6 EXPERIMENT RESULTS AND INTERPRETATION
In chapter 5 the test canvases are discussed in terms of their yarn, weave, and wettability properties. Based on these properties the canvases’ cusping behaviour is predicted. This chapter is dedicated to the three experiments. The research objectives and experiment execution are discussed, and results are presented and interpreted. Detailed protocols of all test canvases can be found in appendix IX.
As a reminder: cusping is defined as fraction change in length. It is quantified by dividing the total height of a cusp (ltip-lbottom) by a reference length (loriginal, distance of
the measured thread to the canvas centre in an unstretched state). 1 Cusping is
expressed in percentage and quantified by the equation: ε cusp = Dl loriginal =
ltip- lbottom loriginal
6.1 EXPERIMENT I: LOW, MEDIUM AND HIGH PULLING
FORCE
In this experiment the Claessenens’ canvases (machine-wc) and the Die Leinweber canvases (hand-wc), all 40 cm x 40 cm, are stretched on a working frame with low, medium and high pulling forces. An average force of 7 N is used for the canvases stretched with low force, 19 N for the canvases stretched with medium force and 26 N for the one stretched with high pulling force. The stretched canvases are sized with a
medium strong rabbit-skin glue size (rabbit skin glue). For the size 10% rabbit skin glue solution is chosen, as the historical concentration is not known and may have varied depending on factors like glue quality, boiling time and ambient temperature2.
A 10 % rabbit skin concentration is a concentration used nowadays by many artists and conservators for sizing. The size is heated to 39° (±1°) and applied warmed. For each canvas, a new solution of rabbit skin glue size is prepared.
The objective of this experiment is to investigate three sub questions:
a) What is the impact of the amount of applied pulling forces on the severity of cusping in stretched canvases?
b) Will the test canvases, at a constant pulling force (equal in both directions) show equal cusping in warp and weft direction?
c) At what (minimum) force are differences in warp and weft cusping noticeable?
As discussed in chapter 4.4 it is expected that the weave anisotropy will play an important role in the cusping severity of the stretched canvases. When woven textiles are stretched, the crimp in the yarns is pulled out. During the period in which the material straightens out the stress strain curve is shallow, see the toe region in fig 25. The area in the stress strain curve in which fabrics decrimp, is called the toe region. When the fibres are straightened out entirely, the stress strain curve becomes very steep; the material is then very stiff. From this follows, that most cusping will already occur at low strain levels because little force is needed to initiate strain and thus thread deformation.
2 Stols-Witlox, Maartje. „Sizing layers for oil painting in western European sources (1500-1900): historical recipes and reconstructions“. In: Kroustallis, S., et al. (eds.) Art Technology. Sources and Methods. Proceedings oft he second symposium oft he Art Technological Source Research study group. London: Archetype, 2008: 147-165.
Results of experiment I: low, medium and high pulling forces
When comparing all six canvases (three hand-wcs and three machine-wcs) some general tendencies can be observed:
• The behaviour of the machine-wc is more anisotropic than that of the hand-wc
• Cusping is greater in the hand-wc
• The increase in cusping is not linear
The bar chart (fig 26) deals with the cusping severity of all six canvases of experiment I in the stretched state. Thereby, the chart is divided into three parts: the machine-wc and hand-wc, which were stretched with low, medium and high pulling forces.
It can be seen that, the higher the pulling forces during stretching, the stronger is the average cusping in all test canvases. The only exception is the hand-wc in warp direction, which shows 0,6% less average cusping when stretched with high force, compared to the canvas stretched with medium force (the final warp cusping value of the canvas stretched with medium force is 5.6%, the final warp value for the canvas stretched with high force is 5%). This might simply be a measurement fault or a consquence of irregularities in the hand-wc.
The force-cusping relationship is, as expected, not linear.3 The effect of decrimping
causes the highest amount of cusping, relative to the pulling force, to occur in those canvases stretched with low force and the least in those canvases stretched with high force.
The warp and weft systems of both canvases show anisotropy in their stress-strain response, but the machine-wc to a much higher extent. The machine-wc exhibits in all cases (at low, medium and high pulling force) significantly more cusping in the weft direction (more than twice as much). The amount of weft and warp cusping in the hand-wc is always similar. In general, the hand-wc is much
3 See glossary for a definition of a non-linear relationship.
Fig 26. Cusping of experiment I in stretched state.
stronger cusped in both weave directions.
These results are only partly in line with the predicted cusping behaviour. It had been assumed in chapter 4.4, that both canvases would cusp more in warp than the weft direction, because of the greater crimp and thus flexibility of the warp yarns. This is not at all true for the machine-wc and right for the hand-wc only at medium and high pulling forces. More factors than just yarn-flexibility consequently influence cusp formation. For the interpretation of these test results, the physicality's of yarn-bending and stress distribution are looked at. The yarn-bending of fibres and yarns is similar to the bending of beams.4 There are three forces involved in the bending of a
beam: a pulling force in the middle, and two points of resistance at the ends of the beam, which act in the opposite direction (fig 27).
In such a three point system, deformation occurs, which is called flexion. The interlacing points (crossover points of warp and weft threads) act, in a way, as points of resistance. The higher the density of interlacing points, i.e. weave density, the higher the counter force will be, which acts in the opposite direction of the pulling force. Comparative research carried out by Zupin and Dimitrovski confirms this hypothesis. They were able to show, that plain weave fabrics have the highest tensile force, compared to other weave patterns. This is because plain weave fabrics have the maximum number of interlacing points, as well as the highest friction between yarns. As the machine-wc is 17% denser in warp direction, the canvas experiences much greater resistance when pulled in weft direction, because more interlacing points are in the way (fig 28). The warp yarns are furthermore 25% thicker than the weft yarns, which increases the friction warp yarns experience during stretching even more.
4 Morton, W.E. and W.S.E. Hearle, Physical Properties of Textile Fibres. Butterworth & Co: Manchester, 1997, p. 399.
Fig 27. Principle of three point bending.
The assumption that, the hand-wc would generally cusp more than the machine-wc is confirmed. In addition to the greater crimp, those properties resulting from the weaving process also play a role here. It is intuitively obvious that, compared to hand looms, in machine looms warp and weft yarn tensions are much higher during weaving, resulting in flatter, stiffer fabrics. The machine-wc is indeed stiffer and has less drape than the lose hand-wc. Due to the higher inter-yarn tension, the friction in the interlacing points is greater in the hand-wc.
The bar chart below (fig 29) shows the cusping of the same test canvases as in chart fig. 26, now in the sized and dried state. In the machine-wc there is a significant difference in cusping between the stretched state and the sized and dried state, but not so in the hand-wc. The increase in cusping furthermore tends to be greater in those canvases that were stretched with lower force. Cusping increases in the machine-wc stretched with low pulling force by a factor 3.6, in the one stretched with medium force by a factor 2.9, and in the canvas stretched with high force by a factor 2.4. In the hand-wc, in contrast, cusping hardly changes upon sizing (factor 0.9 - 1). The impact of sizing is discussed in more detail in section 6.2.
Fig 28. This detail of the machine-woven canvas illustrates that over a length of 4mm, any chosen point (marked red) is hindered to move by 5 interlacing points in vertical direction and 6 points in horizontal direction.
To summarize, the sub-questions of this experiment can be answered with:
a) Most cusping occurs already at low stress levels, thus when little pulling force is applied. The cusping effect is inversely proportional to the pulling force.
b) In most cases warp and weft threads differ in their amount of cusping when stretched with constant pulling force, due to warp-weft anisotropy.
c) Differences in warp and weft cusping are already visible at low pulling forces.
6.2 EXPERIMENT II: WEAK, MEDIUM AND STRONG SIZE
In experiment II, the Claessenens’ canvas (machine-wc) and the Die Leinweber canvas (hand-wc) are stretched in the same manner as in experiment I. The variable
that is investigated in this experiment is the concentration of the rabbit skin glue solution used while sizing. Four stretched canvases (two machine-wcs and two hand-wcs) are sized with one layer of 5% and 20% solutions of warm rabbit skin glue, and compared to the machine-wc and hand-wc of experiment I, which are stretched with medium pulling force and sized with a 10% rabbit skin glue solution. The research question, which is hoped to be answered, is, whether the concentration of rabbit skin glue solution affects cusping severity?
