La végétation forestière de la Kroumirie

International audience

英语. The present thesis deals with the study of the mechanical properties of pit props relative to various factors of forestry production.Its purposes are as follows : t . to show how wood used as supporting membres in a minebehaves in static bending and in sag, also to give some figures for the strength wood opposes to these stresses. 2. to find how these mechanical properties will vary (as well as warning signs given by wood before failure, and the aspect of the wood fiber where the supporting member has failed relative to the main factors of forestry production: the species, the location of the tree and the type of management adopted. 3. to draw the relation that exists between mechanical strength, warning signs and the aspect of wood fibers at rupture on one hand and between certain aspects or properties of wood such as density, width of growth rings, moisture content, dimensions and differences in shape, structure or chemical composition on the other hand. In spite of considerable progress achieved these last few years by metal props, wood retains an important place in the mine when difficult working conditions, the type of exploitation or kinds of stresses call for this material. Wooden props consumption seems to reach a limit between 20 to 25 dm3 per ton of coal produced. These figures, considering the increasing coal production will always result in a large use of wood. But mainly when one makes a comparison with supporting members or metal frames it is likely that users will be more and more particular about the quality of wood props, which is quite fair and which one should wish. Mine props will be chosen from better selected wood. Therefore, it is important to have a good knowledge of wood mechanical properties to investigate the factors that influence these mechanical properties and to work out new forestry managments that will produce a better raw material in the best conditions. About fourty different species have been investigated, mainly indigenous species from France but also foreign species which have been cultivated in our country for a various length of time. Furthermore, two bunches of mine props from Finland (spruce and Scotch Pine) and a bunch of mangrove from French Guinea have been tested for comparison. Among the fourty species studied there were 16 coniferous and 24 deciduous species. In each of the two categories one single species (Scotch pine for the softwoods and oak for the hardwoods) was chosen as a reference species which was studied in details and then compared with other species. As most of the species tested here grow in many parts of the country under various conditions, depending of the station and of the type of management adopted it has been necessary to consider various test specimens for each species which were chosen according to special conditions so that the number of test specimens exceeds 350. For certain important species, the number of test bunches was particularly high (42 bunches of Scotch pine, 30 bunches of Maritime pine (Pinus Pinaster). Finally, in order to draw valuable means and to study within a bunch of specimens the interaction of certain variables such as storage, or the influence of defects, etc... it has been necessary to prepare bunches of specimens that include a great many pieces. Sampling amounted to an average of 25 to 100 pieces. In all, over 15 000 pieces have been experimented upon.Testing Procedures Tests have been performed in compression along the grain andin bending. The ratio between length and diameter generally exceeded io, compression tests were actually sag tests. Specimens tested in bending laid freely on two supports and the load was applied at center by means of a rounded-edge knife. The testing machine consisted in hydraulic or mechanical presses with a measuring device of the scale type, including a dynamometer of manometer and a measure of the corresponding deflection or crush. The load at rupture was either registered during the test on drum register or plootted afterward for each result obtained, each one corresponding to a'point on the curve. Presses used enabled us to perform tests on specimens as long as 2.50 meters under a pressure exceeding one hundred tons. Individual diagrams give the following information: 1. the speed at wich the load was applied before rupture, 2. a general aspect of the elastic portion which takes place at the beginning of the test, 3. the value of the load applied on the test specimen (warning load) at the precise time when it begins to give the warning signs (either by sight or hearing), 4. the maximum load and corresponding deflection, 5. the aspect of the rupture, 6. the work at failure which can be divided into various fractions as indicated on graph No. 1 & by parallels to the « y »axes. The portion of the curve called of