Category Tensile Surface Structures: A. Practical Guide to Cable and. Membrane Construction

The present book has attempted to give an overview of the current state of the art in the implementation of tensile sur­face structures. The purpose of the work has been to permit the recognition of fundamental interactions between the production and the erection of flexible structural elements.

The investigation of the factors affecting the construction process, as result from the manufacture of flexible structural elements, illustrates the interdependency of production and erection processes. The emphasis has been the production and assembly technology of coated textile sheet materials.

These investigations have been based on the collection and structuring of the highly specialised knowledge of designers, manufacturers and erectors. This, and also the classification and definitions of the criteria relevant for construction, can be seen as the main contribution of this book to the practical implementation of wide-span, lightweight structures

Starting from the textile characteristics and internal structure of flexible materials, the first main part of the book describes their composition and manufacture and explains how the technicalities of their manufacture determine the complex material behaviour and how they function as structural ele­ments A summarised description of the mechanical behav­iour of coated fabrics was necessary for the evaluation of the influence of the materials used as flexible structural elements on the construction process. In addition to the internal struc­ture of coated fabric, which is constructed to meet its specific purpose, it is mainly the weave type used in the production process, which has a substantial effect on the deformation behaviour under the influence of load, time and temperature during erection. The interaction of the thread directions de­pending on the calculated compensation value is of central importance for the erection of membrane surfaces.

Another example of the reciprocal influence of the mechani­cal properties of the material between production and erec­tion is the patterning of textile sheet elements. The definition of criteria for the patterning showed how the determination how the design of the spatial cutting-out pattern of the indi­vidual fabric strips on the membrane surface affects the load transfer and the erection. It was demonstrated here what ef­fects the selection of material and also the determination of the tensioning direction and the tensioning sequence have on the erection of membrane surfaces. It can be stated re­garding the topographical and structural criteria that the type of material, the shape of the surface, strip layout, strip shape, seam geometry, seam arrangement and the geome­try and type of edging all interact, which considerably affects the behaviour and interaction of the structural elements.

Statements about these relationships are, however, only pos­sible after detailed investigations and tests. Regarding the parameters of form and load-bearing behaviour, which af­fect the patterning, there is a need for more detailed analytic investigations into this complex area of work. One possible starting point would be to research the relationship of rise height and main curvature to possibly achievable spans and strip lengths of anisotropic membrane surfaces.

A further section illustrates in the first part examples of ways of jointing the surface and methods of introducing force into the edge, and their effects on the structural behaviour and special factors affecting manageability and ease of erection. The starting point for further considerations in this area was the detailing of corner, edge and surface connections so as to make them buildable and manageable.

The second main part dealt with the erection of flexible structural elements. The description of the economic and technical conditions for construction management looked into the planning forecast of the construction process. The factors resulting from production practicalities on the schedul і ng and the influences and areas of responsibility of construction planning concerning the design and construc­tion of membrane structures. Ways of modelling the erection activities and aspects of detailing in such a way as to increase the practicality of erection were also discussed.

An illustration and a description of the equipment used de­scribes the role of the tools and equipment used in transport, lifting and tensioning, and explains the resulting parameters for the design process.

Building on the foundation of production and construction technology parameters, the erection principles and influen­tial factors affecting the erection process were defined and investigated. In addition to the type and the jointing of the materials used, the local conditions on the construction site, the type of equipment used for erection and the practicali­ties of production, delivery and transport, it is above all the working principle of the structural system, which decisively influences the choice of a suitable erection procedure. The emphasis of this section is therefore a description of tempo­rary states during the erection of form-active structural sys­tems seen from the practical point of view. The nature of the interaction of primary structure, sub-structure and second­ary structure elements was investigated regarding the erec­tion technology. Stabilisation measures for the erection of form-active structures were discussed through a comparison of the loading of structures and structural elements during erection with parameters from the safety and compensation plan, taking construction technology into account.

In a further section, erection procedures for mechanically tensioned membrane structures were described and illus­trated with examples, and the erection sequence and con­struction activities for the erection of characteristic forms of structure were explained. In order to do this, it was also nec­essary to go into construction problems in the areas of deliv­ery, handling and jointing and categorise and describe the individual erection implementation activities. The documen­tation of completed projects and schematic diagrams visual­ises these themes.

