supply and configured using a DSpace DS1104 DSP board and real-time control using simulink. The DSpace DS1104 DSP board is a stand alone computer that resides on a PCI board inside the PC desktop that operates independently of the PC and Windows.  It was into this board that the input from the Wheatstone bridge would be directed.  Simulink is a block diagram control system modeling program that allows one to simulate a closed loop control system using a block diagram as opposed to programming.  In order to use these  two pieces of software it was required that we had access to the MatLab program. This is becuase it was from there that Simulink was opened and a model was created, as shown in figure 3, as well as the verification graphs. 

Figure 3

The gain values were set at 100 for each gauge and a filter was designed to replicate the conditions needed in a signal amplifier and conditioner.  The filter consisted of a minimum order Butterworth low-band pass filter with a sampling frequency of 5000hz, band pass of 2hz. and band stop of 200hz.  This filter was chosen because it provides a smooth response at all frequencies while still remaining maximally flat.  This model was then "built" in Matlab.  The DSpace control desk, which is the program that receives the incoming information from the DSpace Board, was then opened and graphs were constructed to represent velocity vs. time.  The bar was then struck by the hammer and readings were taken.   These readings were then saved and a program was written using Matlab to display our findings.  

From there, the same physical setup of the bar with the mounted strain gauges, 12V power supply and the two Wheatstone bridge circuits were connected to the DAQ board that led to the computer that ran the LabView program.  Through programming using block diagrams, in a similar fashion to that used in Simulink, the LabView program was written.  The completed block diagram can be seen in figure 4 and the portion of the program written specifically for data acquisition is pictured in figure 5.  As seen in figure 5 a gain amplifier and filter were also added using the same specifications as in the previous program.

Figure 4

Figure 5

Again the bar was struck and readings were taken to prove that LabView has the capability to acquire voltage vs. time graphs from the bar.  It also proved that LabView also offers the possibility of eliminating other equipment such as the oscilloscope and signal amplifier and conditioner, thereby outputting a clearer more precise reading without excess noise and internal resistance.  A reading taken from LabView can be seen in figure 6.

Figure 6

From this verification we are now able to fully recommend LabView as a data acquisition and analysis software to Eglin Air Force Base.  In the future we hope to mount semi-conducting strain gauges to the same bar and conduct the same type of analysis on them and then compare the results to the results obtained from the foil strain gauges.