QUALITY TEST & INSPECTION

S uccess in today’s marketplace requiresimprovements in efficiency, quality andaccuracy of testing facilities and testing

equipment. Testing machines are used to devel-op better information on known materials or todevelop new materials and maintain the qualityof the materials. For materials suppliers, themechanical properties tested by these machinesare an important measure of product quality,and testing is required for certification.

In a broad sense, strength refers to the abilityof a structure to resist loads without failure.Failure may occur by rupture because of excessivestress or may take place owing to excessive defor-mation. Tensile properties include the resistanceof materials to pulling or stretching forces. Theamount of force required to break a material andthe amount it extends before breaking are impor-tant properties. For most materials, the initialresistance to force, or modulus, and the point ofpermanent deformation, are obtained from plotsof force against elongation. Analysis of force-elongation or stress-strain curves can conveymuch about the material being tested, and it canhelp in predicting its behavior.

Machine design There are many types of testing machines. Themost common are universal testing machines,which test materials in tension, compression orbending. The primary use of the testing machineis to create the stress-strain diagram. Once thediagram is generated, a pencil and straight edgeor computer algorithm can be used to calculateyield strength, Young’s Modulus, tensile strengthor total elongation.

There are two classes of testing machines,e lectromechanical and hydraul ic . Theelectromechanical machine uses an electricmotor, gear reduction system and one, two orfour screws to move the crosshead up or down.A range of crosshead speeds can be achieved bychanging the speed of the motor. A micro-processor based closed-loop servo system can beimplemented to accurately control the speed ofthe crosshead.

A hydraulic testing machine uses either asingle- or dual-acting piston to move thecrosshead up or down. In a manually operatedmachine, the operator adjusts a needle valve tocontrol the rate of loading. In a closed-loop

Tensile Testing Basics,Tips and TrendsMechanical

properties are an important

measure ofproduct quality,

and tensiletesting is just

one way tocertify the

product.

BY RICHARD GEDNEY

| TECH TIPS |◆ There are two classes

of testing machines,electromechanicaland hydraulic.

◆ ASTM requires thatload and strain meas-uring devices be cali-brated annually ormore frequently.

◆ Measuring smallstrains typical of ahigh-strength metalstest—0.0001 inch orless—is the task of anextensometer.

There are two classes of testing machines—electromechanical and hydraulic. Photo: ADMET Inc.

Reprinted from Quality, Copyright January 2005.

hydraulic servo system, the needle valveis replaced by an electrically operatedservo-valve for precise control.

In general, the electromechanicalmachine is capable of a wide range oftest speeds and long crosshead displace-ments, whereas the hydraulic machineis a cost-effective solution for generat-ing high forces.

Reporting resultsIn the laboratory, there are many thingsto consider. For those signing theirname to the report, it’s important toquestion the assumptions and methods

used to obtain results. Check the ASTMInternational (American Society forTesting and Materials) test methods orother test specifications, and be surethe correct test speeds, loading profileand calculations are used.

The rate at which a test is per-formed can have a significant effect ontensile properties. Tensile propertiesfor plastics, polymers and steels arevery sensitive to testing rates, but alu-minum alloys exhibit little sensitivity.Materials that are sensitive to strainrates exhibit higher tensile strengthsand lower elongations at faster speeds.

Therefore, it is important that all test-ing rates are within the limits specifiedby the governing standard. If compar-ing results across lots or batches ofsimilar materials, insist that the testingrates be identical.

ASTM requires that load and strainmeasuring devices be calibrated annu-ally. If any device has been damaged oris subjected to excessive use, more fre-quent calibrations should be done.Calibration records should be kept todetermine when it’s time to replace orrefurbish a device. Each device has itsown certified range and, therefore, no

What is the strength of a material? Strength refers to the ability of a structure to resist loadswithout failure because of excessive stress or deformation.Typical points of interest when testing a material include:ultimate tensile strength (UTS) or peak stress; offset yieldstrength (OYS) which represents a point just beyond theonset of permanent deformation; and the rupture (R) orfracture point where the specimen separates into pieces.

TheoryA graphical description of the amount of deflection underload for a given material is the stress-strain curve.Engineering stress (S) is obtained by dividing the load(P) at any given time by the original cross sectional area(Ao) of the specimen.

S = P/Ao Eq. 1

Engineering strain (e) is obtained by dividing the elonga-tion of the gage length of the specimen (∆l) by the origi-nal gage length (lo).

e = ∆l/lo = (l - lo)/lo Eq. 2

The figure depicts a typical stress-strain curve. The shapeand magnitude of the curve is dependent on the type ofmetal being tested. Point A represents the proportionallimit of a material. A material loaded in tension beyondpoint A when unloaded will exhibit permanent deformation.The proportional limit is often difficult to calculate, there-fore, two practical measurements, offset yield strength(OYS) and yield by extension under load (EUL), were devel-oped to approximate the proportional limit. The initial por-tion of the curve below point A represents the elasticregion and is approximated by a straight line. The slope (E)of the curve in the elastic region is defined as Young’sModulus of Elasticity and is a measure of material stiffness.