The main emphasis of the discussion of the construction process was the categorisation and description of the proce­dures for introducing the pretensioning into flexible structur­al elements from the practical point of view. The tensioning of flexible structural elements is of central importance in the erection of membrane structures. Sufficient stability of the structural element and the structural element is not created until the tension has been applied to the linear and sheet el­ements. The growth and creation of the structure through the successive introduction and constant increase of the pre­tensioning forces differentiates the erection of wide-span, lightweight structures intrinsically from conventional con­struction procedures, where the erection of the structure occurs through constant addition of new elements. In addi­tion to the preassembly, lifting and assembly, the tensioning process is a major characteristic of the erection of wide-span lightweight structures.

Finally, an overview was presented of processes for measur­ing forces in ropes and membranes on the construction site.

In summary, it can be observed that an overall assessment of the erection of wide-span lightweight structures is very com­plex, on account of the multitude of influential factors. This type of structure is a unique product, with each new project producing new conditions and different aims for erection. The main parameters affecting the erection of membrane structures are the deformation behaviour of the material used, its jointing and edging and the determination of the spatial cutting patterns. The manner of interaction with the primary structure is also essential for the installation of the flexible elements.

In the opinion of the author, great potential for optimisa­tion of the erection lies in the production of the sheet tex­tile elements. New developments in materials, which offer more satisfactory solutions for the elastic and load-bearing behaviour under the effects of load, time and temperature, jointing methods and the requirements arising from the load-independent influences on the material, could con­siderably simplify the construction technology. Above all the deformation behaviour of coated fabrics and the effect of it on the tensioning process could offer potential for future development.

In summary, it can be observed that the erection of form – active structures is a very complex field. In order to avoid mistakes and shortcomings in the implementation on site, the importance of erection practicalities should be award­ed sufficient importance at an early stage by architects and engineers responsible for the design of wide-span light­weight structures.

[1] Ferguson, E. S. (1993)

[2] Buckminster Fuller, R. (1973)

[3] Definitions according to the Austrian standard ONORM M 9500

[4] Definition according to European Standard EN 12385-2

[5] Trurnit, P. D. (1981)

[6] PFEIFER company material (2003)

[7] Peil, U. (2002)

[8] Feyrer, K. (1986)

[9] Scheffler, M. (1994)

[10] Stauske, D. (2002)

[11] Mogk, R. (2000) 3 Verreet, R. (1996-1)

[12] Gabriel, K.; Wagner R. (1992)

[13] Mogk, R. (2000)

[14] Verreet, R. (1996-2)

[15] Vogel, W. (2002)

[16] Mogk, R. (2000)

[17] Stauske, D. (2002)

[18] Stauske, D. (1990)

[19] Detailed descriptions of tensioning equipment and devices for ropes can be found in section 3.3.2

[20] Fluoroplastics films, which are primarily used in pillow construction,

must in contrast fulfil all functions.

[22] Funk, J. (2005)

[23] Domininghaus, H. (1992)

[24] Fitz, H. (1989)

[25] Fitz, H. (2004) 4 Herrmann, H. (1986)

[26] Fitz, H. (2004)

[27] Schwarz, O.; Ebeling, F.-W.; Furth, B. (1999)

[28] Nentwing, J. (2000)

[29] Schlaich, JWagner, A. (1988)

[30] Blum, R. (1990)

[31] Blum, R. (1990)

[32] Remark: fabrics, which are held in weft direction during weaving, form an exception here, as their thread crimp is almost equal in warp and weft direction.

[33] The tension test shown in Fig 73 was performed by Labor Blum in Stuttgart

(project engineers IF-Group, Reichenau).

