Betonda deney hızı-dayanım ilişkileri ve etken faktörleri

Abstract

ÖZET Bu çalışmada, sertleşmiş çimento hamuru, harç ve beton numunelerinin deneylerinde uygulanan yükleme hızının daya nıma etkisi araştırılmış, özellikle YÜK HIZI-DAYANIM iliş kisine etki eden bazı faktörler üzerinde durulmuştur. Bilindiği gibi, hız etkisi uzun zamandan beri araştırı lan bir konudur. Araştırıcılar daha çok deney hızının.yani; yük hızı veya deformasyon hızının dayanıma etkisini içeren çalışmalar yapmışlardır. Ancak, bazı araştırıcılar (9-12) de ney hızının dayanımı etkilemediğini, buna karşın bazı araş tırıcılar da (1-8) deney hızının dayanımı etkilediğini ve hız yükseldikçe daha büyük dayanım elde edildiğini belirt mektedirler. Deney hızının etkisiyle oluşan dayanım değişi minin hangi faktörlerin etkisiyle oluştuğu ve hızın neden etkili bir faktör olduğu henüz tam açıklığa kavuşmuş de ğildir. Bu çalışmada önce dayanım yük hızı ilişkisine serbest su miktarı, numune yaşı ve saklama koşulunun etkileri ince lenmiş, genel olarak deney süresi 1 saniye ile 4 saat ara sında değişen yük hızları uygulanmış tır. Buna ilaveten yük hızının gerilme seviyesine göre oluşturduğu dayanım deği şimini saptamak amacıyla değişik tipte ön yüklemeli deney ler de yapılmıştır. Yük hızı - dayanım ilişkisine serbest su miktarının etkisini belirlemek amacıyla, çimento hamuru, harç ve beton bileşiminde numuneler hazırlanmış, deneyden birkaç gün önce numuneler 105 'C sıcaklıkta farklı- bir süre bekletilerek su miktarı farklı olan gruplar elde edilmiştir. Bunu taki ben herbir grubun aynı şekilde deneyi yapılmıştır. Suya doy gun olarak deneyi yapılan numunelerin dayanımları yaklaşık %35 mertebesinde, yük hızı etkisiyle değişim gösterirken, su miktarı azaldıkça değişim miktarının azaldığı ve kuru halde deneyi yapılan numunelerin dayanımlarında oluşan de ğişimin önemsenmiyecek kadar az olduğu saptanmıştır. Bu so nuçlar; serbest su miktarının yük hızı-dayanım ilişkisine etki eden önemli bir faktör olduğunu yansıtmaktadır. özel likle iç yapısında serbest su molekülleri bulunan numune lerin, kuru hale oranla, yüksek hızlarda daha büyük dayanımVI ermesi fiziksel olarak yorumlandzğmdâf su moleküllerinin atı faza yardımcı rol oynadığı ve bunun bir sonucu olarak u ihtiva eden numunelerden daha büyük dayanım elde edildi- i kanaatine varılmıştır. Dayanım -yük hızı ilişkisine numune yaşının etkisini geren deneyler de yapılmıştır. Su ihtiva eden ve farklı faşlarda deneyi yapılan numunelerin dayanımları yük hızm- îan farklı miktarda etkilenmişlerdir. Dayanım-yük hızı ilişkisine numune saklama koşulunun etkisini belirlemek amacıyla yapılan deneylerden elde edi len sonuçlara göre ise; saklama koşulu farklı olan