Dr. Wes Womack, Director of Engineering from Epsilon Technology talks to AZoM about non-contact strain measurements for materials testing and how to make measurements accurate and consistent to avoid common errors.
Since we last spoke can you tell us about any new updates and products you have released at Epsilon Technology?
Interest in non-contacting extensometers has been on the rise for a number of reasons. We have placed a great deal of attention on this technology in the last year and recently announced a new product series with some significant innovations.
Why are materials testing labs turning to non-contacting extensometers?
Two prominent reasons are a desire to test more specimens per hour (higher throughput) and to cover many gauge lengths and measuring ranges with a single sensor. Non-contact methods are also attractive in applications where specimens are very thin, compliant, or may be damaged by contacting extensometers, and for tests with large energy release at failure where a mechanical extensometer might be damaged if left in place.
What technologies are typically used for non-contact strain measurement in materials testing applications?
Video cameras are the most widely used sensor platform, followed by traditional laser-scanning systems. Of the video systems, point-to-point video extensometers are used for practical quantitative materials testing strain measurements, while full-field DIC methods provide displacement data or qualitative strain data for structures and components.
What barriers have prevented more widespread use of non-contacting strain measurement methods?
Primary challenges have been achieving sufficient resolution, accuracy and speed in real-time use. Additionally, many users of video imaging systems have complained that they take too long to set up and use, and require the storage and post-processing of massive video files.
Testing standards such as ASTM D3039 specify extensometer resolution and accuracy minimums, and some non-contact extensometers have struggled or failed to meet those specifications. Getting high frame rates and data rates has been another challenge; yet those features are necessary for performing non-contacting strain control which is increasingly desired for metals testing.
Additionally, many lab managers require simplicity for their operators - managers want minimal interaction with the extensometer in order to increase throughput. Most video extensometers available today require a lot of time for training, tuning and calibration to optimize the extensometer’s settings and use them correctly, as well as software interactions every time the specimen is changed. Consequently, two of our major design goals were to eliminate the need for software interactions when performing repetitive tests, and elimination of any video data post-processing.
What range of technologies are used in the Epsilon Technology system?
We found that achieving the resolution and accuracy required for small strain measurements typical with composites and metals required a combination of several technologies. Managing the light path, use of telecentric optics, fast and accurate image processing algorithms, specialized optical calibration techniques and even extensometer mounting on the test machine all turned out to be important factors. Using a high-speed, high-resolution camera sensor is necessary but we found that this is not sufficient - combining all of the technologies mentioned is critical.
In addition, we designed a system architecture that eliminates a second computer, providing several significant benefits. These are key technologies, especially for challenging high speed, low strain applications. The new system also integrates lasers for fast and accurate optics setup.
What are the most important considerations for reliable and accurate non-contact strain measurements? Why?
Getting good performance depends on having enough resolution and speed, as well as avoiding calibration mistakes. Users will need to pay attention to resolution in order to meet the required accuracy class. For example, ISO 6892 and 527 require resolution of 1μm or better, depending on the application, and not all non-contacting extensometers can do this. Similarly, ISO 6892 recommends running tests in strain control, which requires very fast image processing for real-time strain measurement. Non-contact extensometers that rely on post-test image analysis cannot be used for strain control.
There are many other considerations such as compensating for out-of-plane motions using telecentric lenses, verifying the system calibration over its entire field of view, and correcting for uneven lighting.
What are the most neglected considerations for reliable non-contact strain measurements?
One of the most common problems is mounting the camera on a tripod. All it takes is a bump or someone thinking they can move the tripod and put it back. Unless the calibration is performed all over again, test data after that moment will be uncalibrated. That’s why rigidly mounting directly to the test machine is essential.
Another source of errors is using conventional lenses when “out of plane” motions are present. For example, a “flat” test specimen will often flex and straighten out throughout a tension test. With conventional lenses, this out-of-plane motion will cause a measurement error because the specimen appears smaller. This error can be quite significant when strain measurements are about 2% or less, common when testing high-modulus materials.
The solution is to use a telecentric lens, a type of lens that results in no change to the image size when the specimen moves out of plane, eliminating this source of error.
Additionally, it is important to check how the non-contact extensometer’s accuracy class was determined. Usually the manufacturer will state a minimum gauge length for a particular accuracy class, and so the gauge length of the test specimen has to be considered. Furthermore, it’s worth checking the extensometer’s verification report to make sure that the extensometer has the necessary accuracy at the strain values to be measured, at the gauge length that will be used. An extensometer verified at a long gauge length or high elongation may not meet the required accuracy class when measuring strains from 0-1% or with a smaller gauge length.
How can video extensometers be used to improve testing cycle times and throughput?
Labs with high testing volumes can look for non-contact extensometers that automatically detect specimen marks and start measuring strain without any pre-test or post-test software interactions. This leads to the potential for semi-automatic or fully automatic strain measurement where the test operator only has to change specimens and never touches the extensometer. The benefits are faster specimen throughput in busy labs and fewer mistakes.
For fatigue testing, faster update rates can be used to speed up the test itself; a video system with 30 Hz data rate will take 100x longer to complete a million-cycle test than a video system with a 3000 Hz data rate.
Are non-contact extensometers used for cyclic fatigue testing or strain controlled-testing? Can they be used for high-rate tensile testing? How?
Non-contact extensometers work like contacting extensometers for fatigue testing and strain control: the extensometer must respond fast enough, in real time, to keep up with the fatigue cycles or the strain control loop. This puts the emphasis on fast processing with minimal latency. Put simply, faster extensometer real-time data rates enable higher fatigue test frequencies and better performance in strain control.
Physical dynamics (vibration) can also be a limiting factor. Some non-contact systems are insensitive to vibration and can outperform contacting extensometers in this regard, but it is important to choose a system with robust, rigid mounting.
With these factors in mind, high speed non-contacting extensometers that have no moving parts can potentially offer much higher fatigue testing frequencies.
Non-contact extensometers with high frame rates and tracking speed also offer new capabilities for high-rate tensile testing. For example, an extensometer with a 3000 Hz frame rate can measure 300 data points in a tensile test that only lasts 100 milliseconds.
What other types of extensometers does Epsilon Technology manufacture?
We make over 30 models of extensometers, COD gages and calibrators to handle many applications - environments from liquid helium to 1600 °C, axial, shear, torsional, averaging, bi-axial, and more.
Where can our readers go to find out more?
For more information about video extensometers, extensometer calibrators, or extensometers in general, visit www.epsilontech.com. Readers who want to keep up to date can also sign up for our newsletter.
About Dr. Wes Womack
Dr. Wes Womack is Director of Engineering at Epsilon Technology. He joined Epsilon in 2011 and his background includes mechanical design, controls, and signal processing. He leads Epsilon’s engineering team as they develop specialized extensometers and improved strain measurement technologies, as well as ensuring compliance of Epsilon’s products with ISO and ASTM standards. Wes is active in ASTM standards committees for mechanical testing and transducer calibration.
Wes is also involved in the Jackson, WY community as a mentor for the high school robotics club. He has also given seminars on materials science to high school students with a career interest in engineering.
Wes holds a PhD in Mechanical Engineering and a PE certification.
Disclaimer: The views expressed here are those of the interviewee and do not necessarily represent the views of AZoM.com Limited (T/A) AZoNetwork, the owner and operator of this website. This disclaimer forms part of the Terms and Conditions of use of this website.