Experiment I has already shown, that the higher crimp in the hand-wc causes more severe cusping. Materials with a high crimp ratio also have a high tension ratio on wetting. It seems, therefore, plausible, that the hand-wc should display a high increase in cusping upon sizing. As this expectation was not fulfilled in experiment I (cusping becomes on average three times greater in the machine-wc and hardly changes in the hand-wc), it becomes apparent that the weave properties of canvases have little influence on the cusping development during size application. As discussed earlier, there are differences in the water absorption properties between the two test canvases. The machine-wc absorbs water 20 times faster than the hand-wc, and water travels along the fibres of the machine-wc more than twice as fast as along those of the hand-wc. In both canvases water travels faster along the warp direction. Another aspect that is of interest in this experiment is drying shrinkage. It seems intuitively obvious that a highly concentrated size will create greater drying stresses. However, as rabbit skin glue dries by evaporation of the solvent (water), a greater proportion of water in the size also causes a greater amount of shrinkage. In the light of these two aspects (wickability properties of test canvases and drying shrinkage of rabbit skin glue), it is expected for this experiment, that the machine-wcs will be influenced most by sizing, and in particular when sized with low concentrations of rabbit skin glue.
Results of experiment II: Weak, medium and strong size
The bar chart below (fig 30) displays the cusping of three machine-wcs and three hand-wcs in the sized and dried state. A purple bar indicates the cusping of the canvases in the stretched state.
Taking the standard deviation into consideration, it can be seen that the strength of size does not influence cusping in the machine-wc in either weft or warp direction. Cusping in the three differently sized hand-wcs does not differ either, with one exception. That is the medium sized canvas, which is cusped significantly less in warp direction.
When looking at the increase in cusping within the same canvases, which is caused by the application of size, it can be observed that the increase in cusping in the machine-wc is always higher in the warp direction (cusping becomes on average 4,1 times greater), than in the weft direction (cusping becomes on average 2,3 times greater). As in experiment I, cusping in the hand-wc again hardly changes upon sizing.
Hintz * UvA * 2014
Fig 30. Cusping severity (cusping) of experiment II in sized and dried state.
During the execution of this experiment, it is observed that the machine-wc absorbed the size muchbetter than the hand-wc. This can be seen on the reverse of the canvases (fig 31), where even darkening in the machine-wcs indicates, that the size penetrates through the canvases evenly. Light patches on the reverse of the hand-wcs show that this was not the case in the latter, which is particularly visible in the canvas sized with highly concentrated rabbit skin glue.
This phenomenon can be explained with the better wettability properties of the machine-wc. The size is applied at a temperature of 38°-39° Celsius. While the liquid size is brushed out on the canvas, it cools down rapidly and becomes a gel. This limits the time frame in which the size can penetrate the canvas and thus affect severity in cusping.
In a magnified cross-sectional view it can be observed that weft yarns of both canvases have increased crimp after sizing (fig 32, 33). Zupin explains that, when a woven fabric is put under uniaxial stress, the crimp decreases in the stressed direction, and increases in the perpendicular direction. Crimp is thus interchanged.5 The friction
in the interlacing points of yarns hinders the movement to some extent. When the canvas is then sized, however, the water in the size unlocks the interlacing points for a moment and mobilizes thus both yarn systems.
Hintz * UvA * 2014
Fig 32. Cross sections of the machine-woven canvas in x230 mag. Each image is from a different location on the canvases.
Top: warp yarn in unstretched state (left) and stretched, sized and dried state. Bottom: weft yarn in unstretched state (left) and stretched, sized and dried state.
Stretched and sized weft (horizontal) Unsized and unstretched weft (horizontal)
Stretched and sized warp (horizontal) Unsized and unstretched warp (horizontal)
6.3
EXPERIMENT
III: POSITION
ON THE
CANVAS BOLT (CENTER PIECE, END OF
BOLT, SELVEDGE)
In this experiment both test canvases (the Claessenens’ canvases (machine-wc) and the Die Leinweber canvases (hand-wc)) are stretched in the standard manner. Canvas pieces of 40 cm x 40 cm are used, stretched with medium pulling force (19 N), and sized with medium strong rabbit skin glue solution (10%). The factor that is varied in this experiment is the position of the canvases on the bolt from which they are cut. In chapter 2.4 it has been described, how nineteenth century painting supports (machine-woven) were found to exhibit very deep cusping in warp threads, which results from the way warp threads were sometimes stretched to high tensions and bundled in the looms. No mention was made of likewise cusping in hand-woven canvases.
The objective in this experiment is to investigate, if the bolt position of a canvas influences cusping patterns. Sub questions, which will be looked at are:
a) Are canvases from the end of a bolt or from selvedge, already cusped as a consequence of the weaving process? And if so, do these thread deviations remain visible when these canvases are stretched?
Stretched and sized warp (horizontal) Unsized and unstretched warp (horizontal)
Fig 33. Cross section of the hand-woven canvas in x230 mag. Each image is from a different location on the canvases.
Top: warp yarn in unstretched state (left) and stretched, sized and dried state (right). Bottom: weft yarn in unstretched state (left) and stretched, sized and dried state (right).
Stretched and sized weft (horizontal) Unsized and unstretched weft (horizontal)
b) Is the severity of cusping, caused by stretching, influenced by the presence of a selvedge or the end of a bolt?
In connection with the bolt position, it is assumed that weave density is an influential factor for cusping severity.
Results of Experiment III: Position on the canvas bolt
Machine-woven canvas:
On both sides of the canvas, along the selvedge, the weft threads are deviated vertically (fig 34). This deviation extends about 17cm into the canvas. This weft deviation could either result from the weaving process, or from handling, for instance from the friction that occurs when the canvas bolt is rolled up.
The three images below show the machine-wc III.3, in which the selvedge is present at the left:
Fig 34. Selfedge of the machine-wc. The grid of individually marked threads makes the deviation of weft threads visible.
It can be seen, that the deviation is also clearly visible in the stretched state, but becomes less noticeable after sizing. This deviation makes the weft threads longer and therewith increases their elongation capacity. There is furthermore a difference in the warp weave density, which decreases towards the selvedge (16-16.5 threads per cm along the edge and 17 threads per cm in the centre of the canvas bolt). One would thus expect the left half of the canvas to cusp slightly more than the right half. That is indeed the case, as can be seen in image 36.
Fig 35. Binary imgages of canvas III.3 (machine-wc with the selvedge on the left), in unstretched (left), stretched (middle), and sized & dried state (right).
Fig 36. Canvas III.3 in stretched state (left) and sized state (right). The numbers around the edges represent the measured strain in each cusp (in percentage). In the center are the average cusping values of each canvas half.
Unfortunately, the ends of the bolt are not present in this roll of canvas. Therefore it is not possible to investigate the deviation of warp threads, which is likely to be present at the end of the bolt.
Hand-woven canvas:
The hand-woven canvas fans out along the selvedge and at the bolt ends. The corners are tipped and warp and weft threads are here deviated (fig. 37). The canvas is to both ends of the bolt 3 cm wider than in the middle.
In fig. 39 and 40 it can be seen, that because of the fanned out edges, the canvases take on non-rectangular shapes when stretched. Clear and unambiguous identification marks of the selvedge or bolt end are not recognizable in the thread movement.
Fig 38. Plain selvedge of the hand-wc.Fig 37. The hand-wc fans out toward s the end of the bolt. Near the bolt end
weft threads are deviated vertically.
Fig 39. Binary imgages of canvas III.1 (hand-wc with the selvedge on the left), in unstretched (left), stretched (middle), and sized & dried state (right).