[35] Minte, J. (1981)

[36] Blum, R. (1990)

[37] Minte, J. (1981)

[38] Sobek, W.; Speth, M. (1995)

[39] Moncrieff, E.; Grnndig, L. (1999)

[40] Essrich, R. (2004)

[41] Ziegler R.; Wagner, W. (2001)

[42] Remark: No publication is known to the author, in w

[43] Essrich, R. (2004)

[44] Fitz, H. (2004)

[45] DVSguideline2225 (1991)

[46] Blum, R. (2002-2)

[47] Buhner, E. (1997)

[48] Essrich, R. (2004)

[49] Otto, F; Happold, E.; Bubner, E. (1982)

[50] Sobek, W. (1994)

[51] Stimpfe, B. (2000) 3 Stimpfe, B. (2000)

[52] Minte, J. (1981) 4 Sobek, W. (1994)

a Filling chain b Hose fabric c Connection fabric

[54] Gropper, H.; Sobek, W. (1985)

Remark: No membrane structures with this type of edge detail are known to the author

[55] Bubner, E. (1997)

[56] Minte J. (1981)

[57] Buhner, E. (1997)

[58] Sobek, Ww Gropper, H. (1985)

[59] Gropper, H.; Sobek, W. (1985)

[60] Albrecht, R. (1973)

1 Petzschmann, E.; Bauer, H. (1991) 2 Cenotec (1999)

[62] Rudorf-Witrin, W. (1999)

[63] Teschner, R. (2004)

[64] To simulate the structural reality using a geometrical model, the relationship of weight to 2 Philipp Holzmann – construction documentation (1988) volume (8-fold increase of weight according to volume) and the relationship of the load­bearing capacity of the structural element with the increase of cross-section (factor 4) must be reconciled. This requires the application of an additional loading to the model.

[65] Petzschmann, E.; Bauer, H. (1991)

[66] DIN536-1 to DIN 15030 (1995)

[67] Drees, G.; KrauB, S. (2002)

[68] Junker, D. (2004)

[69] Inauen, B. (2003)

[70] Eberspacher company information (2003)

[71] Imgruth, H. (2002)

[72] Remark: Weighted hanging roofs made of fabric are practically never built, or else they would also appear in this group.

[73] Siokola, W. (2004)

[74] Miller, P. W. (2000)

[75] Albrecht, R. (1973)

[76] Seliger, P. M. (1989)

[77] Sischka, J.; Stadler, F. (2003)

[78] Zechner, M. (2005)

[79] Lorenz, T; Mandl, P.; Siokola, W.; Zechner, M. (2004)

[80] Bergermann, R.; Goppert, K.; Schlaich, J. (1995)

[81] Bergermann, R.; Goppert, K.; Schlaich, J. (1995)

[82] Alpermann, H.; Gengnagel, C. (2003)

[83] Holst, S. (2006)

[84] Sobek, W. Linder, J.; Krampen, J. (2004)

[85] Rudorf-Witrin, W, Stimpfe, B.; Blumel, S.; Pasternak, H. (2006)

[86] Blumel, S., Stimpfe, B.; Rudorf-Witrin, W. Pasternak, H. (2005)

[87] Inauen, B. (2003)

[88] Habermann, K. J.; Schittich, C. (1994)

[89] Durr, H. (1988)

[90] Bergermann, R.; Sobek, W. (1992)

[91] Schlaich, M. (2000)

[92] Bukor, S. (2003)

[93] Ryser, R.; Badoux, J.-C. (2002)

[94] Ryser, R.; Badoux, J.-C. (2002)

[95] Durr, H. (2002)

[96] Baumuller, D. (2000)

[97] Basic statements about the influence of cutting patterns on the tensioning of fabrics with dissimilar stretch properties have been made in section 24.3.1.

[98] Essrich, R. (2004)

[99] Bohmer, C. (2004)

[100] Lenk, S. (2004)

[101] Labor Blum company material

[102] Labor Blum company material

[103] Labor Blum company material

[104] Ramberger, G. (1978)

[105] Durr, H. (2000)

[106] Blum, R. (1982)

[107] Plunger

[108] Rotating disc

[109] Ring force gauge

[110] Base ring with partial vacuum

[111] Potentiometer

[112] Membrane

In order to avoid permissible stresses being exceeded, check measurements have to be performed on installed and ten­sioned membranes. The measurement of membrane stress is, however, problematic.

