Vous êtes sur la page 1sur 5


EXPERIMENT: COMPRESSION TEST OBJECTIVES: (1) To conduct a compression test on three types of wood and obtain material properties for the tested samples. (2) To perform statistical analysis on the results to obtain sample means, standard deviations, confidence intervals, and other appropriate statistics. INTRODUCTION: In this experiment, three types of wood will be tested to failure in compression. Several material properties will be determined for each specimen. BACKGROUND: When a simple compressive load is applied to a specimen, the following types of deformation may take place: elastic or plastic shortening in ductile materials, crushing and fracture in brittle materials, a sudden bending deformation called buckling in long, slender bars, or combinations of these. Ductile materials, such as mild steel, have no meaningful compressive strength. Lateral expansion and thus an increasing cross-sectional area accompany axial shortening. The specimen will not break: excessive deformation rather than loss of strength often characterize failure. Brittle material, such as the wood specimens that are to be tested in this lab, commonly fracture along a diagonal plane which is not the plane of maximum compressive stress, but rather one of high shear stress which accompanies the uniaxial compression. Strain is a measure of the intensity of deformation (deformation per unit length). Normal strain, , measures the contraction (or elongation) of a body during deformation. avg = n/L Factors that affect the properties of wood include the arrangement of the grain and the amount of heartwood (the dark core wood of the tree). Irregularities also affect material properties. There are three important classes of defects: 1.) knots, 2.) checks, and 3.) shakes. Knots are the areas of the trunk in which the wood surrounds the base of the branch as the tree grows. Checks are longitudinal cracks that run normal to the growth rings and shakes are cracks that run parallel to the growth rings. Wood is anisotropic which means that properties will be different in different directions. When wood is loaded in compression parallel to the grain direction, it will resist large forces. However, if it is loaded transverse to the grain direction it can be quite weak. Wood, when loaded in compression parallel to its grain, is one of the strongest structural materials in proportion to its weight. The compression strength to weight ratios of some woods exceed that of structural steel by as much as 80 percent and exceeds 2024-T4 aluminum alloy by more that 50 percent. Wood is relatively weak in shear parallel to the grain, and will often fail in this mode. THEORY: Figure 1 illustrates the test specimen under compression loading. In cases where the deformation is non-uniform the average strain may be significantly different than the true strain at a given point. WOOD: Wood consists of tube-like cells which are tightly cemented together to form a basically homogeneous material. The cells, which mostly run in the same direction, form fibers, which constitute the grain. Important physical properties are moisture content and density. These properties are related to the mechanical properties of the wood.



8". Exact measurements must be made for each specimen prior to testing. P EQUIPMENT TO BE USED: b d L MTS Testing Machine 55,000-lb capacity. SAFETY: !! USE SAFETY PLEXIGLASS SHIELD ON MTS MACHINE AT ALL TIMES WHEN TESTING WOOD !! Never operate the MTS machine when someone's hands are between the grips. Make sure all lab participants are clear of equipment before beginning or resuming testing. P FIGURE 1 The test consists of uniaxial loading, and therefore the stress is calculated by: P A P is the applied load A is the cross-sectional area Specific gravity is defined as the density of a material, , divided by the density of water. SG= water Where: water= 1.94 slugs/ft3 [ 1 g/cm3] = PROCEDURE: TEST PREPARATIONS: The weight, length, and cross-sectional dimensions of each specimen must be measured prior to testing. TEST DATA: The student will need to produce a stress-versusstrain diagram for each of the three tests, similar to the one shown in Figure 3. Note that the load versus stroke curve may not contain an initial straight-line portion. If not, you will need to estimate the best fit tangent to the curve to obtain the Modulus of Elasticity, E. In doing this you may find that the tangent line intercepts the horizontal axis to the left of the curve. If this is the case, the point where the tangent line intercepts the horizontal axis should be selected as the location for the origin. Thus, it is advisable to zero the plotter pen about an inch from the left-hand side of the graph paper.