The selvedge is created from the same warp and weft yarns as the fabric body, with the difference that the warp thread density is higher at the edges (18-19 threads per cm in contrast to the usual 16, fig 38). On the one hand restrains the higher warp thread density cusp formation along the selvedge, and, on the other hand, is cusp formation favoured by the lower weft yarn density, which causes the fanning out along the selvedge (weft density along the selvedge is 13.5 threads per cm and 13.7 threads per cm along the center of the bolt).
Cusping in the stretched canvases with selvedge and bolt end varies ambiguously, and no clear tendencies are recognisable.
6.4 DISCUSSION
Due to the complexity of the stretching experiments, many of the test results are
Fig 40. Binary imgages of canvas III.2 (hand-wc with the selvedge on the left and bolt end at the top), in unstretched (left), stretched (middle), and sized & dried state (right).
Fig 41. Schematic depiction of the fanned out selvedge. The lower weft thread density causes the fanning out effect.
ambiguous and could not be fully interpreted within this research project. In this chapter connections are made between those test results which are conclusive. Results are discussed and linked to prior research and issues which are relevant to the field of conservation and restoration.
Cusping formation during stretching and sizing depends very much on the inherent structural properties of a canvas. De-crimping of yarns causes already, at low stress levels, strong cusp formation in the hand-wc. It is thus to conclude that the more a canvas is crimped, the more pronounced is the cusping in a stretched state. The application of warm size, however, causes the machine-wc with lower crimp to cusp equally (at medium and high pulling load), and even more severely than the hand-wc (at low pulling load). The response to sizing depends exceedingly on the wettability properties of canvases. Cusping in the hand-wc, with a low absorption rate, changes little or not at all upon sizing. The different size concentrations, which are tested (5%, 10% and 20%), have no influence on cusping.
An interesting test outcome is the stronger cusping in weft direction in the machine-wc. Johnson’s observation that cusping is generally stronger in warp direction (due to the greater flexibility of warp yarns), is in these experiments only confirmed by the hand-wc stretched with medium and high pulling force. All other test canvases show stronger cusping in weft direction. The lower weft thread density explains greater cusping in weft direction. From this follows, that the weave density is more decisive for cusping anisotropy than crimp.
Differences in weft and warp yarn properties are furthermore relevant to consider when re-stretching or keying out paintings. It is relatively easy to falsely associate stress with force only. However, stress is force over area, which is determined by the density and thickness of the carrying thread: 000000
s = pullingForce
threadDensity´ threadDiameter0000 σσσ
Therefore, in order to re-stretch a painting at a constant tension in both weave directions, the yarn and weave properties need to be taken into consideration. In the case of the machine-wc weft yarns are thinner and less dense. Therefore, the weft yarns undergo greater stress than the warp yarns; the same amount of force is exerted on a lower area. Over the years, the canvas will become considerably weaker in weft
direction and might then also become vulnerable to the relatively low forces that conservators use when re-stretching or keying out paintings.
6.5 STUDY OBJECT REVIEWED
The possibility to interpret the cusping patterns found in the study object (the painting of St. Francis of Assisi) unambiguously remains limited. The question about the original format of the painting cannot be answered yet however assumptions can be made, based on the study of the painting’s x-ray, and findings of the experiments. The study of the x-ray reveals the weave characteristics of the paintings’ original canvas support. The characteristics of the yarn cannot be deduced from the x-ray. The thread density and thickness were measured with the ruler tool in Adobe Photoshop CS5. The canvas is woven in plain weave, like the test canvases, but its canvas fill is lower (88.3%, the total fill of the test canvas is 94%). The vertical and horizontal yarns are of different thickness and density. The vertical yarns are thicker (0.6mm), but less dense (13 threads per cm). The horizontal yarns are thinner (0.4mm) and
denser (14 threads per cm).
Density vertical threads 13 (threads/cm) Density horizontal threads 14 (threads/cm)
Vertical yarn thickness 0,6 mm
Horizontal yarn thickness 0,4 mm
Canvas fill (total) 88,3 %
Vertical canvas fill 71,12 %
Horizontal Canvas fill 59,34 %
Weave plain weave
Being a seventeenth century painting, the canvas of this painting must be hand-woven. Therefore, it can be better compared to the hand-woven test canvas. Several
features suggest that the warp-weft orientation in this painting is not, as usually the case, warp-vertical and weft- horizontal, but the other way around. When preparing paintings of larger scale, the artist would be more inclined to pay attention to the warp-weft orientation of his painting support. Stronger yarns are chosen for the warp because they experience greater stresses during weaving. Warp yarns are, therefore, also better capable of carrying the weight of large paintings, and are ideally orientated vertically in the painting. In small scale paintings the canvas has to carry a lower weight, which is why the weft-warp orientation is less important. With this in mind it appears unlikely that this painting is a fragment of a much larger composition. However, in the seventeenth century it was not uncommon to orientate also large paintings with the warp in horizontal direction.6
In hand-woven canvases warp threads are typically more evenly spun, more tightly twisted and thinner. The weft, often also referred to as fill, is thicker and contains coarser fibres. For this reason it is assumed that the vertical yarns in this painting are weft yarns (thicker: 0.6mm and less dense) and the horizontal yarns are warp yarns (thinner: 0.4mm and denser). In regard to weave density, this would also correspond to the hand-woven test canvas, which has 16 threads per cm in warp, and 14 threads per cm in weft direction.
If the painting were cut out of a much larger composition, one would expect no cusping to be present at all. However, in vertical direction yarns are deviated in small waves. These waves are too big to be weft snakes, but they resemble secondary cusping that Van de Wetering found in a Rembrandt painting, fig 42. The cusp in horizontal yarn direction, in the left bottom corner of the painting is believed to be primary cusping because of its comparably great width, fig 43.
6 The canvases of the seventeenth century, large scale paintings (7,5m x 7,5m and 2m x 3m) in the Oranjezaal, in Den Haag, for instance, are orientated with the warp running horizontally, along the width.
From private conversation with Lidwien Speleers, conservator and co-author of ’Jordaens and the Oranjezaal in Huis ten Bosch Palace. The paintings and the letters’, in: B.U. Münch en Z.Á. Pataki (red.), Jordaens. Genius of grand scale / Genie großen Formats, Stuttgart 2012, pp. 131-163.
Fig 42. Left: X-ray detail of Portrait of a Couple. Boston, Isabella Stewart Gardner Museum. Right: X-ray detail from above the ear of St. Francis.
Images do not have the same scale.
Fig. 43. Detail of the left bottom edge of the thread mapping, showing what is believed to be primary cusping.
As mentioned before, the author believes this painting to be intended as a small scale picture, because of the weave orientation. If that was indeed the case, then the painting has not been reduced in size, or only very little. It is assumed that the canvas was prepared with a ground layer in a larger format and the canvas fragment for this painting is cut out from somewhere further away from the larger canvas edge. The cusping in the left bottom area seems to be primary cusping and indicates that there, the canvas fragment is closest to the original edge.
To gain more clarity about this painting, stretching reconstructions could be carried out, that include re-stretching of canvases onto stretchers. Thereby the development of secondary cusping could be investigated. It would furthermore be interesting to compare the x-ray of this painting to x-rays of seventeenth century paintings of which the weft-warp orientation is certainly known.
7 CONCLUSION
This project aimed at shedding more light on the formation of cusping patterns in stretched canvases, by means of experimental testing. The validity of prevalent presumptions, concerning cusp formation, has been investigated and relativized. It was found that weave density and not crimp is the decisive factor that influences anisotropic cusping in the strongly orthotropic machine-woven test canvas. The effect of decrimping caused canvases to cusp already at low stress levels, in particular the strongly crimped hand-woven test canvas. Sizing increases cusping considerably in the highly absorbent machine-woven canvas, but not in the less absorbent hand-woven canvas. The size concentration was found to have no influence on cusping. The bolt position only influenced cusping in the machine-woven canvas. The method of visually recording strain development with photographs, proved to be effective. The reproducibility of the stretching experiments is restrained because both the stretching and the application of size is handwork and, therefore, includes variation, impossible to reproduce. Many of the test results cannot be interpreted with just one experiment, because of the complexity the experiments. Therefore, the test
results can only indicate tendencies. With more repetitions of the same experiments, canvas behaviour could be interpreted more meaningfully. Ideally, one would want to use spring scales, which can measure the pulling forces continuously over longer time periods.