Measurement of the strain with strain gauges is difficult to handle and evaluate. The strip has to be attached to the membrane surface when this is stress-free in order to be able to determine the strains after the loading has been applied. It is hardly practical to protect the sensitive strain gauge dur­ing the process of erection. The evaluation of the strains in order to determine the stress is also very difficult on account of the anisotropic and viscoelestic behaviour of the fabric. Other procedures for measuring membrane stress also have problems with the curvature of the membrane surface and the inhomogeneous (locally not constant) stress conditions to be expected.[106]

Fig. 368: Three-point bending measurement

Three methods of measuring stress developed by Labor Blum are described below

The first method is based on the observation of the spread of acceleration waves, which start from a short impact on the membrane surface. Two concentric circles are drawn round the impact point at defined diameters and accelerometers are fixed at the intersection of the circles with rays project­ing from the centre (Fig. 369). After the impulse has been generated, the movement condition of the membrane is re­corded, forwarded to an oscilloscope as electrical signal and displayed on a screen. On the basis of three measurements in any direction, the maximum and the minimum time and thus the speed of the elliptical spread of the wave, which is needed as the basis for the calculation of stress.[107]

In the second process for determining the stress situation, a pressure depression is made in the surface by applying a de­
fined force. A plate-shaped device is used for this, which presses a plunger within a ring against the membrane surface (Fig. 370). The deflections and stresses arising in the membrane surface can then be calculated.

The third measurement method describes a measuring de­vice developed for test purposes at Labor Blum, which is con­structed with two potentiometers, a ring force measuring device, a turning disc and an annulus with press-down ring.

To perform a measurement, the base ring is pressed onto the membrane by a partial vacuum and a defined point load is applied in the axis of the base ring. The rotating disc with two potentiometers mounted on it is set turning and the de­flection of the concentric rings is measured. During the rota­tion, the two potentiometers show the deflection curve as a function of radius, with which the relationship to the defined load can be calculated (Fig. 371).[108] [109] [110] [111] [112]

The best-known method of determining the force in ropes in steel erection is the measurement of the compression in hy­draulic cylinders installed in the flow of the forces during the pretensioning process. Various intermediate measurements are undertaken in the course of force introduction into the rope by hydraulic presses controlled according to force or travel. The compression force in the press is read from a cali­brated manometer until the permissible press lift as speci­fied in the erection plan has been reached, and is recorded together with the relevant piston travel. The applied force can be calculated with knowledge of the data for the device
(area of the press cylinder). After the full pretension has been reached and the press compression has dropped off, the tensioning equipment can be removed. It is often necessary to readjust the rope length when using this process.

One disadvantage of this method is the size and weight of the presses and the associated work in setting up in a new location. The high weight of presses can well require use of a crane.

Vibrating wire process

In this electrical measurement process, the resonance fre­quency of a string clamped into a holder is measured dur­ing alteration of length and converted. When a tension force is applied, the tension in the wire changes and thus its resonance frequency. The wire can be resonated by a piezo­electric or an electromagnetic component transverse to the

I ongitudinal axis, the measured electrical voltage converted and fed to an electronic counter device.

The vibrating wire sensor is clamped to the rope in a stress – free state. Stress-free ropes have a slight bend. In order to compensate for this bending and achieve exact values, it is necessary for this process to mount 2 sensors in one plane (Fig. 367). The strain in the rope then results from the aver­age resonance frequency value of the shortened and the lengthened wires. The advantages of this method are long­term stability and high precision. It is therefore often used for long-term monitoring. One disadvantage is that the sensor needs a relatively high setting force for adjustment.1

Three-point bending measurement

This method of measurement is performed by applying a deflection of defined extent to the rope and measuring the required force. The rope is laid on saddles and is given a de­fined angle by a hand-operated central tensioning shoe. If the rope is under load, the force in it works against the angle of deviation and acts upon a built-in electric load cell, which sends a signal dependant on the force to a process-con­trolled digital display (Fig. 368). This measurement is not suit­able for long-term use.

