3 Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٤٤ USE OF COMBINED NONDESTRUCTIVE TEST METHODS TO PREDICT CONCRETE COMPRESSIVE STRENGTH CAST WITH AGGREGATE OBTAINED FROM SOUTHERN PARTS OF IRAQ Ass. Prof. Sallal R. A. Al-Owaisy, Prof. Ghazi F. Kheder, and Ass. Lect. Ahmed A. Abbas Wasit University , Mustanseriyah University, Shatra Technical Institute Abstract Concrete compressive strength is one of the most important concrete requirements that can be used to decide if the concrete is structurally acceptable or not. In several cases there is a need to estimate the concrete compressive strength on the site during construction or later on during the life of concrete. There are several methods used for this purpose, among the mostly used methods are the Ultra sonic pulse velocity and the Schmidt hammer rebound number. In this work six different fine aggregate and two different coarse aggregate were obtained from different parts of southern Iraq. Using these different aggregate combinations, 120 different concrete mixes with mix proportions of 1:2:4 or 1:1.5:3 and W/C ratios ranging between 0.40 to 0.60 were cast into 152 mm cubes. The compressive strength, ultrasonic pulse velocity, Schmidt hammer’s rebound number and concrete density were measured. These results were introduced into nonlinear multiple variable regressions to obtain correlation relationships to predict the concrete compressive strength. Two groups of regressions were formulated, the first used only the Ultrasonic pulse velocity and rebound number in the regressions, and separate regressions were prepared for each single source of aggregate. The results of the predicted strength was in good agreement with the experimentally measured values, the value of the standard errors of these regressions were less than 10% of the lowest concrete strength investigated (20MPa). In the second group of regressions, the data from all concrete mixes with different aggregate sources were combined together to obtain the correlation regressions. These regressions were formulated because in many cases in practice the source of aggregate may not be known exactly. Two subgroups were developed, with different independent variables combinations. The standard error of this group was higher than for the first group, its best value was 16% of the minimum value of concrete strength investigated. This clearly proves the importance of the aggregate source on the predicted concrete compressive strength values. Key words: Nondestructive test, ultrasound pulse velocity, rebound number, combined NDT test, strength evaluation, concrete compressive strength. لخرسانة مصنعة من ركام االنضغاططرق الفحص االإتالفي المشترك لتقییم مقاومة استخدام من المنطقة الجنوبیة للعراق غازي فیصل خضر. م.م، أحمد عبدالهادي عباس. د. أ،صالل راشد عبد العویسي. م.أ الشطرة/هد الفنيالمع، المستنصریةجامعة ال، جامعة واسط Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٤٥ :الخالصة .و الدیمومةالخرسانة أحد أهم الخواص الالزمة لتحدید مدى قبولیة الخرسانة من الناحیة اإلنشائیةانضغاطتعتبر مقاومة . االنجازالخرسانة في الموقع سواًء أثناء التنفیذ أو خالل عمر المنشأ بعد انضغاطحاجة لتقییم مقاومة ظهر في عدة حاالت ت سرعة األمواج فوق الصوتیة ورقم هي قییم مقاومة انضغاط الخرسانة و من أكثر هذه الطرق شیوعًا تلالأتالفیة عدة طرق ستخدم ت في هذا البحث تم استخدام ستة أنواع مختلفة من الركام الناعم ونوعین من الركام الخشن من مصادر طبیعیة .مطرقة شمیتارتداد خلطة خرسانیة مختلفة بنسب ١٢٠استخدمت هذه األنواع المختلفة من الركام في أعداد . من العراقمختلفة من المنطقة الجنوبیة مم ١٥٢استخدمت مكعبات قیاسیة بأبعاد . ٠.٦إلى ٠.٤سمنت تتراوح من /وبنسب ماء) ١:١.٥:٣(أو ) ١:٢:٤(خلط وزنیة طرقة شمیت وكثافة الخرسانة على هذه النماذج ثم بعد ذلك تم قیاس سرعة األمواج فوق الصوتیة ورقم ارتداد م. كنماذج للفحص استخدامت نتائج هذه الفحوصات في إجراء تحلیل الخطي متعدد المتغیرات إلیجاد . تم قیاس مقاومة أنضغاطها أتالفیا لحین الفشل في ، وعتین من عملیات التحلیلتم صیاغة مجم.عالقات ارتباط لتخمین مقاومة انضغاط الخرسانة بأستخدام الفحوصات الالأتالفیة أجریت . المجموعة األولى استخدم سرعة األمواج فوق الصوتیة ورقم ارتداد مطرقة شمیت فقط لتحدید مقاومة أنضغاط الخرسانة طأ عملیات التحلیل لكل نوع من أنواع الركام وكانت القیم المستنبطة لمقاومة انضغاط الخرسانة مقاربة للنتائج العملیة وبنسبة خ أما في المجموعة الثانیة من عملیات التحلیل فقد تم . من أقل قیمة مسجلة لمقاومة انضغاط الخرسانة% ١٠قیاسي التتجاوز أعتماد نتائج كافة الفحوصات و لجمیع الخلطات بغض النظر عن نوع الركام والسبب في ذلك أن مصدر الركام في كثیر من أعطت نتائج المجموعة الثانیة نسبة خطأ أعلى من المجموعة األولى حیث . رسانة المنشأاالحیان قد یكون مجهوًال عند تقییم خ هذه النتیجة تثبیت أن تحدید نوع و مصدر الركام . من أقل قیمة مسجلة لمقاومة االنضغاط% ١٦كانت أفضل قیمها تساوي .