Further research
It is recommended that further research be undertaken to investigate the following issues:
• Spacing between pulling points and their influence on the depth of cusping. • The influence of staggered/opposite positions of pulling points along the edges of stretched canvases on shearing stresses and cusp formation.
• The impact of the stretching order on cusping patterns. This concerns primary and secondary cusping. Here it is of interest to see whether the order at which a canvas is laced or tacked to a working- or stretcher frame influences the deformation of threads. With stretching experiments investig the the impact of the stretching order it might be possible to find an explanation for irregular cusping that fades out from one side to the other, as it can be observed in Vermeer’s The Art of Painting (fig 2, paragraph 1.2).
• Factors that influence secondary cusping. This implies differences in the mounting methods, for instance: use of nails/tacks, different tacking spacing, use of canvas pliers, folding canvas around the edge or fixing it at the front of a stretcher. • Structural fabric properties produced by contemporary weaving machines (air-jet, rapier and projectile) and finishing procedures (calendering).
The methodology with which cusping in stretched canvases is assessed could be improved by collaborating with the The Thread Count Automation Project. Within the TCAP project automated analysis has so far been conducted on radiographs.
According to R.G. Erdmann, however, it should be possible to produce automated angle maps directly from high resolution photographs of canvases.7
BIBLIOGRAPHY
Ackroyd, Paul. ‘The Structural Conservation of Canvas Paintings: Changes in Attitude and Practice since the Early 1970s’, Reviews in Conservation, no. 3, IIC, 2002, pp.3–14.
Bischoff, Gudrun. Das De Mayerne-Manuskript. Diploma thesis. Institut für Technologie der Malerei an der Staatl. Akademie der Bildenden Künste, Stuttgart, 2003.
Buckley, Spike. ‘The classification of craquelure patterns.’ Stoner, Joyce Hill, Rebecca Rushfield (ed.). Conservation of Easel Paintings. Oxon: Routledge, 2012, pp. 285-288.
Erdmann, Robert, e.a. Reuniting Poussin’s Bacchanals Painted for Cardinal Richelieu through Quantitative Canvas Weave Analysis. Presented 29 June 2013 at the AIC Annual Meeting, Indianapolis. 24 January 2014.
<http://erg.mse.arizona.edu/Erdmann_Reuniting_Poussins_Bacchanals.pdf >. Gaskell, Ivan and Michiel Jonker, Vermeer Studies. New Haven: Yale University Press, 1998.
Harley, Rosamond D., ‘Artists' prepared canvases from Winsor & Newton 1928-1951’. Studies in Conservation, vol. 32, no. 2, May 1987, pp. 77-85.
Price, Arthur and Allen C. Cohen, J.J. Pizzuto's fabric science. 6th edition. New York: Fairchild Publications, 1994.
Johnson, Don H., e.a. ‘Automated Thread Counting’. In: Vellekoop, Marije, e.a. Van Gogh at Work. München: Hirmer, 2013, pp. 142-155.
Johnson, Richard C., e.a. ‘Interpreting Canvas Weave Matches’. ArtMatters International Journal for Technical Art History, no. 5, 2013, pp. 53-61.
Johnson, Richard C., e.a. ‘Detecting Weft Snakes’. ArtMatters, no. 5, 2013, pp. 48-52. Lamers, Maranthe, Condition and treatment report St. Franciscus van Assisi, Antoon van Dyck. University of Amsterdam, 2014.
Liedtke, Walter, e.a. ‘Canvas Matches in Vermeer: A Case Study in the Computer Analysis of Fabric Supports’. Metropolitan Museum Journal, no 47, 2012, pp. 99-106.
Merrifield, Mary Philadelphia, Medieval and Renaissance Treatises on the Arts of Painting, Original Texts with English Translations, 2 vols. bound as one. New York: Dover, 1849, 1967, 1999.
Morton, W.E. and W.S.E. Hearle, Physical Properties of Textile Fibres. Butterworth & Co: Manchester, 1997.
Rouba, Bogumila J. ‘Die Leinwandstrukturanalyse und ihre Anwendung für die Gemäldekonservierung’. Restaurtorenblätter, no. 13, 1992, pp. 79-89.
Van de Wetering, Ernst. Rembrandt: The Painter at Work. Revised Edition. Berkeley & Los Angeles: University of California Press, 2009.
Van der Werf, Inez, ‘Het spanraam: een essentieel onderdeel van het schilderij’, Kunstenaars Materialen, no. 8, 1993, pp. 6-9.
Young, Christina. Measurement of the Biaxial Tensile Properties of Paintings on Canvas. Diss. thesis. University of London, 1996. The Courtauld Institute of Art. 21 November 2013. <http://www.courtauld.ac.uk/people/young-christina.shtml>. Young, Christina. Accelerated ageing of fabric supports: is it possible? Keynote Paper. AHRB Textile Conservation Centre Conference Post Prints, October 2005, Archetype Books, pp 111-116.
Young, C. ‘History of fabric supports’. Stoner, Joyce Hill, Rebecca Rushfield (ed.). Conservation of Easel Paintings. Oxon: Routledge, 2012, pp. 147-116.
Zenker, Evelyn, ‘Über Kett- und Schussfaden’, Zeitschrift für Kunsttechnologie und Konservierung, no. 2, 1998, pp. 338-350.
Zupin Ziva and Krste Dimitrovski, ‘Mechanical Properties from Cotton and
Biodegradable Yarns Bamboo, SPF, PLA in Weft’, Dubrovski, Polona Dobnik (ed.), Woven Fabric Engenering, INTECH, 2010, 3 June 2014. <http://bit.ly/1pAWURx>.
GLOSSARY OF TERMS
Anisotropy: the property of being directionally dependent, as opposed to isotropy
(uniformity in all orientations).
Auxiliary support: the framework/stretching frame over which canvas is stretched.
Usually a wooden stretcher or strainer.
Bar: in a stretching frame; a principal component/member that can be joined together
with three or more bars to a frame.
Binary image: a digital image that has only two possible values for each pixel, e.g. black and white.
Bolt: canvas bolt; the biggest possible roll of canvas, as it comes from the weaving
loom.
Canvas fill: density times thread thickness over unit area: noThreads´ threadDiameter ´ measuredLength
measuredLength´ m.Width ´100%
Connoisseurship: intuitive and delicate discrimination skill of the styles and
Creep: cold flow, slow movement resulting in permanent deformation of a solid
material.
Crimp: waviness.
Drape: textile property, which can be measured with a drape meter. It describes the
loseness of a textile and its ability to fall in folds.
Elasticity: capacity of a material to deform in a non-permanent way. Isotropy: uniformity in all orientations.
Keys: small wedges inserted into slots at the inner corners of a stretcher that are used
to expand the stretcher.
Lacing: method of stretching canvases into a working frame with cords.
Leno weave: the warp yarn twist back and forth in pairs around the weft threads. Nonlinear relationship: a type of relationship between two quantities in which
change in one quantity does not correspond with constant change in the other quantity.
Orthorectification: the process of correcting distortions in image geometry due to
non-vertical angles from the camera. The orthorectification process makes a positional correction of each point in an image.
Plain weave: or linen weave, (1/1) one up one down.
Selvedge: the edge of a canvas that is parallel to the warp direction; the lengthwise
edge of a fabric.
Stiffness: resistance to deformation.
Strainer: a rigid frame usually composed of wooden bars, which are joined in the
corner with non-expandable joins.
Stress: force over area; an internal force associated with a strain. Directly
proportional to strain.
Stretcher: a rigid frame usually composed of wooden bars, with expandable corners. Tacking margin: turnover edge; the section of canvas that extends onto the outer
edge and/or the reverse side of the auxiliary support, where the canvas is attached.