E = ∆S /∆e = (S2-S1)/(e2-e1) Eq. 3

Point B represents the offset yield strength and is foundby constructing a line X-B parallel to the curve in theelastic region. Line X-B is offset a strain amount O-X thatis typically 0.2% of the gage length. Point C representsthe yield strength by extension under load (EUL) and isfound by constructing a vertical line Y-C. Line Y-C is off-set a strain amount O-Y that is typically 0.5% of gagelength. The ultimate tensile strength, or peak stress, isrepresented by point D. Total elongation, which includesboth elastic and plastic deformation, is the amount ofuniaxial strain at fracture and is depicted as strain atpoint Z. Percent elongation at break is determined byremoving the fractured specimen from the grips; fittingthe broken ends together and measuring the distancebetween gage marks. Percent elongation at break reportsthe amount of plastic deformation only. The gage lengthused for measurement is reported with the result.

elongation at break(%) = ez = 100*(lz-lo)/lo Eq. 4

Reduction of area, like elongation at break, is a measureof ductility and is expressed in percent. Reduction of areais calculated by measuring the cross sectional area atthe fracture point (Az).

reduction of area(%) = (Ao-Az)/Ao Eq. 5

ENGINEERING STRESS-STRAIN CURVE

results should be reported outside thecertified ranges.

Changes made in 2001 to ASTM E4 Standard Practices for ForceVerification of Testing Machines in-creased the minimum certifiable forcefor a given range from 100 to 200 timesthe force resolution. Resolution on dialgages and pen recorders depends on thedistance between graduation marks andthe width of the needle or pen. Often,with these devices, it is not clear whatthe resolution should be, and determin-ing it is subject to much interpretation.Based on the new E4, tests formerlyconducted at the low end of a givenforce range may no longer be valid.

Worn machine components canresult in misalignment that createsbending stresses, lowering tensile stressreadings. Check the test machine’s align-ment and play to ensure concentricity ofthe crosshead over the full travel range.

With the advent of microprocessor-based test systems, applied loads inad-vertently can be “zeroed out,” resultingin lower stress readings. To preventthis, clamp the specimen in the uppergrip, then zero the load and close thelower grip.

Potential pitfalls When testing most metals, strains areusually too small to be measured by usingtesting machine crosshead or piston dis-placement methods. Measuring smallstrains typical of a high-strength metalstest—0.0001 inch or less—is the task ofan extensometer.

If yield values are incorrect, reviewthe stress-strain diagram; the exten-someter may have slipped on the spec-imen during the test. To help preventextensometer slippage, the clampingforce and the zero point should bechecked regularly and worn knifeedges replaced.

Wedge action grips are the mostcommon style used in metals testing. Asthe axial load increases, the wedge acts toincrease the squeezing pressure appliedto the specimen. Wedge grips are manu-ally, pneumatically or hydraulically actu-ated. For high-volume testing, it is rec-ommended that pneumatic or hydraulicactuated grips be used. Worn or dirtygrip faces can result in specimen slip-page, which often renders the stress-strain diagram useless. The grip facesshould be inspected periodically. Worn

A stress-strain curve shows extensometer slippage. Inset: An extensometeris attached to a flat specimen. Photo: ADMET Inc.

Strain rate sensitivity example -Wire tie testWire ties were pulled until fracture. Table and force-extension curves showhow test speed affects peak load and maximum elongation. As the speedwas doubled from 2 to 4 inches per minute, the maximum elongation wasreduced by almost half. As the speed was increased from 2 to 16 inches perminute, the maximum force increased by 7.5%.

Speed Peak Load ElongationCurve# (in/min) (lb) (in)

1 2 29.5 3.902 4 29.6 2.013 8 30.4 1.754 16 31.7 1.72

inserts should be replaced and dirtyinserts cleaned.

Correct alignment of the grips andthe specimen, when clamped in thegrips, is important. Offsets in alignmentwill create bending stresses and lowertensile stress readings. It may even causethe specimen to fracture outside thegage length.

Most ASTM or similar test methodsrequire a shaped specimen that willconcentrate the stress within the gagelength. If the specimen is incorrectlymachined, fracture could occur outsidethe gage length and result in strainerrors. Incorrect reading of specimendimensions will create stress measure-ment errors. Worn micrometers orcalipers should be replaced and careshould be taken when recording speci-men dimensions. Some computer-based test systems will read themicrometer or caliper directly, thuseliminating data entry errors.

Can software do it all?Today’s software algorithms are good atcalculating the correct results for agiven stress vs. strain curve. It is usuallythe data fed into in the algorithm—such as the shape and magnitude of thestress vs. strain curve—that producesthe bad results. That is why the testtechnician’s job is important.

The technician must be responsiblefor doing the test according to specifica-tions. He should be aware of the poten-tial errors introduced by the machine,extensometer, grips and specimenirregularities, and also should alert thelab supervisor when problems arise. Inshort, the technician should be trainedin correctly generating the stress vs.strain curve for a given test method.

One indirect way of checking theresults is to send specimens to at leasttwo other testing laboratories so resultscan be compared.

The design and manufacture of newmaterials is broadening the scope ofmaterials testing and complicating itspractice. To remain competitive, reducetest times, eliminate data entry errorsand speed the time of product ship-ments. Using testing systems that auto-matically perform the tests according tospecification, automatically calculateresults, and seamlessly communicatewith other computers and programsrunning on the corporate network, willbe paramount to success. Q

Richard Gedney is founder and president ofADMET Inc. (Norwood, MA). For moreinformation, he can be contacted at rged-ney@admet.com or (800) 667-3220.

This diagram shows specimen fullyengaged in grip faces and grip facescorrectly supported inside grip pockets.Correct arrangement. Photo: ADMET Inc.

1/2"

Potential area of damage

Potential area of damage

The diagram shows a specimen partiallyengaged in grip faces and grip facescorrectly supported inside grip pockets.Arrangement can cause permanentdamage to grip inserts and may lead tospecimen slippage. Photo: ADMET Inc.

The diagram shows a specimen fullyengaged in grip faces and grip faces notfully supported inside grip pockets.Arrangement can cause permanentdamage to grip pockets and may lead tospecimen slippage. Photo: ADMET Inc.

Stress-strain curve shows numerous areas of specimen slippage causinginvalid test. Photo: ADMET Inc.