خرسانةالمستخدم في الخلطة الخرسانیة عامل مهم في عملیة تحمین مقاومة انضغاط ال Introduction It is often necessary to test concrete structures after the concrete has hardened to determine whether the structure is suitable for its designed use. Ideally such testing should be done without damaging the concrete. The tests available for testing hardened concrete range between the completely nondestructive, where there is no damage to the concrete, through those were the concrete surface is slightly damaged, to partially destructive tests, where the concrete surface had to be repaired after testing. The range of properties that can be assessed using nondestructive tests is quite large and includes such fundamental parameter as density, elastic modulus, compressive strength, surface hardness and absorption as well as reinforcement size and location. Concrete compressive strength is one concrete property that is widely needed to be evaluated during the progress of concrete structures execution. Among the most used nondestructive test methods in assessing concrete compressive strength are the Ultrasonic pulse velocity and the Schmidt hammer rebound number. These two methods are known for more than 50 years [(Carino ,1994) (Bungey and Millard ,1996)]. The first method is used to measure the sound velocity in concrete and concrete compressive strength, while the second method evaluates concrete compressive strength through measuring its surface hardness. These two methods have been known for more than 50 years and gained wide spread use worldwide for their low cost and simple and fast test procedures. Numerous reports and researches have been published on these two methods aiming to obtain mathematical formulations to be used to determine the concrete compressive strength. From previous literature it can be recognized that there is no unique mathematical relationship that can be used worldwide for this purpose. This is because the readings and results of these two methods are largely affected by many factors; among these factors are the elastic properties of aggregate (aggregate source) and their proportion in the concrete, concrete density and moisture content. Thus Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٤٦ large number of mathematical relationships was obtained aiming to give good assessment to concrete compressive strength [(IAEA ,2002)( ACI Committtee 228.1R-95)]. In this research work, it was aimed to obtain mathematical relationships that can be used for the assessment of compressive strength of concrete cast using fine and coarse aggregate obtained from different sources in the southern parts of Iraq. Six different fine aggregate and two coarse aggregate sources were investigated. In addition, two nominal concrete mix proportions with different water / cement ratios were also included as a variable in this research. These two nominal mixes are commonly used in the southern part of Iraq. Experimental Work Two well known nondestructive test methods were used in this work, these methods are the ultrasonic pulse velocity and the Schmidt hammer. The readings and the accuracy of these two methods are very much affected by the elastic properties and proportions of aggregate in the concrete mix. In order to obtain a clear image on the effects of these factors on the accuracy of predicting concrete compressive strength using combined nondestructive test methods, several types of coarse and fine aggregates, from different sources in southern parts of Iraq were used. Testing Program The testing program was planned to obtain information about the effect of mix proportions (aggregate to cement and water to cement ratios) and type of coarse and fine aggregates and concrete density on the results of the UPV and RN methods and on their accuracy in predicting concrete compressive. Two nominal concrete mixes 1:2:4 and 1:1.5:3 mixes with W/C ratio in the range of 0.40 to 0.60 were investigated. These mixes