Theta θ: variable representing an angle of unknown degrees.
Threshold operation: (in Photoshop) the histogram of an image is split in a valley
between the black- and white- peaks.
Twist: denotes the rate of the rotation of fibres in a yarn. Yarn: one or more interlocked fibres.
Warp and Weft: the warp is the carrying thread and is stretched lengthwise in the
loom. The weft is the spooled filling thread and in a passed back and forth through the warp threads in a shuttle.
Weave: weave pattern or motif in a fabric, which is created by alternately raising and
lowering different threads in the warp.
Fig 41. Schematic depiction of weft and warp threads.
Weaver&Loom. Handmade Rug Boutique. 2014.
25 January 2014.
<http://www.weaverandloom.com/blog/the- weaving-guide/area-rugs-toronto-the-weaving-guide-on-the-pile/attachment/warp-and-weft>.
Wickability: The property of a fiber that allows moisture to move rapidly along the fiber surface and pass quickly through the fabric.
Wrap angle: The wrap angle (θ1 in fig 42) describes the contact area of one yarn, which surrounds another. The denser a fabric is woven, the greater is the warp angle of the yarns.
Fig 42. Schematic depiction of the geometry of a plain woven fabric.
G.A.V. Leaf, ‘The mechanics of plain
woven fabrics’, International Journal of Clothing
Science and Technology, 2004, vol. 16 Iss: 1/2,
pp. 97 – 107.
<http://www.emeraldinsight.com/journals.htm? articleid=875557>.
LIST OF IMAGES
Images are by the author, unless indicated differently.
Fig 1. Left: Magnified detail from an x-ray of Falling Leaves (F651) by Vincent van
Gogh. Right: Reversed grey scale image of the same detail.
Johnson, Don H., e.a. Advances in Computer-Assisted Canvas Examination: Thread Counting Algorithms, Presented 21May 2009 at the AIC Annual Meeting, Los Angeles. Revised June 18, 2009. 18 January 2014.
<http://people.ece.cornell.edu/johnson/aic-pap.pdf>.
Fig 2. Weft-thread angle map of The Art of Painting by Johannes Vermeer. There is
cusping at the top and bottom of the canvas and a horizontal weft snake about one-third from the top.
Liedtke, Walter, e.a. ‘Canvas Matches in Vermeer: A Case Study in the Computer Analysis of Fabric Supports’. Metropolitan Museum Journal, no 47, 2012, p. 101.
Fig 3. Combined warp- and weft-thread angle maps for Van Gogh’s People strolling in a park (F225).
Johnson, Richard C., e.a. ‘Interpreting Canvas Weave Matches’. ArtMatters International Journal for Technical Art History, no. 5, 2013, p. 56.
Fig 4. Two depictions of working frames, in which the strings are wraped around the
bars of the strainer.
Besides written sources and paintings with original stretching methods, such
depictions are the only sources that tell us about stretching method of that time. Left: A detail of A Painter in his Studio, by G. Dou. Right: A Painter in his Studio, etching by V. van der Vinne.
Van de Wetering, Ernst. Rembrandt: The Painter at Work. Revised Edition. Berkeley & Los Angeles: University of California Press, 2009, p. 117.
Fig 5. Detail of Assumption of the Virgin, by Wouter Crebeth II, 1628. Van de Wetering, Ernst. Rembrandt: The Painter at Work. Revised Edition. Berkeley & Los Angeles: University of California Press, 2009, p. 120.
Fig 6. Detail of the reverse side of Portrait of Dirck Hendricksz van Swieten,
unknown artist, Northern Netherlandish School, 1626. Amsterdam Rijksmuseum. The bottom edge of the canvas has been seamed. This is one of the rare cases that a painting was stretched directly into the picture frame instead of an auxiliary frame.
Van de Wetering, Ernst. Rembrandt: The Painter at Work. Revised Edition. Berkeley & Los Angeles: University of California Press, 2009, p. 122.
Fig 7. Schematic depiction of the threading of a loom.
Johnson, Van Gogh at Work: 149.
Fig 8. Illustration of a fabric with crimped warp and straight weft.
IFI CLAIMS Patent Services. Google Patents. ‘Warp crimp fabric. US 20060157138 A1.’ 2012. 16 June 2014.
<http://www.google.com/patents/US20060157138>.
Fig 9. Weft angle map of The Art of Painting by Johannes Vermeer. There is cusping
at the top and bottom of the canvas and a horizontal weft snake about one-third from the top.
Liedtke, Walter, e.a. ‘Canvas Matches in Vermeer: A Case Study in the Computer Analysis of Fabric Supports’. Metropolitan Museum Journal, no 47, 2012, p. 101.
Fig 10. Francis of Assisi, attributed to Anthony van Dyck, seventeenth century.
Museum Catharijneconvent Utrecht, s10076301. Photo courtesy of Maranthe Lamers, fellow student.
Fig 11. X-ray of the painting Francis of Assisi (left) and a mapping of the canvas
Fig. 12. Detail of the left bottom edge of the thread mapping. The horizontal threads
are pulled down towards the left side.
Fig. 13. Detail of the x-ray with errors indication where vertical threads change
direction.
Fig 14. Stretching sequence for the canvases on the working frame. Fig 15. Additional srews aid in optimizing the pulling force.
Fig 16. Camera set-up.
Fig 17. Location of the measuring locations, the zero point and the central reference
lines.
Fig 18. Example of two layered canvases.
Fig 19. Fringe type selvedge of the Claessens canvas. Fig 20. (left) Thick and thin warp threads.
Fig 21. (above) The upper yarn is a weft thread,
the lower yarn a wavy warp thread.
Fig 22. An abnormally thick warp thread of 1,3 mm diameter.
Fig 23. Warp thread fibres of the hand-wc (top) and the machine-wc (bottom). Fig. 24: Wickability test with warp strips.
Fig 25. A typical stress strain curve for crimped organic materials.
Korhonen, Rami K. and Simon Saarakkala.
’Biomechanics and Modeling of Skeletal Soft Tissues.’ Intech. 2011, p. 119. <http://cdn.intechopen.com/pdfs-wm/22189.pdf>.
Fig 26. Cusping of experiment I in stretched state. Fig 27. Principle of three point bending.
Anonymous. ‘Reactive transport and gel formation in two-phase systems and porous media’. 15 August 2007. 16 June 2014.
<http://www.phys.tue.nl/nfcmr/Restop25.html>.
Fig 28. This detail of the machine-woven canvas illustrates that over a length of 4mm,
any chosen point (marked red) is hindered to move by 5 interlacing points in vertical direction and 6 points in horizontal direction.
Fig 29. cusping of experiment I in sized and dried state.
Fig 30. The reverse of the machine-wcs in sized and dried state. Fig 31. The reverse of the hand-wcs in sized and dried state. Fig 32. Cross section of the machine-woven canvas in x230 mag.
Top: warp yarn in unstretched state (left) and stretched, sized and dried state. Bottom: weft yarn in unstretched state (left) and stretched, sized and dried state.
Fig 33. Cross section of the hand-woven canvas in x230 mag.
Top: warp yarn in unstretched state (left) and stretched, sized and dried state (right). Bottom: weft yarn in unstretched state (left) and stretched, sized and dried state (right).
Fig 34. Selfedge of the machine-wc. The grid of individually marked threads makes
the deviation of weft threads visible.
Fig 35. Binary imgages of canvas III.3 (machine-wc with the selvedge on the left), in
Fig 36. Canvas III.3 in stretched state (left) and sized state (right). The numbers
around the edges represent the measured strain in each cusp (in percentage). In the center are the average cusping values of each canvas half.
Fig 37. The hand-wc fans out toward s the end of the bolt. Near the bolt end weft
threads are deviated vertically.
Fig 38. Plain selvedge of the hand-wc.
Fig 39. Binary imgages of canvas III.1 (hand-wc with the selvedge on the left), in
unstretched (left), stretched (middle), and sized & dried state (right).