numune lerin hem DAYANIM - log(HIZ) bağıntısının biçimi hem de yük hızının dayanıma etki miktarı farklı olmuştur. Bu çalışmada ayrıca farklı gerilme seviyelerinde yük hızının ani değiştirilmesiyle oluşan dayanım değişimini belirlemek amacıyla, yeniden çimento hamuru numuneleri farklı tip yükleme altında deneye tabi tutulmuştur. Aynı seriye ait numuneler 9 gruba ayrılarak, birinci grup, deney süresi 1 saniye olan Hl sabit yük hızı, ikinci grup, deney süresi 10 dakika olan H2 sabit yük hızı altında deneye tabi tutulmuştur. Diğer gruplar önce H2 hızıyla bel li bir gerilme seviyesine kadar yüklenmiş sonra Hl hızına geçilmiş ve böylece herbir grup için bu hız değişimi fark lı gerilme seviyelerinde yapılmıştır. Elde edilen deney so nuçlarına göre; maksimum gerilmenin %30 seviyesine kadar olan gerilme seviyeleri bölgesinde H2 hızı ve sonra Hl hı zı uygulandığında daha büyük dayanım elde edilmiştir. Buna karşın sözkonusu hız değişimi %30 seviyeden daha büyük ge rilme seviyelerinde yapıldığında, dayanımın azaldığı, özel likle maksimum gerilmenin %85 seviyesinden daha büyük ge rilme seviyelerinde hız değişimi yapıldığında dayanımdaki azalmanın çok arttığı gözlenmiştir. Diğer bir yükleme tipi olarak, yalnız küçük gerilme bölgesinde hız değişimi yapılmıştır. Yani, bir grup numune deney süresi 1 saniye olan Hl yük hızı altında deneye tabi tutulmuş ve tfı gibi bir dayanım elde edilmiş, diğer grup lara 0.20 tfı seviyesine kadar Hl den küçük H2,H3,H4 gibi sabit yük hızları uygulanmıştır. Her bir grup için 0.20 Wıvu gerilme değerlerinden büyük gerilmelerde Hl sabit yük hızı uygulanmıştır. Bu yükleme tipinden elde edilen sonuçlara gö re; 0.20 Wı seviyesine kadar daha küçük yük hızları uygu landığında daha büyük dayanım elde edilmektedir. Diğer bir yükleme tipi olarak, büyük gerilme seviyesin de hız değişiminin dayanıma etkisini belirleyecek tipte yükleme seçilmiştir. Aynı kalitede numuneler deney sırasın da iki gruba ayrılmış ve gruplardan birisi sıfırdan itiba ren Hl,H2,H3,H4 gibi sabit yük hızları altında deneye tabi tutulmuş, diğer gruplarsa,önce maksimum gerilmenin %50 sevi yesine kadar H3 hızıyla yüklenmiş sonra 0.50 Wı seviyesinden itibaren Hl,H2,H3,H4 gibi sabit yük hızları uygulanmıştır. Elde edilen sonuçlara göre$0.50 Wı seviyesinden itibaren H4 hızı uygulamak veya sıfırdan itibaren H4 hızı uygulamak, sonuçta yaklaşık dayanım değerleri vermektedir. Buradan, yük hızı etkisinin maksimum gerilme seviyesine yakın geril melerde oluştuğu kanaatine varılmaktadır. Bu farklı tipte uygulanan yükleme deneyleriyle elde edilen sonuçlar fizik sel olarak yorumlandığında yük hızı etkisinin numune iç ya pısındaki çatlakların gelişmesiyle arttığı sonucuna varıl maktadır.