were chosen to represent those widely used in Iraqi construction projects. Also two natural types of coarse aggregate and six types of fine aggregate obtained from different sources in southern part of Iraq was used in the preparation of the concrete mixes. The details of the materials used are given below: Cement (C) Two types of cement, Ordinary and Sulphate Resisting Portland cements conforming to Iraqi standard IQS 5 [6] was used in this work. Fine Aggregate (FA) Six fine aggregate types conforming to Iraqi standard IQS 45 [7] were used. The sources of these aggregate and notation are given in the Table 1. Coarse Aggregate (CA) Two coarse aggregate types conforming to Iraqi standard IQS 45[7] were used. Table 1 also shows the source and notation of these aggregate. Concrete Mixes As detailed in Table 2, hundred and twenty different concrete mixes with aggregate from different sources were investigated; the general characteristics of the mixes are given below: 1. Mix proportions 1:2:4 or 1:1.5:3 2. Nominal water/cement ratio (W/C) from 0.4 to 0.6 (the effective W/C ratio depended finally on the natural moisture condition of the aggregate and ranged between 0.444 to 0.628) Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٤٧ Casting and Curing of Test Specimens 152x152x152 mm concrete cubes were prepared and cast to measure concrete compressive strength; six cubes of each mix were cast in steel moulds then covered for 24 hours by polyethylene sheets for 24 hour. The cubes were then stored in curing tanks for a total period of 28 days. After the 28 days of curing the cube specimens were removed out of water and tested immediately. Testing of Concrete Cube Specimens Concrete Density The densities of the concrete cubes were measured according to ASTM C 138-02 [9]. Each value of density of each mix represents the average of densities of six cubes. Values of concrete densities are given in Table 3. Ultrasonic Pulse Velocity The ultra sonic pulse velocities of the cast concrete cubes were measured according to ASTM C 597-02 [10]. Two readings on each cube were measured (using the opposite smooth surfaces of the cube). Thus each mix result of ultrasonic pulse velocity represents an average of twelve readings. Table 3 gives the ultrasonic pulse velocities of all the mixes investigated. Rebound Number The rebound number was measured on the cube specimens using Schmidt hammer and according to ASTM C 805-02 [11]. Each cube was fixed in the compression machine, and a pressure of 7 MPa was applied on the cube. Five readings were taken on each two opposite smooth surfaces of the cube, thus a total of 10 readings were taken on each cube. The final reading of rebound number of each mix was therefore the average of 60 readings. Table 3 shows the values of the rebound numbers of all the concrete mixes investigated. Concrete Compressive Strength The compressive strengths of the concrete mixes were determined using a compression machine with ultimate capacity of 3000 kN and according to IQS 248[8]. The compressive strength of each mix was the average of the compressive strength of six cubes. The results of the compressive strengths of all the 120 concrete mixes are given in Table 3. Experimental Results Table 3 shows the experimentally measured properties of all the 20 concrete mixes investigated with their ranges of actual W/C and aggregate to cement ratios. These results were fed into the Statistica program in different combinations to find the constants of the multiple regressions (a0, a1, a2, a3, a4, a5). Multiple Non-linear Regressions for Prediction of Concrete Compressive Strength In practice, it is advantageous to use more than one method of non destructive testing (NDT) at a time in predicting or monitoring concrete strength and quality. Using more than one method is beneficial especially because the variations in properties and composition of concrete (aggregate type and source) largely affect the test results of the NDT. Both the Schmidt hammer and UPV are affected by the mix proportions of the concrete, aggregate elastic properties and also by its moisture condition each in a certain manner [(ASTM C 805-02) (Kaplan,1959)]. These factors may result in an increase or decrease in the value of the estimated concrete strength (error). Such as the case of the presence of moisture in concrete: presence of moisture in concrete increases the UPV, but on the Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٤٨ other hand, it decreases the rebound number recorded by the Schmidt hammer (Neville,2005), so when both methods are used together, the error in one method will correct the error in the second method. The presence of moisture in concrete will increase the UPV reading but at the same time will decrease the rebound number, so readings of the UPV and RN will correct each other and the effect of moisture will be eliminated in the estimation of concrete strength. Another factor that largely affects the NDT results are the concrete mix proportions, for the same compressive strength, mixes with higher coarse aggregate will result in an increase in the UPV and rebound number(Bungey and Millard,1996). There have been numerous attempts from different researchers throughout the world to find mathematical relationships that can predict the concrete compressive strength by using the Ultra Sonic Pulse and the Schmidt Rebound Hammer either separately or combined [(Facacoaru ,1984) to (Tanigawa and etl,1984)]. All these methods used local materials and cannot be applied for concrete cast using aggregate from other different sources. The predicted concrete strength values using the previously developed relationships will show large scatter compared to the experimentally measured values when aggregate from other sources are used in casting concrete. In order to obtain accurate relationships to predict concrete strength, multiple linear regressions were used. Different forms of relationships with different combinations of independent variables can be obtained to predict the concrete compressive strength. These independent variables are: Ultrasonic pulse velocity, Schmidt rebound number, water/cement ratio, aggregate / cement ratio and concrete density, Depending on the availability of these information on the concrete mix characteristics and aggregate origin (source). The general form of the regression is given below: fcu = ao . (UPV)a1 . (RN)a2 . (A/C)a3 . (W/C)a4 . (ρ)a5 Where: fcu : Concrete compressive strength in MPa. UPV : Ultrasonic pulse velocity in km/sec. RN : Rebound number. W/C : Water to cement ratio by weight. A/C : Aggregate to cement ratio by weight. ρ : Concrete density in (kg/m3). ao, a1 to a5: regression constants. The mathematical regressions were divided into two groups. The first group considered each type of aggregate from a particular source in Southern part of Iraq individually, while the second group combined all the data of concrete specimens, regardless to the aggregate source, in an attempt to get more practical and easier to use regressions, and also, because in some cases, the source of aggregate may not be known. This of course will affect the accuracy of the regressions adversely, but the engineer must take this into consideration in his assessment to concrete strength. All regressions considered either Ordinary Portland cement or Sulphate Resisting Portland cement. Group 1: Regressions for Aggregates from Known (Particular) Sources Group one was divided into 20 subgroups, ten groups for each of the Ordinary or Sulphate Resisting Portland cements concrete. In each subgroup, three regressions were derived. Each regression included different either the ultrasonic pulse velocity or rebound number separately or combined depending on the availability nondestructive method used in testing the concrete. In each table, the values of regression constants (ao, a1, a2,) are given. These tables also give the standard errors of estimates in the compressive strength (SE) and the multiple variable correlation coefficients (R) of each regression to show its accuracy. Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٤٩ Table 4 gives the regressions constants for each type of aggregate source for concrete mixes cast with Ordinary Portland cement, while Table 5 gives the constants for concrete mixes cast with Sulphate Resisting Portland cement. Each group included both 1:2:4 and 1:1.5:3 mixes. From Tables 4 and 5, it can be clearly seen that the regressions gave excellent prediction, using UPV test only, the value of standard error was between 0.32 to 3.41 MPa, while when using the Schmidt hammer the this error was between 0.94 to 2.59 MPa. Combining both nondestructive test methods, the maximum standard error value is decreased to 0.27 to 2.03 MPa. The latter value is less than 10% of the lowest concrete compressive strength investigated (20MPa). Introducing the mix proportions in the regression (aggregate/cement, water/cement and density) improved the regression, but to a limited extent. Therefore, using the combined UPV and RN regression (with particular reference to aggregate source) was found to be sufficient and practical to predict the concrete compressive strength. Group 2: Regressions for Aggregate from Unknown Source In order to extend the validity of the regressions derived for particular sources of aggregates (Tables 4 and 5), other combinations of regressions were derived, in these regressions; the source of aggregate was overlooked. Table 6 and 7 give regressions constants for predicting concrete strength cast with Ordinary or Sulphate Resisting Portland cements respectively, regardless to the aggregate source. It is important to highlight here that these regressions must only be used for concrete cast with aggregate from the southern parts of Iraq. In these two groups of regressions (Ordinary or Sulphate Resisting Portland cements), the maximum value of standard error was 4.05 MPa when using UPV method only, and 4.44 MPa when using Schmidt hammer only, the value of standard error decreases to 3.52 MPa when using the combine UPV and Schmidt hammer test methods. When the mix properties are included, the standard error value was further decreased to 3.36 MPa. The four values of standard errors for these regressions give errors of about 20%, 22%, 19% and 17% respectively. Figure 1 and Figure 2 shows the observed versus predicted compressive strength for the first regression of the two groups (Regressions N1 and S1). It is important to mention here that the age of concrete was not included in the regressions, because it is more preferable to depend only on the result of the UPV and rebound number in addition to the mix proportions if available to represent the condition of concrete hardened properties, in many practical site cases the age of concrete may not be known exactly. Limitations of The Developed Regressions In order to obtain a realistic predicted value for the concrete compressive strength, the general ranges of the independent variables introduced in the derivation of these regressions must be taken into consideration. These final ranges are given in Table 8. Conclusions On the basis of the experimental results obtained in this work, using two different nondestructive test methods for predicting concrete compressive strength, cast with aggregate obtained from different sources in the southern part of Iraq, following conclusions can be withdrawn: 1. Changing the source of aggregate affects the results of the ultra sonic pulse velocity and the rebound number of the Schmidt hammer. There is no generalized formula that can be used Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٥٠ for predicting concrete compressive strength using nondestructive testing. The stiffness of aggregate largely affects the readings of the UPV and RN methods. 2. The combined usage of UPV and RN methods improves the predicted values of concrete compressive strength, several factors that causes variations in the readings of these methods eliminate each other, thus resulting in more accurate predicted values of concrete strength. Further introducing information on concrete mix proportions and density into the mathematical regressions can improve the accuracy of the predicted value. 3. Several regressions were derived for each type of aggregate source used in the concrete mix. These regressions gave excellent accuracies especially when both UPV and RN methods were used together. In most cases, the standard error of the regression was less than 10 % compared with the minimum concrete strength investigated (20 MPa). It was found that there is no need to introduce the mix proportions in this case, since the regressions gave good and acceptable accuracy. 