For the measurements in connection with the rope fapade of the Cologne airport, a rope force meter based on the three – point bending test was further developed, which in addition
to the force can also measure the deflection. This makes it pos­sible to work with smaller deflections. Development work is also aimed at using hinged saddles to distribute the unavoid­able increase of strain at this location more uniformly over the entire length of the rope. A built-in microprocessor digitises the measured data of force and deflection and makes it avail­able for further processing. This device can be used to meas­ure ropes up to 38 mm diameter and forces of 250 kN.[105]

One method of determining the forces in ropes is the mea­surement of strain, with the rope itself being used as force measurement element. To do this, an extensometer is at­tached to the rope in the unstressed state, which measures the initial length and the length under tension. This enables the force in the rope to be calculated from the strain and the strain stiffness. A whole range of piezoelectric or induc­tive sensors or also vibration sensors can be used for this pro­cess. It is essential that the sensor is clamped to the rope in a stress-free state and is not disturbed during erection (a in Fig. 366).[102]

Another way of determining the force in a rope is to mount the extensometer on the pretensioned rope and then to un­load the rope (b in Fig. 366). The force is determined analog­ously to the above example. This method can, however, only be used with suitable tensioning apparatus and is therefore restricted to relatively thin ropes and rope forces.

Edge ropes in rope nets mostly have large diameter and high rope forces, which make an unloading unnecessary. In this case it is possible to measure the forces in the net rope run­ning into the edge rope using the unloading method and determine the force out of the equilibrium conditions with the geometrical and loading values for net and edge rope.[103]

Direct or indirect methods of measurement can be used to determine the tension forces in ropes. A direct method of measuring the tension forces can be carried out with dyna­mometers installed between hydraulic presses or by meas­uring the pressure with a manometer. Indirect methods can mean measuring the geometrical deflection sag or by meas­uring the resonance frequency.[104]

The assembling of different materials and the involvement of all parts of the structure in load transfer places high de­mands on the geometry of lightweight structures. The de­formations of the load-bearing elements in tension result­ing from form-active interaction limit each other mutually. The coordination of the differing strains of the materials is therefore fundamental for the load-bearing behaviour of structures loaded in tension. This situation demands that measurements of force and geometry are performed on the individual parts of the structure during the tensioning pro­cess. In the course of applying the loading, unintended load transfer could lead to constraint forces and can lead to local overloading and destruction of the ropes. In order to avoid this, force measurement members are provided between the primary construction, tensioning tools and the flexible load­bearing elements, and the values occurring during tension­ing are compared with the intended values and recorded. The final state and the geometry also have to be monitored. In cases where long-term monitoring is needed, permanent­ly installed measurement devices are used.

There are a number of processes for the measurement of rope forces, but it is not always possible to determine an ide­al method of measurement. The suitability of a measurement process depends mainly on the rope measurements, the de­gree of pretension, the work involved in installation and local conditions for force measurements. Often several methods are combined for one structure in order to achieve economi­cally justifiable and sufficiently exact measurement results. For example, four different measuring procedures were used for the measurements of rope forces during the erection of the Forum roof at the Sony Center in Berlin.1

In addition to the type of rope, the rope diameter and the ex­pected force in the rope, other criteria like current tempera­ture, wind conditions and manageability can influence the selection of measuring method.

Forces in structural elements are not directly measurable, they can only be defined by their effect on suitable measur­ing equipment. Because of the dependence of the force on the mass and acceleration of a body, there is a direct relation­ship of force to defined physical values like elasticity (strain, extension), pressure or piezo-electricity.

A commonly used method for measuring force is the exploi­tation of the elastic deformation of solid bodies. Mechani­cal, hydraulic or electrical measuring devices record the force
to be measured through deformation of the measuring ele­ment directly in the flow of force. The force measurement can be undertaken with measuring elements, which work with elongation, bending or shear.[101]

Further methods of checking rope forces are the setting off of vibrations in the rope, where the force is calculated from the measured resonance frequency, and the measurement of the static deflection, from which the force can be calculated if the rope length, cross-sectional diameter and material data of the rope are known.

The best-known process in strain measurement technology is measuring with strain gauges. These tolerate a high number of load cycles, only have a small dead weight and are used in a multitude of applications as measurement sensor. They are applied by sticking them to the body to be measured.

Strain gauges work on the basis of the alteration of the resis­tance of an elongated wire or strip of metal foil. The me­ander-shaped resistance wire, which is fixed to a carrier foil or a corresponding metal foil, which is similarly structured by etching or vaporisation, is elongated by the strain. This makes the cross-sectional area of the wire smaller and alters its crystalline structure. This effect on the wire alters its elec­trical resistance and this can be calculated from the specific resistance of the material and its length.