U M M A R In this work, the effect of loading rate upon the strength of hardened cement paste, mortar and concrete was studied. A great deal of research has been carried out in the past sixty years, to find out the effect of testing speed (i.e. strain rate or loading rate) on the mechanical properties of concrete. Most of the attention has been paid to the effect of testing speed on strength. According to some investigators (9-12) the effect of testing speed on strength can be neglected* On the other hand some researchers (1-8) indicated that the effect of testing speed is not negligible and strength increases up to 80 percent with increasing testing speed. In this research, first the effect of moisture content, age and curing regime of the specimen upon the strength-loading rate relationship was studied. Also the reduction in strength, due to a change to slower loading rates was investigated at various stress levels and a physical explanation for this behaviour was given.X According to the research plan, more than seven hundred cement paste, mortar and concrete specimens were prepared altogether and tested under uniaxial compression. Dimensions of the cement paste and mortar specimens were 4 x 4 x 16 cm. The concrete specimens were cylinders, 10 cm. in diameter and 20 cm. in hight. Mix ratios (by weight) for various specimens were: Cement paste specimens: Water/Cement ratio - 0.38 Mortar specimens group I: Water/Cement ratio s 0.41 Sand /Cement ratio -2.0 Mortar specimens group II: Water /Cement ration 0.50 Sand /Cement ratio -2.0 Concrete specimens : Water/Cement ratio -0.53 Sand /Cement ratio -2.08 Gravel/Cement ratio-3.15 The cement used was K.P.Ç. 325 (Portland cement) which was obtained from the `Trabzon Cement Factory `: Sand and gravel were local materials obtained from the Black Sea Cost. The tests were carried out at the `KARADENİZ TEKNİK ÜNİVERSİTESİ'* YAPI ve MALZEME LABORATUARI` and a `LOSENHAUSENWERK UNIVERSAL TESTING MACHİNE` with a capacity of 60 tons was used. First, the influence of moisture content upon the strength loading rate relationship was investigated. For this purpose, a series of cement paste specimens were prepared and cured in water for 28 days. Then the specimens were divided into groups and each group owen dried (105 'C) for different duration from I to 14 hours. This meant that, each group of specimens had a different moisture content at the time of testing. At the age of 32 days, each group of specimens were tested under costant rates of loading, the rate varying bepween 500 to 0,035 kgf/cm^ per second, and failure was reached in periods ranging from I second to 4 hours, for different specimens.XT The observed data indicated that strength increased with increasing rate of loading, and the amount of this change in strength depended on the moisture content of the specimen tested. For the saturated specimens, aproximately 35 percent increase in strength values was observed,. when loading time was reduced from 4 hours to I second. Similar observations were made for specimens with lesser moisture contents but the increase in strength was not so pronounced. At lower loading rates all groups had same strength, however under high rates of loading moist specimens had remarkably higher strengths than dry ones. An other groups of cement paste, mortar and concrete specimens were prepared and after curing, they were all owen dried (105 'C) for 48 hours to evaporete all the free water in them. They were tested in the same way as before. The obtained data indicated that this time the effect of loading rate was negligible. Therefore, there was evidence to believe that the amount of free water present in a specimen, determined the amount of the change in strength due to a change in rate of loading. It may be said that the water phase played a contributing part in the strength of the composite meterials. We may assume that, when a low loading rate was employed, most of the the free water flowed from compressed voids to the uncompressed empty regions. On the other hand, when a high rate was employed, water did not find enough time to flow, and most of it remained locked up in its original position and acted as a solid. This action contributed to the overall strength of the specimen. Secondly, the effect of age upon the strength-loading rate relationship was investigated. For that purpose; hardened cement paste specimens were first prepared and tested under constant rates of loading at ages of 2, 4, 8, 15 and 29 days. These test were repeated with different loading rates varying between 0.035 to 500 kgf/cm? per second. All specimens displayed higher strengths when load at 500 kgf/cm? per second, compared to those measured at a loading rate of 0.035 kgf/cm? per second. The difference between strengthXII values at these two loading rates was taken as the measure of the effect of loading rate on strength, and expressed as a percentage of the higher strength values. The magnitudes of these effect values so calculated for the specimens at ages of 2, 4, 8, 15 and 29 days were 40, 42, 35, 28, and 22 percent respectively. Also two series of mortar specimens were prepared and one of them was tested at ages of 2 1/2, 5 and 8 days. The magnitudes of the effect values were 27,39 and 38 percent respectively. The second series was tested at the ages of 2, 12 and 34 days, and the magnitude of