4. The accuracy of the regressions decreased when all the data from the different aggregate sources were used, due to the variation in the elastic properties of the concrete. The standard error maximum values exceeded 20% when using UPV or RN methods separately. This error decreased to less than 19% when using the combined tests together. A further decrease in the standard error was obtained when the mix proportions and concrete density was introduced into the regressions, the maximum value of the standard error became less than 17%. 5. In using the derived regression, the engineer must be aware not to tolerate the limits of the independent variables used in the regression. This may result in nonrealistic predicted values. Aknowledgement The authors would like to thank the Iraqi Ministry of Higher Education and Scientific Researches for providing the fund to carry out the work reported in this paper. Thanks are also to the staff of the structural and material laboratory of Shatra Technical Institute for providing the technical support for this work. References Carino, N.J., “Nondestructive Testing of concrete: History and Challenges”, SP 144-30 (American Concrete Institute, Detroit 1994) pp.623-678. Kolek, J., “Nondestructive testing of concrete by hardness methods”, Symposium on Nondestructive testing of concrete and timber, (The Institution of Civil Engineers, (London, 1970) pp.19-22. Bungey, J.H. and Millard, S.G., “Testing of Concrete in Structures”, Third Edition, United Kingdom, Glasgow, 1996, 350 pp. IAEA: International Atomic Energy Agency, “Guidebook on nondestructive testing of concrete structures”, Vienna, 2002, pp.62-66. ACI Committtee 228.1R-95, “In-Place Methods to Estimate Concrete Strength”, (American Concrete Institute, Detroite, 1995). IQS 5/1984, ''Portland Cement,'' Iraqi Organization for Standards and Specifications. Baghdad, Iraq, 1984. Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٥١ IQS 45/1984 "Aggregate From Natural Sources for Concrete," Iraqi Organization for Standards and Specifications. Baghdad, Iraq, 1984. IQS 248 "Method for Determination of Compressive Strength of Concrete Cubes," Iraqi Organization for Standards and Specifications. Baghdad, Iraq, 1992. ASTM C 138-02, "Test for Unit Weight, Yield, and Air Content (Gravimetric) of Concrete", ASTM International. American Society of Testing Materials, USA. ASTM C 597-02, “Standard Test Method for Pulse Velocity through Concrete”, ASTM International. American Society of Testing Materials, USA. ASTM C 805-02, “Standard Test Method for Rebound Number of Hardened Concrete”, ASTM International, American Society of Testing Materials, USA. Kaplan, M.F., “The effect of age and water/cement ratio upon the relationship between ultrasonic pulse velocity and compressive strength”, Mag. of Concrete Researches, Vol. 11, No. 32, July 1959, pp. 85-91. Bullock, R.E. and Whitehurst, E.A., “Effect of certain variables on pulse velocities through concrete”, Highway Researches Board, Bull., 206, 37, 1959. Neville, A.M., “Properties of Concrete”, fourth and final edition, United Kingdom, 2005. Facacoaru, I. , “Romanian Achievements in Nondestructive strength testing of concrete”, In-situ nondestructive testing of concrete, SP-82, (American Concrete Institute, Detroite, 1984), pp.33-56. Knaze, P. and Beno, P., “The use of combined nondestructive testing methods to determine the compressive strength of concrete”, Materials and structures, Vol. 17-No. 99, May-June 1984, (RILEM, Paris, 1984), pp.207-210. Kehder, G.F. , “A two stages procedure for assessment of in-situ concrete compressive strength using combind nondestructive testing”, Materials and structures, Vol. 32, July 1999, (RILEM, Paris, 1984), pp.410-417. Sturrup, V.R., Vecchio, F.J. and Caratin, H., “Pulse velocity as a measure of concrete compressive strength”, In-situ nondestructive testing of concrete, SP-82, (American Concrete Institute, Detroite, 1984), pp.201-227. Swamy, N.R. and Al-Hamed, A.H., “The use of pulse velocity measurements to estimate strength of air-dried cubes and hence in-situ strength of concrete”, Malhotra, V.M., Ed., ACI SP-82, (American Concrete Institute, Farmington Hills, MI, 1984), 247 pp. Teodoru, G.V., “Mechanical Strength Property of Concrete at Early Ages as Reflected by Schmidt Rebound Number, Ultrasonic Pulse Velocity and Ultrasonic Attenuation”, Property of Concrete at Early Ages, ACI SP-95, (American Concrete Institue, Detroit, 1986), pp.139-153. Leshchinsky, A.M., “Combined Methods of Determining Control Measures of Concrete Quality”, Materials and structures, Vol. 24, 1991, pp.177-184. Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٥٢ Tanigawa, Y., Baba, K. and Mori, H., “Estimation of Concrete Strength by Combined Nondestructive Testing Methods”, In-situ nondestructive testing of concrete, SP-82, (American Concrete Institute, Detroite, 1984), pp.57-75. Table 1: Materials sources and notations* item Cement type Fine aggregate source Coarse aggregate source 1 Ordinary N Najaf1(Wilayat Ali) A Badrah B 2 Sulphate resisting S Badrah B Basrah S 3 Najaf2(Khamas) K 4 Zubair Z 5 Jabal Sanam S 6 Al Ukhaider G NBS: Concrete mix with Ordinary Portland cement, FA from Badrah and CA from Basrah. Table 2: mix proportion details and notations Mix notation Mix proportion Effective W/C ratio Mix notation Mix proportion Effective W/C ratio Mix notation Mix proportion Effective W/C ratio Mix notation Mix proportion Effective W/C ratio NBB1 1:2:4 0.557 SKB1 1:2:4 0.628 NAS1 1:2:4 0.600 SZS1 1:2:4 0.600 NBB2 1:2:4 0.507 SKB2 1:2:4 0.578 NAS2 1:2:4 0.550 SZS2 1:2:4 0.550 NBB3 1:2:4 0.457 SKB3 1:2:4 0.528 NAS3 1:2:4 0.500 SZS3 1:2:4 0.500 NBB4 1:1.5:3 0.544 SKB4 1:1.5:3 0.600 NAS4 1:1.5:3 0.578 SZS4 1:1.5:3 0.578 NBB5 1:1.5:3 0.494 SKB5 1:1.5:3 0.550 NAS5 1:1.5:3 0.494 SZS5 1:1.5:3 0.528 NBB6 1:1.5:3 0.444 SKB6 1:1.5:3 0.500 NAS6 1:1.5:3 0.444 SZS6 1:1.5:3 0.478 SBB1 1:2:4 0.600 NZB1 1:2:4 0.557 SAS1 1:2:4 0.557 NSS1 1:2:4 0.600 SBB2 1:2:4 0.550 NZB2 1:2:4 0.507 SAS2 1:2:4 0.507 NSS2 1:2:4 0.550 SBB3 1:2:4 0.500 NZB3 1:2:4 0.457 SAS3 1:2:4 0.457 NSS3 1:2:4 0.500 SBB4 1:1.5:3 0.578 NZB4 1:1.5:3 0.544 SAS4 1:1.5:3 0.544 NSS4 1:1.5:3 0.578 SBB5 1:1.5:3 0.528 NZB5 1:1.5:3 0.494 SAS5 1:1.5:3 0.494 NSS5 1:1.5:3 0.528 SBB6 1:1.5:3 0.478 NZB6 1:1.5:3 0.444 SAS6 1:1.5:3 0.444 NSS6 1:1.5:3 0.473 NAB1 1:2:4 0.500 SZB1 1:2:4 0.600 NKS1 1:2:4 0.600 SSS1 1:2:4 0.557 NAB2 1:2:4 0.450 SZB2 1:2:4 0.550 NKS2 1:2:4 0.550 SSS2 1:2:4 0.550 NAB3 1:2:4 0.400 SZB3 1:2:4 0.500 NKS3 1:2:4 0.500 SSS3 1:2:4 0.457 NAB4 1:1.5:3 0.550 SZB4 1:1.5:3 0.600 NKS4 1:1.5:3 0.578 SSS4 1:1.5:3 0.544 NAB5 1:1.5:3 0.500 SZB5 1:1.5:3 0.550 NKS5 1:1.5:3 0.528 SSS5 1:1.5:3 0.494 NAB6 1:1.5:3 0.450 SZB6 1:1.5:3 0.500 NKS6 1:1.5:3 0.478 SSS6 1:1.5:3 0.444 SAB1 1:2:4 0.600 NSB1 1:2:4 0.557 SKS1 1:2:4 0.557 NGB1 1:2:4 0.600 SAB2 1:2:4 0.550 NSB2 1:2:4 0.550 SKS2 1:2:4 0.507 NGB2 1:2:4 0.550 SAB3 1:2:4 0.500 NSB3 1:2:4 0.500 SKS3 1:2:4 0.457 NGB3 1:2:4 0.500 SAB4 1:1.5:3 0.578 NSB4 1:1.5:3 0.578 SKS4 1:1.5:3 0.544 NGB4 1:1.5:3 0.578 SAB5 1:1.5:3 0.528 NSB5 1:1.5:3 0.528 SKS5 1:1.5:3 0.494 NGB5 1:1.5:3 0.528 SAB6 1:1.5:3 0.478 NSB6 1:1.5:3 0.478 SKS6 1:1.5:3 0.444 NGB6 1:1.5:3 0.478 NKB1 1:2:4 0.628 SSB1 1:2:4 0.600 NZS1 1:2:4 0.600 SGB1 1:2:4 0.600 NKB2 1:2:4 0.578 SSB2 1:2:4 0.550 NZS2 1:2:4 0.550 SGB2 1:2:4 0.550 NKB3 1:2:4 0.528 SSB3 1:2:4 0.500 NZS3 1:2:4 0.500 SGB3 1:2:4 0.528 NKB4 1:1.5:3 0.600 SSB4 1:1.5:3 0.589 NZS4 1:1.5:3 0.578 SGB4 1:1.5:3 0.578 NKB5 1:1.5:3 0.550 SSB5 1:1.5:3 0.539 NZS5 1:1.5:3 0.528 SGB5 1:1.5:3 0.528 NKB6 1:1.5:3 0.500 SSB6 1:1.5:3 0.489 NZS6 1:1.5:3 0.478 SGB6 1:1.5:3 0.478 Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٥٣ Table 3: Range of measured properties of the tested concrete mixes Mix notation Compressive strength MPa UPV m/sec RN W/C Ratio Agg/cement ratio Density kg/m3 NBB 25.02-45.2 4573-4870 21.2-27.3 0.444-0.557 6 or 4.5 2296-2381 SBB 19.64-41.42 4310-4813 23.7-30.7 0.478-0.60 6 or 4.5 2309-2379 NAB 25.91-34.22 4412-4839 23.4-31.3 0.45-0.55 6 or 4.5 2353-2415 SAB 20.27-31.84 4412-4639 25.3-27.4 0.478-0.60 6 or 4.5 2295-2399 NKB 18.82-33.41 4167-4518 18.8-33.4 0.50-0.628 6 or 4.5 2296-2379 SKB 24.12-33.19 4265-4478 24.1-33.2 0.50-0.628 6 or 4.5 2305-2360 NZB 22.40-31.45 4412-4545 22.4-31.5 0.444-0.557 6 or 4.5 2343-2418 SZB 28.02-43.08 4335-4663 28.0-43.2 0.50-0.60 6 or 4.5 2315-2389 NSB 23.24-35.48 4369-4545 23.2-35.5 0.478-0.557 6 or 4.5 2335-2383 SSB 26.77-35.94 4455-4615 26.8-35.9 0.489-0.60 6 or 4.5 2337-2396 NAS 20.93-42.12 4186-4687 20.4-42.1 0.444-0.60 6 or 4.5 2357-2441 SAS 29.02-42.28 4434-4687 29.0-42.3 0.444-0.557 6 or 4.5 2349-2451 NKS 27.95-32.94 4360-4545 29.3-29.5 0.478-0.60 6 