Note: specific weight is defined as: = g

MATERIALS TO BE TESTED: Three types of wood will be tested; red oak, yellow birch, and ponderosa pine. Specimens have been precut into blocks of approximately 1-3/4" x 1-3/4" x



Stroke 0.02 Time
Best-Fit Estimate Compressive Strength

in min



TESTING PROCEDURE: 1.) Create specimen file comp*. 2.) Center specimen on lower loading platen. 3.) Lock MPT and select specimen. 4.) Start scope. 5.) CLOSE SAFETY SHIELD 6.) Press `RUN' and let test proceed until rupture. The load will drop off at this point. It is not desired to crush the specimen beyond the first major rupture. 7.) Press `STOP'. 8.) Unlock MPT. 9.) Adjust SET POINT to 0.0. 10.) Remove specimen 11.) Repeat procedure for each remaining specimen. 12.) Turn hydraulics `OFF' . 13.) Copy data files to diskette. c:\em327data\comp*\specimen.dat 14.) Delete specimen comp*. REPORT: The report outline found in Appendix A should be used. REPORT REQUIREMENTS: (1) Stress versus strain plots from MTS data file complete with: Curve Fitting to obtain Tangent Modulus Labels and Title (2) Determine the following properties for each specimen: a. Proportional Limit, pl b. Compressive Strength, c c. Modulus of Elasticity, E


Proportional Limit

Note that Load vs Stroke (Pv.) will be recorded at the MTS. These values can be converted to stress-strain (v.)


Several other calculations will be required based on the results of the test. For additional information, please consult the tension test experiment write-up. E= / pl = Ppl/A c = Pmax/A UR = 1/2(pl)(pl) = pl2 2E SG= water

MTS SET-UP: 1.) Follow Start- up Procedures Station Manager compress MPT compress.000 2.) Turn hydraulics on. 3.) Make sure 'MANUAL OFFSET' = 0 for Stroke. 4.) Adjust 'SET POINT' to 0.0 5.) 'AUTO OFFSET' Load. 6.) Set-up Scope to plot a/b. Load 5000 lbf -10,000

d. Modulus of Resilience, UR



e. Specific Gravity, SG (3) Compare a,b,c, and e to reference values (4) Tabulate all values calculated in (3) and (4) and neatly show one set of sample calculations. (5) Your instructor will provide you with a data sheet from previous compression tests done on wood samples like the ones you tested. Add your results to the appropriate data and perform statistical analysis to determine the mean, median, mode (if applicable), standard deviation, and confidence interval for a 90% confidence level for each data set. (A brief review of statistical terms and calculations can be found on the next page) Show one set of sample calculations and tabulate your results. Discuss the statistical results and their implications. Examine the mean, median, and mode values and discuss which of the three may be more representative for the different data sets and why. What does the standard deviation tell you about the scatter in the data of each set? What kind of conclusions can you make about the material properties of wood based on your statistics? The EXCEL worksheet is located at: http://www.iastate.edu/~em327/ experiments/compression/database (6) Discuss sources of error as well as their impact on the design process. (7) Describe the types of failure observed for each specimen. SKETCHES OF THE FAILURES ARE REQUIRED. (9) Answer the questions your instructor assigned. QUESTIONS: (1) Are the compressive strength and the specific gravity related? If so, what trends do the data indicate. (2) Strain calculations based on the measured stroke may not be very accurate when premature crushing occurs at the ends of the specimens. Discuss how this would effect the experimental values determined in (3).



STATISTICS: A measure of central tendency refers to a measure indicating the value to be expected of a typical data point in a data set. Three measures of central tendency will be considered: ARITHMETIC MEAN: The central tendency measure representing the arithmetic average of a set of observations. MEDIAN: The middle point of an ordered set of data. (If an even number of data exists the median is the average of the middle two) MODE: The value(s) most often repeated in the data set. May or may not be applicable. Units for each of these are the same as the data units. In a large enough sample size, which does not violate statistical assumptions of randomness and independence one would expect these values to be the same. Thus, one can use these values to get a qualitative estimate of the true measure of central tendency by examining how close they are to each other, or if one is much different. In the latter case the data set can be examined for outlying data points which can strongly affect the mean. Outliers much less impact the mode and median. The STANDARD DEVIATION, , is the positive square root of the variance and represents the scatter (dispersion) in the data set. The units of the standard deviation are the same as the data set. The standard deviation of a normally distributed data set is given by the following equation: would include 95% of the observations, and within 3 standard deviations one would expect to find 99% of the observations. Typically, engineering applications consider a 3 standard deviation limit. When more scatter exists in the data, the bell shape will be flatter, where a data set with little scatter would have a steep bell shape. Note that the tails of the curve approach the x-axis, but will never reach it, thus, it is possible to obtain any value for a given data set, in theory.

Normal Distribution
0.10 0.08 0.06 0.04 0.02 0.00 0 40 80 120

n xi2 ( xi ) n(n 1)

where n is the number of observations. The figure illustrates the shape of a normal distribution. Within one standard deviation from the mean value one would expect to find 68% of the observations, two standard deviation within the mean