The disadvantage of the strain gauge for measuring the strain in ropes is lack of practicality, considering the often uneven surface and the measurement results. It is often un­clear which strain is being measured, because the surface of a rope can show a different strain then the strands or the insert. For measuring the strain in ropes, the strain gauge is mostly installed in force measurement boxes.

Synclastic surfaces cannot be pretensioned by tangen­tial displacement of elements in the surface. They have to be tensioned by applying surface loading perpendicular to the plane. The pretensioning of rope hanging roofs is done by ballasting. This creates a loading condition, which en­sures that no compression will occur in the tension elements through external influences like wind uplift or snow. The application of surface loading by ballasting can be seen not as actual pretensioning, but as a loading condition.

Pneumatic construction makes use of the medium air as pressure dissipating element for the envelope. The surface loading results from the low difference of pressure between the air inside and outside. The forms of pneumatic construc­tions always follow the general formula for the stress in the walls of a pressure vessel. Loads acting inwards are support­ed by the pneumatic envelope and lead to a reduction of the stress in the membrane. Loads acting outwards are dis­tributed in the plane of the envelope and mostly increase the stress.1 The types of pneumatic construction are air-sup­ported halls, large pillows and compartmental pillows. Suit­able materials for the envelope are PVC-coated polyester fabrics and, for limited spans, ETFE foils.

Air supported halls have to be anchored around the peri­meter foundation. The form of the foundation line has to
correspond to the correct form of the edge of the envelope under long-term stress. The factors, which determine the edge detail, are the stretch properties of the envelope mate­rial and the surface curvature in the edge area.

The envelope expands in all directions during the inflation process. The enlargement is, however, not uniform. It can happen that the stretch in one direction is very large while the material in the other direction tends to shrink.2 The risk of wrinkles forming in the area near the edge therefore has to be investigated. A high arch rise and steeply inclined side walls lead to strong deformation under loading, which load the anchoring elements at the edge. The resulting rotations at the bearing have to be taken by the edge element. The standard edge detail for air-supported halls is a steel tube in a membrane sleeve. Other possible anchorages are clamp­ing plate edges and laced edges at the foundation.

Flat dome forms can be connected to a bearing compression ring, which can also take horizontal forces. Ballast bodies for temporary foundations can be fabric tubes filled with earth, sand, gravel or excavation spoil, or precast elements filled with water (Fig. 363).

To maintain the air pressure in pneumatic structures, blower units are used, which can be regulated and controlled. These supply the membrane envelope though a duct system with

an excess pressure of 200 – approx. 500 Pa. One exception is high-pressure tyres, which are supplied with pressures of about 0.5 bar.

Several blowers are normally used together to supply air – supported halls and pillow constructions, blowing air into the hall either alternatively or through a common connection. When they work alternatively, non-return flaps are installed in the blower ductwork, so that if one blower stops the air will not escape back though it. In order to protect the mate­rial of the envelope, higher pressures should only be used when the corresponding loading case arises. When the pres­sure falls under the required minimum value, further blowers can be switched on to give support. A pressure transmitter can be installed for pressure measurement.

In order to achieve the best possible transparency, ETFE foils are being increasingly used for constructing pillows. Inflow of moisture can, however, obscure the light coming through. In order to make sure that as little moisture gets into the pil­low as possible, there should be a dehumidifier before the blower. The inclusion of a condensate separator, which has to be checked and emptied regularly, can prevent rainwater or condensate forming in the pressure measurement hoses.

The erection of air-supported halls starts with the construc­tion of the foundations and the assembly of the inflation
equipment. After the protected laying out of the membrane envelope, the assembly joints are made and the membrane is clamped to the edge of the lock and the rings of the blower connections. Once the perimeter edging has been fixed and any ballast bodies have been filled, the preparation works for inflation are complete. The pressure monitoring system is activated and the inflation can begin.

