Strength and material law of concrete under biaxial impact
Table of contents
Project data
|
Titel | Title |
Report in the yearbook 2012
Fast and loud
Destroyed concrete specimen after a trial in the biaxial Split-Hopkinson bar
There is not much to see, but there is some audible - something like this can be thought of a trial at the Split-Hopkinson bar. A short, loud bang and an 80 mm long concrete cylinder bursts into a thousand fragments. That‘s impressive, but it raises the question: Why bother?
With the help of such an experimental setup material properties should be studied under very high loading rates, such as occur, for example, in a vehicle collision. At the Institute of Concrete Structures since two years a classical uniaxial Split-Hopkinson bar is available for dynamic pressure and tensile tests on concrete samples. Moreover, the world‘s first biaxial Split-Hopkinson bar is developed. The aim of this is to load the concrete specimen impulsively in two directions at the same time. What initially sounds quite simple turns out on closer examination to be a very complex task, as the two pressure pulses have to be synchronized to within a few microseconds to assume a really biaxial loading of the specimen. Conventional techniques for the loading and triggering, like spring guns, swing hammers, bursting membranes or explosions do not provide sufficiently precise release. Therefore, compressed air guns were chosen to accelerate the impactors.
For the measurement of all interesting system parameters, the experimental setup is equipped with 16 semiconductor strain gauges, four pressure sensors and two light barriers. The control of the trigger times with microsecond accuracy is taken over by a computer-controlled microcontroller, by which constant time differences of about 3 ms between the two accelerator facilities can be compensated. The dependence of the time delay from the air pressure in the acceleration system must be taken into account. Other variations and inaccuracies in the synchronicity could be minimized by reducing the pneumatic ways, the exact positioning of the impactors in the air guns and the startup of the machine by several hundred trials. So it has been possible to generate load impulses in both axes with a time delay of less than ten microseconds. The further optimization will occupy us in the future.
Report in the yearbook 2011
High-speed dynamic behaviour of concrete
Preparing the test
The basic concept of the Split Hopkinson Bar (SHB) was developed in 1949 to study materials under dynamic conditions. Today this experimental setup is an established method to characterise the behaviour of materials under high strain rates (impact). The classic test setup consists of an acceleration device, that accelerates an impactor to a designated speed, and two long, usually cylindrical rods, between which the sample is sandwiched. By the strike of the impactor on the first rod, a pressure wave is induced, which runs through the first rod, the sample and the second rod. This allows loading a specimen with very high uniaxial strain rates that correspond, for example, to a vehicle collision, plane crash or falling rocks.
Previous researches to the velocity dependent behaviour of various materials have shown that the compression and tensile strength of materials increase with rising strain rate. For concrete, the increase of tensile strength, by a factor of six to seven, is significantly larger than in compressive strength which can only be increased two- to threefold. It is also known that the concrete strength increases under multiaxial compressive loading.
That’s why the question arose: Can these two effects superimpose, so that under multiaxial high dynamic loading concrete can resist higher loads than under static conditions? A current research project focuses on the answer to this question. For this, cubical-shaped specimens shall be loaded simultaneously in two directions by an impact load. Because impacts occur within very short periods of time, usually a few milliseconds, the biggest challenge for the planned experiments is the exact synchronisation of the loads from the two directions. This accuracy shall be achieved by two gas guns that accelerate the impactors by compressed air. The supply with compressed air and the release of the gas pressure guns are controlled by computer-based microcontrollers.
The stress-strain behaviour of the concrete specimen is determined indirectly via the strain in the rods. Therefore semiconductor strain gauges are installed in the middle of all rods, which can be read with very large sample rate. Additionally, the displacements of the specimen edges can be detected by a high-speed extensometer.
Report in the yearbook 2010
Concrete behaviour under impact
Test setup of the Split Hopkinson Bar
Under which conditions do the strength properties of materials change? This project investigates the influence of the load velocities on the strength of concrete. To analyse this aspect a Split-Hopkinson-Bar is used. The experimental setup consists of a system, to accelerate an impactor, and two long aluminium transmission bars with a concrete test specimen between them. In the acceleration system, the cylindrical impactor is accelerated and strikes the first aluminium bar. Thereby a compression wave is triggered in the aluminium bar. This pressure wave proceeds with the velocity of structureborne sound through the transmission bar, the concrete test specimen and, finally, the second transmission bar. As in a black-box process, over the incoming, outgoing and reflected test results, a conclusion can be drawn on the properties of the concrete test specimen during the experiment. The aluminium bars behave elastically during the trial, while behaviour of the specimen is plastically.
Previously known studies have investigated the influence of high load velocities in the uniaxial case. In this research project specifically the effect of the biaxial loading of a test specimen shall be studied. For this purpose a concept for an experimental setup was designed last year that shall enable the synchronised loading of the concrete test specimen from two directions. Different loadings and triggering concepts were compared and checked for their feasibility. Based on these analyses, compressed air was selected as suitable propellant for the impactor. The trial is pneumatically controlled. Semiconductor and laser measurement, providing sampling rates in the MHz range, make it possible to measure and analyse such fast pulse operations.
This concept has already been tested in a classical uniaxial Split-Hopkinson-Bar. After the appropriate adjustments for the biaxial experimental setup, now the biaxial test rig is set up.
Report in the yearbook 2009
Concrete under dynamic load
Concrete can be subjected to impact or alternating loads in different situations. For example, pieces of rock can fall onto concrete roads; vehicles can crash into concrete structures during accidents; or supporting structures of factories and manufacturing plants can be subjected to dynamic loads from production processes or machine malfunctions. High strain rates occur in such load cases resulting in rather quick changes to concrete strains.
The goal of this project is to investigate the behavior of concrete under high load velocities with an emphasis on the increase in strength properties of concrete. In particular, which resistance concrete can appose to dynamic load compared to quasi static loading.
One method used to investigate high strain rates is the Split-Hopkinson-Bar method in which the passage of a pressure or tensile wave through a concrete object is analysed. The concrete test specimen is located along a straight axis directly between two long transmission bars. A pressure impulse is applied to one end of the transmission bar which then travels as a pressure wave, first, through the bar, then through the concrete test specimen and, finally, trough the second transmission bar. By measuring the wave as it passes from in front of and behind the concrete test specimen conclusions can be made about how the wave changed inside of the specimen. A sufficiently large sampling rate of signal recordings must be made to accurately analyze high strain rates; therefore, a sampling rate of at least 500,000 per second must be reached.
The Split-Hopkinson-Bar will be enhanced to two dimensions in a future stage of this research project in which concrete behavior under bi-axial dynamic load is to be investigated. Contrary to the one-dimensional case, here the concrete test specimen is to be located in two axes between four transmission bars. The triggering of the load impulse is an enormous challenge for the experimental set-up as impulses must be precisely synchronized so that the load is applied at both bar ends simultaneously (i. e., transmitted to both axes so that the two pressure waves reach the concrete specimen at the same time and phase).
An additional goal of this research project is to incorporate the knowledge from experimental work using the Split-Hopkinson-Bar method as part of a multi-axial material model.