Thesis (MScFor (Dept. of Forest and Wood Science)--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: In this study 1345 pieces of 38x114 mm timber sawn from 16-20 year-old Pinus patula trees were obtained from a wide variety of sites along the Mpumalanga escarpment in South Africa. The samples were tested for various mechanical and physical properties. The objectives of this study were (1) to determine the variation in the bending strength, tension strength, and stiffness of the sawn timber, (2) to assess the efficiency of the current visual and mechanical grading rules on this sample, and (3) to evaluate the potential of some indicator properties to be used as structural grading parameters on this resource. A large number of non-destructive measurements were performed on all the samples, including scanning of the boards with a commercial X-ray density scanner, moisture content measurement, growth ring measurements, warp measurement, acoustic frequency measurement and measurement of the stiffness using a mechanical grader. The data from the X-ray density scanner was used to calculate a number of knot-related variables for each board. Visual grading according to the SANS 1783-2 (2005) specifications for structural timber was performed by a certified grader on all the boards. The timber was divided into two groups for destructive testing, one group each for bending and tension tests. Tests were conducted according to the SANS 6122 (1994) method for in-grade testing. Due to the sampling method followed, the destructive tests effectively used a random placement of defects with regard to the load application position. From the destructive tests the modulus of elasticity (MOEedge), bending strength (MOR) and tensile strength were determined. The study showed that the MOEedge of the sawn timber was far below the requirements of the current national standard (SANS 10163-1) for all of the visual and mechanical grades. The mean MOEedge of the visually graded S5 timber was about 30% lower than required. The 5th percentile values for bending strength of all the visual and mechanical grades were above the required values of SANS 10163-1. The 5th percentile values for tensile strength of all the visual and mechanical grades were similar to that required by the SANS 10163-1 standard. Correlations between flatwise measured MOE (MOEflat) and edgewise measured MOE (MOEedge) were smaller than expected, as well as the correlations between both MOEedge and MOEflat with MOR. Dynamic MOE (MOEdyn), calculated from acoustic frequency tests on the timber, was found to be the best single predictor of MOEedge, MOR and tension strength. Multiple regression analysis showed that a combination of MOEdyn, density and knot parameters can be used to improve the predictability of some of the strength and stiffness characteristics of the timber. It is recommended that a comprehensive study on the structural grading of SA Pine be performed which includes (1) an analysis of market requirements in terms of strength and stiffness properties of timber, (2) in-grade testing of a representative sample of structural timber in South Africa, and (3) a review of the standards used in South Africa to regulate the structural grading of timber. / AFRIKAANSE OPSOMMING: In hierdie studie is 1345 stukke 38x114 mm 16-20 jaar-oue Pinus patula planke, afkomstig van ‘n wye verskydenheid groeiplekke teen die Mpumalanga platorand in Suid Afrika, gebruik. Die planke is getoets vir verskeie meganiese sowel as fisiese eienskappe. Die doelwitte van hierdie studie was om (1) die variasie in buigsterkte, treksterkte en styfheid van die gesaagde planke te bepaal, (2) die effektiwiteit van die huidige visuele -en meganiese graderingsreëls op hierdie monster planke vas te stel, en (3) die potensiaal te evalueer van sommige eienskappe wat gebruik kan word as strukturele graderingsparameters. ‘n Groot hoeveelheid nie-destruktiewe toetse is op alle planke uitgevoer, wat ingesluit het skandering van planke met ‘n kommersiële X-straaldigt-heidskandeerder, metings van voginhoud, groeiringwydtes, deformasie, akoestiese frekwensie en die bepaling van styfheid met behulp van ‘n meganiese gradeerder. Die data van die X-straalskandeerder is gebruik om ‘n aantal kwasverwante veranderlikes vir elke plank te bereken. Visuele gradering is op alle planke uitgevoer ooreenkomstig met die SANS 1783-2 (2005) spesifikasies vir strukturele hout deur ‘n gesertifiseerde gradeerder. Die hout is in twee groepe opgedeel vir destruktiewe toetse, een vir buigtoetse en een vir trektoetse, ooreenkomstig met die SANS 6122 (1994) metode vir binnegraadse toetse. As gevolg van die monsternemingsmetodiek wat gevolg is, is daar effektief gebruik gemaak van ‘n lukrake plasing van defekte met betrekking tot die las-aanwendingsposisie. Die modulus van elastisiteit (MOEedge), buigsterkte (MOR) en treksterkte is deur middel van destruktiewe toetsresultate bepaal. Die studie het aangetoon dat die MOEedge van gesaagde hout aansienlik minder as die vereiste van die huidige nasionale standaard (SANS 10163-1) is vir al die visuele en meganiese grade. Die gemiddelde MOEedge van die visueel-gegradeerde S5 planke was omtrent 30% laer as vereis. Die 5de persentiel waardes vir buigsterkte van alle visuele en meganiese grade was hoër as die vereiste waardes soos voorgeskryf deur SANS 10163-1. Die 5de persentiel waardes vir treksterkte van alle visuele en meganiese grade was gelykstaande aan wat vereis word deur SANS 10163-1. Korrelasies tussen MOE, gemeet op die wydte sy (MOEflat), en MOE, gemeet op die diktesy (MOEedge), asook die korrelasies van beide MOEedge en MOEflat met MOR van die planke was laer as verwag. Dinamiese MOE (MOEdyn), wat bereken was vanaf die akoestiese frekwensie resultate, is vasgestel as die beste enkele indikator van MOEedge, MOR en treksterkte. Meervuldige regressie analise het aangetoon dat ‘n kombinasie van MOEdyn, digtheid en kwasparameters gebruik kan word om die voorspelbaarheid van sommige van die sterkte- en styfheids eienskappe van die hout te verbeter. Daar word aanbeveel dat ‘n omvattende studie gedoen word op die strukturele gradering van SA Dennehout, wat insluit (1) ‘n analise van die markbehoeftes in terme van die sterkte- en styfheids eienskappe van hout, (2) binnegraadse toetsing van ‘n verteenwoordigende monster strukturele hout in Suid Afrika, en (3) die hersiening van standaarde in gebruik in Suid Afrika om die strukturele gradering van hout te reguleer.