Fig 40. Binary imgages of canvas III.2 (hand-wc with the selvedge on the left and bolt
end at the top), in unstretched (left), stretched (middle), and sized & dried state (right).
Fig 41. Schematic depiction of the fanned out selvedge. The lower weft thread density
causes the fanning out effect.
Fig 42. Left: X-ray detail of Portrait of a Couple. Boston, Isabella Stewart Gardner
Museum.
Right: X-ray detail from above the ear of St. Francis. Images do not have the same scale.
Van de Wetering. Rembrandt.The painter at work: 114.
Fig. 43. Detail of the left bottom edge of the thread mapping, showing what is
APPENDIX I – Images of the study object
APPENDIX II – Protocol of the canvas analysis
Claessens Artists’ Canvas
• Warp thread thickness: avg 0,54mmMeasured in the centre of cross points. Each reading is of a different thread. Magnification x50.
• Warp twist angle: avg 21,4° Magnification x226.
• Warp density: 8.17 (thread per 0,5 cm) • Warp wrap angle: 25,6°
• Weft wrap angle: 16,3°
Measured in the centre of cross points. Each reading is of a different thread. Magnification x50.
• Weft twist angle: avg 12,4°
• Weft density: 7 (thread per 0,5 cm) • Canvas fill:
For the calculation of the canvas fill Rouba’s equation was used.8
fillweft/warp=
noThreads´ threadDiameter ´ measuredLength
measuredLength´ m.Width ´100%
The total fill is calculated with the following equation:
filltotal=
fill(weft) + fill(warp) - fill(weft) ´ fill(warp) 100
For the warp fill it is thus calculated: fillwarp= = 86.4% (91,8 if density is 17)
fillweft= 7´ 0.42mm´ 5mm 5mm´ 5mm ´100% = 58.8% filltotal= 58.8%+ 86.4% -58.8%´ 86.4% 100 = 94.4%
• Tensile elongation of indevidual threads- Single yarn stretch test
The elasticity of the weft and warp threads of the two test canvases is investigated by putting single weft and warp threads under a load of 500 grams. The threads are marked at two points and the increasing distance between these two points is measured. A number of 15 yarns was tested in that way and the average elongation during stretching, and the shrinkage after the romaval of the load is given below. a = un-stretched, b = stretched, c = relaxed after stretching
Warp thread un-stretched [mm] a stretched [mm] b elongation stretched (b-a) [mm] relaxed after stretching [mm] c Elongation relaxed (c-a) [mm] avg 100 107,7 7,7% 106,4 6,4%
• Single weft thread stretch test
Weft thread un-stretched [mm] stretched [mm] elongation (b-a) [mm] relaxed after stretching [mm] schrinkage (c-a) [mm]
avg 100 101 1% 100,3 0,3%
Protocol Die Leinweber canvas analysis
• Warp thread thickness: 0.54 mm
Measured in the centre of cross points. Each reading is of a different thread. Magnification x50.
• Warp twist angle ‘Z’: avg 18,8° Magnification x226.
• Warp density
Beginning and end of the bolt: 15,4 Middle of the bolt: 16,2
• Weft thread thickness: 0.49 mm
Measured in the centre of cross points. Each reading is of a different thread. Magnification x50.
• Weft twist angle: 14,2° Magnification x226-230. • Weft density: 13,7 Weft selvadge threads: 13 • Canvas fill:
Fillwarp= 16,2´ 0,5mm´10mm 10mm´10mm ´100% = 81% Fillweft = 13, 7´ 0,5mm´10mm 10mm´10mm ´100% = 68,5% Filltotal= (81%+ 68,5%) -81%´ 68,5% 100 =94%
• Single warp thread stretch test
Warp thread a) un-stretched [mm] b) stretched [mm] elongation (b-a) [mm] c) relaxed after stretching [mm] schrinkage (c-a) [mm] avg 94 109,3 16,3% 107,3 14,1%
• Warp wrap angle: 35,5° • Single weft thread stretch test
Weft thread un-stretched [mm] stretched [mm] elongation (b-a) [mm] relaxed after stretching [mm] schrinkage (c-a) [mm] avg 100 110 10% 108 8%
• Weft wrap angle: 33,7°
Wickability: Strips of the two linen fabrics (2 cm x 10 cm in dimension) are vertically dipped to a depth of 1 cm into water and the time needed to reach the wicking height of 2 cm is measured.
strips of 2cm width hand-woven Die Leinweber canvas
machine-woven Claessen’s canvas
warp direction 1min 45 sec 40 sec
weft direction 3min 1min 25 sec
The wettability of the test canvases is tested by means of a water-drop test, whereby single drops of water are placed on the surface of the canvases. The time from when a water drop is placed on the canvas, until its disappearance in the fabric is recorded (ten readings from different locations). The machine-wc absorbs a drop of water on average within one second, whereas water drops on the hand-wc are only fully absorbed after an average of 21 seconds.
Protocol St. Francis canvas analysis
• Warp density: 12,7 (thread per cm)• Weft density: 13,8 (thread per cm) • Warp thread thickness: 0.56 mm • Weft thread thickness: 0.43 mm
Fillweft/warp =
noThreads´ threadDiameter ´ measuredLength
Fillwarp= 12,7´ 0,56 ´10mm 10mm´10mm ´100%= 71,12 % Fillweft = 13,8´ 0,43´10mm 10mm´10mm ´100%= 59,34 % Filltotal =
fill(weft) + fill(warp) - fill(weft) ´ fill(warp) 100
= (71,12 % + 59,34 %) -
71,12%´ 59,34%
APPENDIX III - Experiment conditions, tools & materials
Experiment conditions
The basic experiment conditions are kept constant. That includes the room climate and the time frames (reading of warp and weft tensions and moment when photographs for weave maps are taken).
Tools & materials
• Hand-woven vanvas (Claessens via Van Beek Art Supply) • Machine-woven canvas (Die Leinweber)
• Hemp cord (to fasten canvases to the working frame) • Cotton yarn (for seaming the canvases)
• Tools needed for stretching (needles etc.) • Drawing lightbox to place the canvases on when marking threads.
• Permanent black marker
• Rabbit skin glue (Kremer)
• Heat plate • Cutting matt
• Thermometer (to control the temperature of the RSG) • Dino-Lite digital microscope
• Digital single-lens reflex camera (DSLR) camera • Tubular spring scales
• Wooden strainer (working frame) • Adobe Photoshop CS5
Documentation
The canvases where photographed after each step of preperation: unstretched, stretched, sized while wet, and after the size had dried. With adope photoshop these images where processed in order to quantify the severity of cusping.
Adobe Photoshop CS5:
• A square of graph paper was placed on every canvas. This graph paper was used to calibrate the scale of the ruler tool. In doing so one cm on the graph paper was marked with the ruler tool and in the option ‘analysis > set measurement scale > custom’ the marked amount od pixels was defined as one cm. A maximum deviation of 0,04 mm was accepted (1cm 0,96-10,04 cm).
• A cutting mat was layed underneath the canvases, so the rectangular lines on the cutting mat could serve as reference for the orthorectification. To do so the images were cropped in perspective.
APPENDIX IX – Experiment results in detail
Unless indicated differently the canvases were sized with 10% rabbit skin glue size (RSG). For the canvases of experiment II the temperature of the size was measured with thermometers. The temperature of the size is stated above the result table. The temperature of the water-bath, in which the size was warmed, is around 41-44° Celsius.
In the following tables the recorded pulling forces are listed for each pulling point. Two columns list the forces (in Kg and N) which were recorded by the spring scales during stretching, named ‘stretched Kg/N’. The two columns named ‘sized wet Kg/N’ list the values of the shrinkage forces, which occured during sizing and measured by the spring scales. The forces measured in the corners are listed seperately. The values that appear under ‘weft’ and ‘warp’ are the strain/cusping measurements. The unit for loriginal, ltip, and lbottom is mm. The strain ε is
calculated as percentage. The method of calculation is discussed in chapter 4.4.
0-SD stands for standard deviation. Avg stands for average. A slash appears in the tables where no values for the pulling force is
available.