During inflation, the pressure must be controlled constantly using a pressure display. The inflation process for halls of average size is normally finished in less than 1 hour. The load­ing of the connections should be watched. Inflation cannot be done with a strong wind. Kinking of the envelope during inflation must be avoided. If it rains or snow falls during in­flation, there is a risk that puddles could form. If the material of the envelope stretches differently when inflated, then bi­axial tests should be performed, and the resulting measures for inflation are important.

The roof should periodically be inspected and checked in an optical control of the roof construction. The membrane should be checked for air-tightness, air pressure and mate­rial condition. The inlet filter in the switching cabinet and the drier filter should be cleaned or replaced. The pressure meas­urement hoses should be checked for condensate formation and blockage.1

Point loads can also be introduced into membrane surfaces by displacing or rotating the stiff primary elements. This ten­sioning process is frequently used for the erection of awnings, where the fabric is pretensioned by the tilting of the masts and the shortening of the stay and edge ropes (Fig. 356).

The fabric is fixed to the mast head by connecting the fittings and hanging the edge rope. In order to introduce the main pretension, the mast can be tilted through the use of tension­ing devices with hydraulic presses or using ratchet pullers, de­pending on the level of the forces to be introduced (Fig. 357). Care must be taken that the tilting is exactly in a plane, so that the eye lugs at the foot of the mast suffer no deformation. The masts need to be stabilised sideways during tilting.

The tensioning is done in the sequence laid down in the erection planning and depends on local conditions, cutting
pattern direction and fabric material. The duration of tension­ing and the sequence are especially critical when tension­ing glass fibre fabrics, where delayed tensioning can enable the stress to distribute in the surface. The force level can be balanced by alternatively tensioning and relaxing.

Because of the large deformations, the rope ends hung at the mast head represent a particularly sensitive area when tilting the masts. It is important that the ropes do not suffer kinking where they emerge from the fittings.

The arrangement of drillings in the foundation linkages for the stay ropes can make the tilting operation much easier (left in Fig. 357). Connecting members and temporary stays can also be fixed to these.

To finely adjust the masts, the stay ropes are shortened with turnbuckles. The fine tensioning of the edges is done by shortening the edge ropes.

Application of load by enlargement of shaft-shaped supporting elements

Structural elements in steel construction can be preten­sioned by processes, which lengthen or shorten the effec­tive axial length of structural elements through the installa­tion of mechanisms in the construction.1 This method can also be used to pretension very efficiently in membrane con­struction. High point constructions with standing or hang­ing columns can be pretensioned centrally by vertically jack­ing the columns and introducing forces into the membrane surface through the connected edges. Fig. 358 shows the displacements at column end points y1 and y2 during this tensioning process. They result from the geometrical stiff­ness, the curvature, the differing strain stiffnesses of ropes and fabric and the level of the forces introduced.

To be able to lengthen the effective axial length, the column has to be made in more than one part. The column can ei­ther be connected through an intermediate member with a c
an adjustable column foot or pushed through the column foot. After fixing the membrane to its edges at corners (1 in Fig. 358), the column is pushed apart in stages by hydraulic presses and the final length fixed with bolts (2 in Fig. 358). If no presses are available, columns hanging from a crane can be ballasted at their foot point. The position of the column can be fixed by retensioning the rope supporting it from underneath (3 in Fig. 358).

One disadvantage of this tensioning method is that the entire pretension force is applied at one point. This means that relatively high forces have to be used. Another way of pretensioning high point structures is to tension the mem­brane solely by peripheral pulling of the edges and corners (b, d in Fig. 359). This requires less force to be applied and a better stress distribution in the surface can be expected, but the amount of work involved in installing equipment for stretching along the edges and at the corners can cause high costs.

і

I

In order to avoid stress concentrations at the high point, the high point radius should be designed to be large for both tensioning methods. Otherwise it would be necessary to re­inforce the fabric in this area.

An example for the central introduction of pretensioning forces in a high point structure supported from underneath by wires is the roofing over the vehicle park at the Munich Waste Management Office (left, right in Fig. 360). The Glass/ PTFE fabric for the roofing was prefabricated at the works in 10-12 m widths and 70 m long strips with seams in the sur­face of 60 mm. The strips were laid out in the transverse di­rection on site and welded to the adjacent strips (see sec­tion 2.5.1.1). In order to be able to compensate for errors, the weld seams were made on site with a width of 150 mm. During the welding work, the hats of the high points were turned inwards and fixed down against wind loads, and in order to avoid puddle formation through rain or snow (right in Fig. 361).