the effect values were 14, 33 and 31 percent respectively. Finally, concrete specimens were prepared and tested at ages of 3, 7 and 36 days with minimum and maximum loading rates of 0.007 kgf/cm2 per second and 150 kgf/cm2 per second respectively. Magnitudes of the effect values for these loading rate limits were 29, 34 and 33 percent for the three different ages. These observations indicated thet the effect of loading rate on strength, varied also with the age of the cement paste mortar and concrete specimens. Thirdly, the effect of curing regime upon the strength- loading rate relationship was investigated. A series of three groups hardened cement paste specimens wore prepared. One hroup was cured in water (16 'C) another group was cured in 98% 2 percent relative humidity (18*2 'C), and the third one was cured in laboratory air (75*8 percent relative humidity and 18%2 'C) for 28 days. Then all three groups were satureted for 3 days and tested under constant rates of loading and these tests repeated with loading rates varying from 0.104 kgf/cm2 per second to 500 kgf/cm2 per second. The results obtained in these investigations indicated that curing regime influenced the strength-loading rate relationship.Xlll In further tests the effect of an abrubt change in lading rate was investigated for this purpose a series : nine groups of specimens were prepared. The first group is tested under a high constant rate of loading, HI»500 jf/cm2 per second and failure was accomplished in 1 second, he second group was tested under a low constant rate of oading, H2*0. 52 kgf/cm2 per second and failure was ccomplished in 10 minutes. The other seven groups of pecimens were initially loaded at the lower rate, H2, tntil a stress level between aproximately 10 to 90 percent >f the strength was reached, and then loading was continued it the higher rate of HI until failure. The observations showed that the initial application of the slower rate, H2, up to a stress level which was below 30 percent of the final strength, caused an increase in strength. However, initial application of the slower rate in excess of this limit, caused a decrease in final strength. This increase in the strength due to initial application of a slower rate of loading was very unexpected, and further investigations had to be made to explain this behavior. Therefore, a series of similar groups of the specimens were prepared, and one group of these was tested under the fast speed, Hl»500 kgf/cm2 per second. The other groups were initially loaded under different slower constant rates of loading, up to a stress level, which was 20 percent of the strength at the fast rate, and then the fast rate was applied until failure. The obtained data showed again clearly, that for this stress level, strength increased with slower rates of loading applied initially. In further tests one series of hardened cement paste specimens was divided in to two groups. One group was tested under constant rates of loading (Hi, H2, H3, H4), and their strengths were measured as Wl, W2, W3, W4 respectively. Another group was loaded under H3 up to a stress level, which was equal to 0.50 Wl, then different constant rates (Hi, H2, H3, H4) were applied.XIV The obtained data indicated that the initial application of the lover rate H3 up to 0.50 Wl, level and then applying the higher rate H4, gave aproximately the same strength- value as when the rate H4 was applied alone from zero to failure. This result proved, that the reduction in strength with slower rates of loading was related to the higher stress levels. A physical explanation for this could be given by considering that crack developpment was also a contribuiting factor in the effect of rate of loading upon the strength of concrete. The following final conclusions can be drown; 1. The strength of a saturated specimen was affected by loading rate and increased about 40 percent when the loading time was reduced from 4 hours to 1 second. 2. The moisture content of the specimen affected the strength-loading rate relationship, and the magnitude of this effect decreased with decreasing moisture content of the specimen. In hardened cement paste, mortar and concrete specimens, with no moisture, the effect of loading rate on strength was negligible. These results indicated that moisture content was one of the most important factors influencing the strength-loading rate relationship. 3. The age of the specimens at testing was found to be an other effective factor upon the stength-loading rate relationship. Reduction in strength due to different constant rates of loading, increased with increasing concrete age, and different types of strength-log (rate) curves have been drawn for different ages. 4. Curing conditions of the specimens also influenced the strength-loading rate relationship, and different types of strength-log (rate) curve have been found for different curing conditions.XV 5. An abrupt change from a slover rate of loading tc a faster one, increased the strength, when it was made at or below a stress level of 30 percent of the final strength, 6. It is possible to interpret the above results by assuming; a) The water phase in the specimens played a contributing part in the strength -loading rate relation of the composite meter ial, b) The effect of loading rate was influenced by the amount of microcracking in the specimen.