Left: The position of pulling points. Canvases were always fastened to the working frame with the warp in vertical direction.
I.1 High pulling force machine-wc
49 grams of 10% RSG applieded at 40° Celsius.
I.1
Position Stretched Kg NStretched Sized wet Kg Sized wet N Corners Stretched Kg Stretched N Sized Kg Sized N
2,1 2,6 25,5 0,6 5,9 1,1 3,7 36,3 0,6 5,9 2,2 2,6 25,5 0,7 6,9 1,2 4,6 45,1 0,1 1,0 2,3 2,9 28,4 0,7 6,9 1,3 4,7 46,1 0,3 2,9 2,4 1,9 18,6 0,8 7,8 1,4 4,7 46,1 0,8 7,8 2,5 2,7 26,5 0,9 8,8 43,4 4,4 2,6 2,3 22,5 0,8 7,8 weft avg 2,5 24,5 0,8 7,4 3,1 2,9 28,4 0,4 3,9 3,2 2,6 25,5 0,8 7,8 3,3 2,9 28,4 0,2 2,0 3,4 2,8 27,4 0,6 5,9 3,5 / / / / 3,6 2,7 26,5 0,4 3,9 warp avg 2,8 27,2 0,5 4,7 total avg 2,6 25,7 0,6 6,1 SD 0,3 2,6 0,2 2,0 Weft l_orig. Stretched l_bot. Stretched l_tip Stretched ε cusp Stretched l_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried top1 193,80 197 202,1 0,026 194,9 191,5 202 0,054 top2 193,80 196,5 201 0,023 195,9 192,2 202,1 0,051 bottom 1 190,70 193,1 197,5 0,023 192,4 190,6 198,6 0,042 bottom 2 191,1 193,8 198 0,022 191,5 190,3 198,3 0,042 weft avg 0,024 0,047 SD 0,002 0,005 vertical
strain l_tip-l_bot cusped middle line
ε strain
Stretched l_tip-l_bot cusped middle line ε strain Dried top1 5,1 199,6 0,030 10,5 196,8 0,009 top2 4,5 198,8 0,026 9,9 197,2 0,006 bottom 1 4,4 195,3 0,024 8 194,6 0,011 bottom 2 4,2 195,9 0,025 8 194,3 0,015 weft avg 0,026 0,010 SD 0,002 0,003 Warp l_orig. Stretched l_bot. Stretched l_tip Stretched ε cusp Stretched l_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried left 1 191,60 193,4 195,9 0,013 193,2 188,2 195,4 0,037 left 2 192,30 194 196,4 0,012 192,5 188,7 195,8 0,037 right 1 192,80 194,6 196,2 0,008 193 188,7 196,1 0,038 right 2 191,9 193,7 195,7 0,010 193,7 188,7 196,6 0,041 warp avg 0,011 0,038 SD 0,002 0,002 horizontal
strain l_tip-l_bot cusped middle line
ε strain Stretched l_tip-l_bot cusped middle line ε strain Dried left 1 2,5 194,7 0,016 7,2 191,8 -0,007 left 2 2,4 195,2 0,015 7,1 192,3 -0,001 right 1 1,6 195,4 0,013 7,4 192,4 -0,003 right 2 2 194,7 0,015 7,9 192,7 -0,005 warp avg 0,015 -0,004 SD 0,001 0,002
I.2 Medium pulling force machine-wc I.2 Position Stretche d Kg Stretche d N Sized wet Kg Sized wet N Corners Stretch ed Kg Stretched N Sized Kg Sized N 2,1 1,9 18,6 0,6 5,9 1,1 1,85 18,1 1,1 10,8 2,2 2,0 19,2 0,7 6,9 1,2 1,87 18,3 0,9 8,8 2,3 1,9 19,0 0,5 4,9 1,3 1,96 19,2 1 9,8 2,4 1,8 17,8 0,5 4,9 1,4 1,8 17,6 1,2 11,8 2,5 1,7 17,1 0,6 5,9 1,9 18,3 1,05 10,3 2,6 1,8 17,4 0,7 6,9 horiz. avg 1,9 18,2 0,6 5,9 3,1 1,9 18,3 1 9,8 3,2 / / / / 3,3 1,9 18,5 0,9 8,8 3,4 1,9 18,5 0,9 8,8 3,5 1,8 17,3 0,8 7,8 3,6 1,9 18,7 1 9,8 vertical avg 1,9 18,3 0,9 9,0 total avg 1,9 18,2 0,7 7,3 SD 0,1 0,6 0,2 1,8 Weft l_orig. Stretched l_bot. Stretched l_tip Stretched ε cusp Stretched l_orig. Dried l_bot. Dried
l_tip Dried ε cusp Dried top1 191,60 194,3 198,2 0,020 187,90 187,2 197,2 0,053 top2 191,80 195,1 198 0,015 188,3 188,3 198,2 0,053 bottom 1 190,80 191,9 196,2 0,023 188,7 186,3 196,3 0,053 bottom 2 191,2 191,2 196 0,025 189 185,9 195,8 0,052 weft avg 0,021 0,053 SD 0,004 0,000 vertical
strain l_tip-l_bot cusped middle line
ε strain Stretched l_tip-l_bot cusped middle line ε strain Dried top1 3,9 196,25 0,024 10 192,2 0,023 top2 2,9 196,55 0,025 9,9 193,25 0,026 bottom 1 4,3 194,05 0,017 10 191,3 0,014 bottom 2 4,8 193,6 0,013 9,9 190,85 0,010 weft avg 0,020 0,018 SD 0,005 0,007 Warp l_orig.
Stretched l_bot. Stretched l_tip Stretched ε cusp Stretched l_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried
left 1 191,2 193,5 195,2 0,009 188 187,2 193,9 0,036 left 2 191,9 193,8 195,6 0,009 188,8 187,4 194 0,035 right 1 192,4 193,9 195,7 0,009 189,5 188,6 194,1 0,029 right 2 191,7 193 194,5 0,008 188,9 188,9 195,6 0,035 warp avg 0,009 0,034 SD 0,001 0,003 horizontal strain l_tip-l_bot cusped middle line ε strain Stretched l_tip-l_bot cusped middle line ε strain Dried left 1 1,7 194,4 0,016 6,7 190,6 0,014 left 2 1,8 194,7 0,015 6,6 190,7 0,010 right 1 1,8 194,8 0,012 5,5 191,4 0,010 right 2 1,5 193,8 0,011 6,7 192,3 0,018 warp avg 0,014 0,013 SD 0,002 0,003
I.3 Low pulling force machine-wc
I.3
Position Stretched Kg Stretched N Sized wet Kg NSized wet Corners Stretched Kg Stretched N Sized Kg Sized N
2,1 0,7 6,86 0,1 1,0 1,1 / / 0,1 0,98 2,2 0,8 7,84 0,1 1,0 1,2 1,2 11,76 0,4 3,92 2,3 / / / / 1,3 1,2 11,76 0,2 1,96 2,4 0,7 6,86 0,1 1,0 1,4 1,2 11,76 0,1 0,98 2,5 0,6 5,88 0,1 1,0 2,6 / / / / weft avg 0,7 6,9 0,1 1,0 3,1 0,6 5,9 / / 3,2 0,7 6,9 0,2 2,0 3,3 0,9 8,8 / / 3,4 0,7 6,9 1,2 11,8 3,5 0,6 5,9 / / 3,6 0,5 4,9 2,2 21,6 warp avg 0,7 6,5 1,2 11,8 total avg 0,7 6,7 0,6 5,6 SD 0,1 1,1 0,8 7,7 Weft l_orig.
