The LVDT jacket that was used to perform tests. A cylindrical sample such as seen in the background was placed in the jacket, then set in the press.
This was designed as my summer employment for 2000. The Mining Engineering department on campus had an antiquated system for exciting and measuring the output of an LVDT (Linear Variable Differential Transformer) jacket that was used in stress-strain tests. (See below) My task was to digitize the system, and make it collect data directly to a PC for furthur analysis. It needed to be able to resolve horizontal and vertical strain to the millionth of an inch, but only needed to sample less than 2 times a second.
I ended up using 2 Analog Devices 24-bit 8-channel delta sigma converters to meet the resolution and sample rate requirements along with multiple Analog Devices integrated LVDT conditioner ICs. 7 channels were measured, 3 for axial strain, 3 for lateral strain, and 1 for the load applied. 2 A/Ds were needed to keep the overall sample rate high enough. A PIC microcontroller was used to drive both converters and provide a serial interface to the host PC.
The software side of the project involved creating LabView programs to calibrate the equipment and perform the tests. It was my first experience with instrumentation user interface design, and was very educational. Working with the professors in the Mining Engineering department taught me many lessons in ergonomics. As part of the project I wrote documentation which not only described how to operate the apparatus, but gave detailed technical drawings and schematics for future expansion and/or repair.
I completed the project on time and $10,000 under budget. (A load cell was budgeted, but I was able to utilize the testing machines built-in load scale instead.)
The LVDT jacket that was used to perform tests. A cylindrical sample such
as seen in the background was placed in the jacket, then set in the press.
This is the assembly that my box replaced.
My equipment with the lid removed.
A close-up view of the inside of the box. Since only one was needed, I
hand-assembled it on a prototyping board.
I have some screenshots of the Labview program. Just need to find them.