actual safety which corresponds to the breaking work in bending between the first warning sign (by sight) and the point where we come closer to maximum load is particularly important and must be taken into consideration. The portion called nerve safety which lay beyond a parallel to the « y » axes corresponding to maximum load is also important and must be translated into figures. During the tests warning signs before failure, either by sight (deflection) or hearing (cracks) were carefully noted. The aspect of the rupture, the moisture content at the time of the test, the density determination, the growth ring counting (age) are also among information which were written down for each test specimen. Presentation and Judging of Results. Results were presented as said above, either as curves or as momentary loads with the corresponding deflections. These figures enables us to draw or calculate the following results: 1. aspect of the average curve at rupture for a given bunch, at a given moisture content, 2. a comparison of curves within a given bunch, arranged according to the decreasing diameters or according to the decreasing moisture contents, 3. the average load for each bunch or part of a bunch of specimensand the girth corresponding to a maximum average load per square centimeter (cm2), 4. in bending, the average load corresponding to the first warning sign by sight (warning load) with the girth corresponding to the average warning load per square centimeter (cm2,5). for important bunches we have drawn a distribution diagramfor maximum load per cm2 with the total result figures obtained in order to study the position variables and dispersion variables (see Table I, graph. 103 and 109). The study of these variables has enabled in many cases to see how the observed distribution comes close to a normal distribution and to figure the percentage of results included within a scale of given values, either near the average or above a given figure, 6. the factor analysis within an certain number of results also enabled us to show clearly the influence of certain occasional variation factors such as accidental variation in shape, or structure, abnormal humidity, stain or dote, etc... 7. Finally, for most of the bunches tested studies of static interdependance have been performed which result in correlative graphs. The main correlativeness drawn was between diameter and maximum load, but we also studied some cases of correlativeness between density, moisture content, soundness in relation to maximum load per cm2. Commentaries on Results: It is not possible of course, in that quick presentation, to give all our results. We will limit ourselves to average results for the main species with some commentaries about the influence of the countries where the timber has grown. a) coniferous species Among coniferous species the major part has been given to four home grown species: Scots pine, spruce, fir and maritime pine. The sag test give comparative results for spruce and fir, 200 kgper square centimeter for pieces measuring between 45 and 70 cm in girth at their middle and 2,50 m long. Scots pine gives inferior results 175 kg, and maritime pine (because of its wide growthrings) is definetely weaker,129 kg. In bending, over 2 meters span still for pieces of 40 to 70 cm in girth results obtained for the above four species are nearly identical ranging between 18 and 18,6 kg per cm2. But, on the other hand, warning signs by sight hearing, given before failure by Scots pine and maritime pine are much more important than those given by either fir or spruce and their rupture is not so fast. It is important to notice that the scattering of results is less pronounced for species such as fir (which has not been cultivated extensively outside its natural area in France) than for spruce (frequently introduced on lower grounds), Scots pine (for which there are very poor examples) or maritime pine (which often show very wide growth rings). The well-known relation between narrow rings, high percentage of summer wood, density and higher mechanical properties has been confirmed and put into figures during tests performed on mine props. With Scots pine, for example, when the width of growth rings changes from 3 to 1 mm, density at 15 % rises from 0,55 to 0,70 and the average maximum load per cm2 in the sag test rises from 125 to 225 kg (see table 8). With spruce resistances measured according to origins vary between 170 and 220 kg and with fir between 152 and 227. With maritime pine, there is a wide range: 85 kg per cm2 for the worstand 185 kg for the best (Forest of Dom de Bormes). Everything that hampers the growth of coniferous trees without having any action upon the regularity of growth rings or the physiologic state of the tree improve the mechanical resistance of the wood. The selection of pines growing on mountains with narrow and even