Then the inverted hats were turned out upwards and the hanging columns were fitted from underneath (this meth­od of installation incidentally only works with appropriate­ly large radius and spacing of the high points). After lifting the hanging columns (right in Fig. 361), the ropes to support them from underneath were pulled in and anchored to the tree column. The staged pushing apart of the hanging col­umns by up to 300 mm introduced the pretension into the fabric and ropes. After adjusting the ropes, the columns were fixed in their final positions. For the duration of the erection process, the structure was stabilised by tirfors and ratchet tie­downs, which were remove after completion of tensioning.

An endless rope was fabricated into the membrane at the upper ring and secured with 3 pegs. This construction en­sures that when sideways movement occurs, no sharp edg­es can damage the fabric and the rope can roll of the ring under loading. There are no clamping plate connections be­tween the 10 x 12 m panels.

Flexible edge details with ropes or belts running in sleeves can be pretensioned by displacing the stiff corner fittings (B in Fig. 353) at points. This can be direct pulling or pushing of the fitting to the anchorage position. With awnings, the dis­placement is mostly produced by shortening the stay ropes, which causes a rotation of the columns.

Tensioning by direct pulling or pushing is performed in many stages. After the edge rope has been pulled in and the corner fittings preassembled, these are displaced towards the fixed primary construction element (1 in Fig. 353). After an altera-

tion of position by distance y1 or y2 and with the reaching of the anchorage position (P in Fig. 353), the corner fittings can be hung from the primary structure. In the second stage, the edge ropes are finely tensioned to the corner (2 in Fig. 353). The rope edge has then moved a distance y3 at its crown.

Corner constructions with elements like tensioning bolts or rollers can make the tensioning process easier. Heavy con­structions are mostly mounted on movable supports (right in Fig. 355).

In order to apply the load at a point into membrane surfaces, the geometry of the corner is important. The more acute the angle is at the membrane plate, the less is the available ma­terial capacity (see section 2.7).

High stresses in the fabric cannot be relaxed there through stretch and angular rotation on account of the short distance to the edge. Insensitive tensioning or imprecise patterning in the area of the membrane plate can therefore lead to extra­ordinary peaks of stress in acute corners under tensioning. The danger of the fabric tearing at the corner is mostly due to such peaks and not the size of the membrane fields to be tensioned. If the breaking tension is exceeded during ten­sioning, this would cause considerable damage to the fabric. It can be advisable to increase the material capacity through fabric reinforcements.

The introduction of linear edge loads into the membrane surface can also be achieved by changing the position of the stiff edge beam. One example of this is the tilting of arch­shaped edge beams. This is an efficient method of tension­ing, which is mostly used with many parallel rows of edge arches. The membrane surface is prestretched in the arch di­

rection and clamped to the arches (1 in Fig. 348). After pull­ing in and bolting the edge ropes, the arches can be pulled apart using rigging ropes or webbing slings with tirfors and fixed (2 in Fig. 348). The surface can be finely tensioned by shortening the edge rope (3 in Fig. 348). To tension a number of arch fields, each field is tensioned alone and secured with temporary construction.

Introduction of loads by moving the edge beams

Linear displacements in the membrane surface can also be achieved with arched structures by pushing up the edge arches.

This is done by first prestretching the membrane edge in the direction of the arch and then clamping it to the arch (1 in Fig. 351). After pulling in and bolting the edge rope, the arches can be jacked up in stages by hydraulic presses, which pretensions the membrane (2 in Fig. 351). The edges can subsequently be finely tensioned by shortening the edge ropes (3 in Fig. 351).

Whether this method can be used depends on the force necessary to introduce the load in relationship to the rise height, the form of the valley and the cutting pattern direction.

If high forces have to be used to push up the arches on ac­count of the strongly shortened membrane surfaces, then this method is uneconomical.[100]

If a row of arch fields are arranged parallel to each other, the membrane is mostly laid over the arches without fixing to the crown of the arch.