Stretched l_bot Stretched l_tip Stretched ε cusp Stretched l_orig. Dried Driedl_bot. l_tip Dried ε cusp Dried
top1 186,30 187,3 189,4 0,011 186 182,6 190,1 0,040 top2 186,30 187,3 190,2 0,016 186,3 182,4 191,5 0,049 bottom 1 198,00 199,1 201,6 0,013 199,3 193,5 202,5 0,045 bottom 2 197,20 198,7 201,3 0,013 198,8 195,9 204,0 0,041 weft avg 0,013 0,044 SD 0,002 0,004 Warp strain l_tip-l_bot cusped middle line ε strain Stretched l_tip-l_bot cusped middle line ε strain Dried top1 2,1 188,35 0,011 7,5 186,35 0,002 top2 2,9 188,75 0,013 9,1 186,95 0,003 bottom 1 2,5 200,35 0,012 9 198 -0,007 bottom 2 2,6 200 0,014 8,1 199,95 0,006 weft avg 0,013 0,001 SD 0,001 0,005 Warp l_orig. Stretched l_bot. Stretched l_tip Stretched ε cusp Stretched l_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried left 1 171,6 172,5 173,5 0,006 172,1 168,2 172,7 0,026 left 2 170,9 170,9 172,2 0,008 171,9 168,4 173 0,027 right 1 211,1 211,7 213,3 0,008 211,3 208 212,8 0,023 right 2 211,9 212,9 213,5 0,003 212,1 207,1 212,1 0,024 warp avg 0,006 0,025 SD 0,002 0,002 Weft strain l_tip-l_bot cusped middle line ε strain Stretched l_tip-l_bot cusped middle line ε strain Dried left 1 1 173 0,008 4,5 170,45 -0,010 left 2 1,3 171,55 0,004 4,6 170,7 -0,007 right 1 1,6 212,5 0,007 4,8 210,4 -0,004 right 2 0,6 213,2 0,006 5 209,6 -0,012 warp avg 0,006 -0,008 SD 0,002 0,0028
I.4 High pulling forces hand-wc
58g of 10% RSG applied at 39° Celsius.
I.4 Position Stretched
Kg Stretched N Sized wet Kg Sized wet N Corners Stretched Kg Stretched N Sized Kg Sized N
2,1 2,5 24,5 1,6 15,7 1,1 4,5 44,1 2,4 23,52 2,2 2,5 24,5 1,8 17,6 1,2 4,3 42,14 2,5 24,5 2,3 2,6 25,5 1,8 17,6 1,3 4,2 41,16 2,5 24,5 2,4 2,5 24,5 1,8 17,6 1,4 4,6 45,08 2,6 25,48 2,5 2,4 23,5 1,6 15,7 4,4 2,6 2,3 22,5 1,6 15,7 weft avg 2,5 24,2 1,7 16,7 3,1 2,7 26,5 1,4 13,7 3,2 2,7 26,5 1,5 14,7 3,3 2,6 25,5 1,4 13,7 3,4 2,7 26,5 1,3 12,7 3,5 / / / / 3,6 3 29,4 1,5 14,7 warp avg 2,8 27,0 1,4 13,9 total avg 2,6 25,4 1,6 15,4 SD 0,2 1,7 0,2 1,7 Weft l_orig.
Stretched l_bot. Stretched l_tip Stretched ε cusp Stretched l_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried
top1 193,0 202,8 211,4 0,045 192,5 201,5 210,8 0,048 top2 193,0 202,1 212,3 0,053 192,7 200,6 209,9 0,048 bottom 1 189,2 201,2 209,4 0,043 187,5 199,2 207,9 0,046 bottom 2 190 201,7 211,1 0,049 187,9 200,3 209,7 0,050 weft avg 0,048 0,048 SD 0,004 0,001
vertical strain l_tip-l_bot cusped middle line ε strain Stretched l_tip-l_bot cusped middle line ε strain Dried top1 8,6 207,1 0,073 9,3 206,2 0,071 top2 10,2 207,2 0,074 9,3 205,3 0,065 bottom 1 8,2 205,3 0,085 8,7 203,6 0,086 bottom 2 9,4 206,4 0,086 9,4 205,0 0,091 weft avg 0,080 0,078 SD 0,006 0,011 Warp l_orig.
Stretched l_bot. Stretched l_tip Stretched ε cusp Stretched l_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried
left 1 188,5 199,4 208,7 0,049 187,8 197,7 207,2 0,051 left 2 189,90 200,8 209,9 0,048 188,9 199,6 209,1 0,050 right 1 191,30 203,8 214 0,053 190,4 202,7 212,1 0,049 right 2 190,1 200,9 210,6 0,051 189 198,8 208,5 0,051 warp avg 0,050 0,050 SD 0,002 0,001 horizontal
strain l_tip-l_bot cusped middle line
ε strain
Stretched l_tip-l_bot cusped middle line
ε strain Dried left 1 9,3 204,1 0,082 9,5 202,5 0,078 left 2 9,1 205,4 0,081 9,5 204,4 0,082 right 1 10,2 208,9 0,092 9,4 207,4 0,089 right 2 9,7 205,8 0,082 9,7 203,7 0,078 warp avg 0,085 0,082 SD 0,004 0,005
I.5 Medium pulling force hand-wc
I.5
Position Stretched Kg Stretched N Sized wet Kg wet NSized Corners Stretched Kg Stretched N Sized Kg Sized N
2,1 1,88 18,4 1,5 14,7 1,1 2,0 19,4 / / 2,2 / / / / 1,2 1,9 18,9 2,19 21,5 2,3 1,8 17,6 1,5 15,0 1,3 2,0 19,2 1,27 12,4 2,4 2,1 20,3 1,8 17,3 1,4 2,0 19,5 1,43 14,0 2,5 1,9 18,3 1,5 15,1 2,0 2,6 1,9 18,2 1,8 17,5 weft avg 1,9 18,6 1,6 15,9 3,1 1,8 17,6 1,2 11,7 3,2 1,9 18,3 1,2 11,6 3,3 1,8 17,5 1,1 11,2 3,4 2,1 20,6 / / 3,5 2,1 20,9 1,4 13,4 3,6 1,9 18,3 1,9 18,6 warp avg 1,9 18,9 1,4 13,3 total avg 1,9 18,7 1,5 14,6 SD 0,1 1,12 0,3 2,5 Weft l_orig.
Stretched l_bot. Stretched l_tip Stretche ε cusp Stretchel_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried
top1 191,00 201,6 209,3 0,040 193 189 198,9 0,051 top2 193,20 203,7 211,4 0,040 195,6 190,8 200 0,047 bottom 1 183,80 193,5 203 0,052 185 186,4 196,5 0,055 bottom 2 187,7 193 201,9 0,047 189,2 186 196,3 0,054 weft avg 0,045 0,052 SD 0,005 0,003 vertical
strain l_tip-l_bot cusped middle line
ε strain Stretched l_tip-l_bot cusped middle line ε strain Dried top1 7,7 205,5 0,076 9,9 193,95 0,005 top2 7,7 207,6 0,074 9,2 195,4 -0,001 bottom 1 9,5 198,3 0,079 10,1 191,45 0,035 bottom 2 8,9 197,5 0,052 10,3 191,15 0,010 weft avg 0,070 0,012 SD 0,011 0,014 Warp l_orig.
Stretched l_bot. Stretched l_tip Stretche ε cusp Stretchel_orig. Dried l_bot. Dried l_tip Dried ε cusp Dried
left 1 191,40 201,3 212,2 0,057 192,8 187,4 193,7 0,033 left 2 193,00 203,7 214,8 0,058 194,5 187,6 193,6 0,031 right 1 190,40 203,8 214,1 0,054 192,8 188,4 194,4 0,031 right 2 194 203 214,1 0,057 196,3 188,6 195,1 0,033 warp avg 0,056 0,032 SD 0,001 0,001 horizontal
strain l_tip-l_bot cusped middle line
ε strain
Stretched l_tip-l_bot cusped middle line
ε strain Dried left 1 10,9 206,8 0,080 6,3 190,6 -0,012 left 2 11,1 209,3 0,084 6 190,6 -0,020 right 1 10,3 209,0 0,097 6 191,4 -0,007 right 2 11,1 208,6 0,075 6,5 191,9 -0,023 warp avg 0,084 -0,015 SD 0,008 0,006