rings, the practise of dense forest, the growth of trees on spoor soil (but properly watered) with not too warm a station, are favorable factors for a harvest of wood with high mechanical resistances. To these well-known coniferous species we were able to compare other species that are not so frequently used but among which many give excellent results. In sag, for example, the following loads have been registered: European larch 194 kg per cm2 Cedar (from North Africa) 193 kg Corsican pine T82 kgDouglas 181 kg. These species compare favorably the species taken as standard. We shall notice that larch that comes from some altitude with narrow rings gave a top figure of 243 kg per cm2. Next to the above species are those with relatively rapid growth which can he placed between Scots pine and maritime pine: Japanese larch 166 kg Austrian pine 154 kg Pinus uncinata 151 kg Laricio from Calabria 142 kg Thuya (western red cedar) 151 kg. A coniferous species with wide rings such as Sitka spruce is classified lower with only 124 kg. The same classification is obtained in bending: Corsican pine 26,1 kg ; Douglas 22,6 ; European larch 20,9 ; Japanese larch 20,6 ; Austrian pine 20,4 ; Cedar 20,3 ; Thuya 18,9, are comparable to the four species studied in this report. Pinus uncinata 17,5, Laricio from 'Calabrica (with wide rings)17,4 and Sitka spruce 14,9 give lower figures. One must note that some species such as European larch, Douglas, Cedar, Corsican pine give very distinct warning signs before rupture (cracking in sag and deflection) and their rupture is not abrupt. The species are durable and as a while have a perfect shape. Such species can actually be « cultivated » to give mine props. b) deciduous species In the sag test some hardwoods give results similar to those obtained with softwoods but the majority remains inferior to softwoods.In bending though, many hardwoods show their superiority oversoftwoods. Here are the results in sag :hornbeam (sound and dry) 187 kgred oak 172 kgcommon oak 169 kg - These species give results fairly close to chose obtained with Scots pine chestnut 144 kg alder (sound and ,dry) 142 kg are comparable in that respect to maritime pine. On the contrary,birch, aspen, lime and even robinia with 120 kg about are inferior. In bending, the results obtained with hardwoods are better: robinia (Robinia pseudoacacia) 32,5 kg ash 29,5 k gred oak 27,6 kg wild cherry 22,6 kg common oak 22,4 kg lime 20,3 kg hornbeam (sound) 20,2 kg chestnut 20,8 kg far behind come: birch 15,8 kg aspen 15,3 kg alder 154 kg As a matter of curiosity let us call mangrove from Guinea whichresisted 300 kg per cm2 in sag and 48,5 kg in bending, but ofcourse, this is a record.Hardwoods give more warning signs by sight before rupture (inbending as well as in sag). Robinia, ash, oak also give a crackingsound (less pronounced though than with pines). Some hardwoods (chestnut, oak, robinia, ash, sound hornbeam)show a slow rupture of their fibers, others have an abrupt rupture(beech, even sound, maple, sorbus aria). Finally, and this is particularlyimportant, some hardwood are not durable and easely attackedby fungi (beech, hornbeam, birch, alder, aspen, Ailanthus)and because of this, lose both: their mechanical properties and theirwarning signs or security advantage.A good many hardwoods could give excellent mine props, buttheir preparation must he very carefully clone : felling when thesap is no more in circulation, grading of the pieces which must bestraight, grooving, storing in well ventilated piles to facilitate airdrying and avoid decay.c) factors (other than species and station of growth) Tc'hich have aninfuence upon mechanical properties.We have tried to evaluate the in fl uence of some factors such as:length (or span)girth (or diameter)straightnessshape of the piece (conicity)egg-like sectioncrooked graincrown of knotsmoisture contentstainetc...1) length or spanIn bending or sag_ , there is generally an inverse relation betweenspan or length and the resistance. which corroborates the generalrules of resistance of materials.One must note, nevertheless, that axial compression test wereactually sag test and not straight compression test along the grain.The relation between length and diameter was alwa.)' over 6.2) diameterIn sag we could not draw any direct relation between load persquare centimeter and section. Among the biggest pieces tested(45 to 70,cm in girth) this relation does not seem to exist (whichis logical as long as one does not change over from sag to compressionalong the grain).Nonetheless, we could notice that the load per square centimetersupported by small pieces (from 18 to 34 cm in girth) was the sameas the load supported by bigger pieces when the former were testedin sag on 1,60 ni only instead of 2,50 ni in length. It wouldthen seem interesting to resume these surveys of relationship basedon the ratio: length over diameter of tested specimens, In bending, we found a relation though not very conspicuous,between the mechanical resistance and thediameter within thehunch of specimens tested over 2 ni span (45 to 70 cm in girth).This relation is not paramount because the girth often brings otherweakness factors such as large knots, greater moisture content,stain, heart checks, etc...The relations studied have enabled us to figure out a shape factorwhich is the « n ]» of the well-known formula of « fatigue »given by the AFNOR standards.3 PLFu —2hh°in whichFt' is the value of fatigue in bending P the load at failure L the spanb and h the horizontal and vertical dimensions of the section n the shape factor.We found for mine props a shape factor slightly lower than the one given by Marcel Monnin for structural lumber n 6/5 instead of 8/6. We have also noticed that smaller pieces (from 18 to 34 cm in girth) showed a resistance in bending per cm2 over a one-meter span identical to the resistance in bending over a two-meter spanshown by large pieces. During the time spent for this survey wecould not study the influence of girth for one single span over a large range of girth (from T8 to 70 cm), but a complementary study could he made on that particular subject.3) straightness Without a detailed study we shall note that one single bend. inferior to 3 cm per meter of length will result in a TO %^ to 20 reduction in the resistance in sag. In the case of a double bend irg S. the reduction will reach 30 %. An exagerated conicity (2 to 2.5 cm per meter, over the diameter) will result in a TS to 20 % reduction in the sag test. If the tree shows too much of an egg shape (1 s 2/3 L) this will also reduce the resistance in bending by 20 °Jo.4) defectsCrown of conspicuous knots of small sizes does not seem tohave any action either in bending or in sag. On the contrary,when knots are over 2 cm in diameter the resistance in bending is reduced, fail re « jumps » toward the crown of knots and becomes abrupt. One notices a reduction in the maximum load ranging from 72 to i8 %. In sag the reduction of resistance appears whenknots are 3 cm in diameter or over. This reduction reaches is to 15 %. Props then crak at the knots whether they are conspicuousor slightly covered. Early artificial pruning which will result in alarger percentage of defectless timber will at the same time increasethe resistance of props.5) moisture contentMoisture content has a definite influence on the resistance ofwood. Standards published by AFNOR give formulae to changethe resistance of a given piece of wood (measured at H% MC) toan average MC (r5 %). For example, the stress at rupture in bending:F15'= FH(r -f- c' (H 1 5))in whichc' is a factor called « humidity factor ».We were able to figure out these factors in sag and in bendingfor some species. We found for example, in sag:2 to 3 % for Douglas fir and mountain larch3 to 4 % for 'fir, spruce and Scots pine4 to 5 % for larch growing in plains and hardwoods (oak, chestnut,hornbeam)5 to 6 % for Austrian black pine and maritime pine with widerings.In bending we found :2 % for Douglas fir, white fir, spruce, larch2 to 3 % for oak, chestnut, hornbeam, Scots pine3 to 4 % for maritime pine and Austrian black pine.This is very important to judge the resistance of a given pieceof timber according to its moisture content. Moisture content variesnot only after felling, to reach the equilibrium with the relativehumidity and the temperature of the storage place but alsoshows a certain up and down variation with an increase in moistureduring cold, wet seasons.6) decaySoftwoods, when they are barked as soon as felled (which isparticularly important in the case of Austrian Black pine) areless exposed to stain. Nevertheless, large pine logs often showstain signs which are not very dangerous in themselves but may hide some worse stain inside the tree. On low grounds one must avoid felling pines when sap is circulating (burnt trees show thesame risks). Decayed wood is easily recognized because it smells of ether and because of a certain hysteresis over summer drying (decayed wood gains more water during wet seasons and loses less duringdry seasons).Many hardwoods (deciduous species) oak, chestnut, locust, etc...are durable, others such as Ailanthus, alder, birch, hornbeam, beech,aspen, etc... are likely to get decayed and sometimes in a very had way. If felled when the sap is in motion they get stained at once. If, on the contrary, they are felled after the sap has stopped circulating inside the tree some of them (birch, hornbeam)can last one year but all get stained during the second year. In that case, the resistance is reduced by half it not more, and rupture is always abrupt (brash) without warning signs .The preparation and use of birch and hornbeam as mine props must be carried out with that serious disadvantage in mind.