or 4.5 2359-2387 SKS 34.60-50.47 4390-4580 28.2-31.1 0.478-0.60 6 or 4.5 2378-2418 NZS 23.05-32.94 4478-4712 27.1-31.7 0.473-0.60 6 or 4.5 2358-2471 SZS 22.50-33.05 4310-4580 27.9-30.4 0.444-0.557 6 or 4.5 2403-2481 NSS 18.07-30.98 4592-4813 27.8-32.7 0.473-0.60 6 or 4.5 2388-2464 SSS 27.18-35.60 4523-4737 27.2-32.6 0.444-0.557 6 or 4.5 2386-2474 NGB 22.83-27.76 4390-4545 28.2-30.4 0.478-0.60 6 or 4.5 2385-2422 SGB 21.91-26.03 4348-4412 27.0-30.6 0.478-0.60 6 or 4.5 2374-2460 Table 4: Regressions constants of groups cast with Ordinary Portland cement Reg. No. a0 a1 a2 S.E MPa R Reg. No. a0 a1 a2 S.E MPa R NAB NKS NAB1 0.445 1.676 0.498 1.39 0.871 NKS1 3.3x10-5 2.797 2.826 0.47 0.990 NAB2 0.272 3.053 1.85 0.759 NKS2 7.8x10-5 8.511 1.16 0.941 NAB3 2.734 0.735 1.69 0.803 NKS3 5.9x10-5 3.908 ٠.66 0.981 NBB NSB NBB1 0.1755 -0.553 1.888 1.04 0.991 NSB1 7.8x10-5 9.136 -0.34 1.11 0.928 NBB2 1.92x10-6 10.722 3.55 0.897 NSB2 1.39x10-4 8.028 1.12 0.927 NBB3 0.096 1.809 1.05 0.991 NSB3 0.028 2.103 1.55 0.854 NAS NSS NAS1 1.87x10-4 4.49 1.545 1.44 0.978 NSS1 0.0671 1.055 1.278 0.88 0.952 NAS2 6.64x10-5 8.614 1.88 0.963 NSS2 0.0330 4.357 1.06 0.931 NAS3 0.0011 3.044 1.91 0.962 NSS3 0.0924 1.659 0.90 0.950 NGB NZB NGB1 0.0012 3.264 1.492 0.43 0.957 NZS1 9.55x10-4 5.147 0.786 0.97 0.963 NGB2 0.010 5.216 1.10 0.823 NZS2 2.48x10-4 7.762 1.48 0.912 NGB3 0.0189 2.138 0.94 0.873 NZS3 0.1595 1.577 1.91 0.847 NKB NZS NKB1 1.29x10-3 7.062 -0.211 0.28 0.998 NZB1 0.01849 7.287 - 1.126 2.03 0.715 NKB2 1.19x10-3 6.656 0.32 0.997 NZB2 0.08616 3.790 2.19 0.655 NKB3 2.88x10-3 2.795 2.55 0.856 NZB3 1.9157 0.794 2.56 0.468 Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٥٤ Table 5: Regressions constants of groups cast with Sulphate Resisting Portland cement Reg. No. a0 a1 a2 S.E MPa R Reg. No. a0 a1 a2 S.E MPa R SAB SKS SAB1 0.0008 4.346 1.176 1.01 0.965 SKS1 1.55x10-4 4.727 1.557 0.84 0.984 SAB2 0.0022 6.258 1.21 0.949 SKS2 1.94x10-7 12.67 1.82 0.922 SAB3 0.001 3.084 1.62 0.908 SKS3 0.0164 2.283 1.18 0.968 SBB SSB SBB1 1.42x10-4 5.774 1.037 0.98 0.991 SSB1 0.0677 1.008 1.396 1.25 0.935 SBB2 1.806x10-5 9.411 1.47 0.981 SSB2 6.44x10-4 7.145 1.70 0.875 SBB3 7.973x10-3 2.459 1.90 0.968 SSB3 0.164 1.588 1.27 0.933 SAS SSS SAS1 6.11x10-3 4.8 0.405 0.67 0.993 SSS1 0.0128 3.961 0.505 0.63 0.98 SAS2 1.67x10-3 6.554 0.97 0.985 SSS2 6.32x10-3 5.539 0.76 0.971 SAS3 0.4256 1.309 1.83 0.945 SSS3 0.173 1.528 1.13 0.933 SGB SZS SGB1 2.19x10-4 5.013 1.272 0.37 0.99 SZS1 3.30x10-5 2.192 3.060 0.81 0.974 SGB2 1.63x10-5 9.644 1.03 0.926 SZS2 3.16x10-3 6.042 1.39 0.922 SGB3 0.02074 2.130 0.91 0.943 SZS3 8.87x10-6 4.422 0.99 0.961 SKB SZB SKB1 0.0029 5.372 0.333 0.27 0.998 SZB1 9.37x10-4 3.037 1.775 0.78 0.987 SKB2 1.75x10-3 6.440 0.51 0.993 SZB2 4.08x10-4 7.496 2.12 0.905 SKB3 0.152 1.580 2.02 0.885 SZB3 6.42x10-3 2.578 1.34 0.963 Table 6: Regressions for concrete with Ordinary Portland cement and all aggregate types Ordinary Portland Cement ( all groups: 60 points : 360 cubes ) All aggregate types RS.EMPaa5a4a3a2a1a0 Regression No. 0.8332.73-3.3240.085-0.5430.7553.069504.882N1 0.8122.89-0.053-0.5520.2673.0550.0857N2 0.8012.96-2.90.7124.31137.621N3 0.7783.100.3374.340.0126N4 0.7613.214.7150.0219N5 0.4404.440.9011.432N6 Table 7: Regressions for concrete with Sulphate Resisting Portland cement and all aggregate types Sulphate Resisting Portland Cement ( all groups: 60 points : 360 cubes ) All aggregate types RS.EMPaa5a4a3a2a1a0 Regression No. 0.8483.20-3.0580.127-0.5241.4742.25669.644S1 0.8183.47-0.032-0.3900.9062.9330.0145S2 0.8313.36-2.0821.5942.9541.266S3 0.8123.521.2203.3820.00315S4 0.7414.056.2490.00247S5 0.7583.941.9690.0421S6 Al-Qadisiya Journal For Engineering Sciences Vol. 3 No. 3 Year 2010 ٢٥٥ Table 8: General ranges of measured properties of the tested concrete mixes UPV m/sec RN W/C Ratio Agg/cement ratio Density kg/m3 4167-4870 18.8-43.2 0.444-0.628 4.5-6.0 2295-2481 20 24 28 32 36 40 44 Predicted Cube Strength MPa 16 20 24 28 32 36 40 44 48 O bs er ve d C ub e St re ng th M Pa Figure 1: Observed versus predicted compressive strength values of regression N1. 20 24 28 32 36 40 44 48 Predicted Cube Strength MPa 16 20 24 28 32 36 40 44 48 O bs er ve d C ub e St re ng th M Pa Figure 2: Observed versus predicted compressive strength values of regression S1.