The determination of the bending stiffness of fibres so far has been cumbersome and complicated. However, a new method is in the final stage of development at Textechno to counter act this.
ITEMAT+ is a new interlace tester for filament yarn using a mechanical sensor. The function of this unit for flat and textured yarn will be explained by means of test results in this paper. The universal filament yarn tester DYNAFIL ME can now optionally be equipped with an additional godet and a new testing method is available in order to more precisely determine crimp and shrinkage of textured and BCF-yarns.
Determination of single-fibre bending stiffness by axial compression
The processing properties of fibres are largely determined by their flexural behaviour. Stiff fibres are more difficult to use for fine yarns because of their high reaction torque. Also the characteristics of the manufactured products are strongly influenced by the stiffness of the single fibres.
In the apparel industry the bending stiffness of fibres is an important factor influencing the wearing comfort. Too high flexural stiffness of the fibre may cause mechanical irritation of the skin. Furthermore, the bending stiffness affects the crease resistance and the overall feel of the textile product.
However, the bending stiffness of a fibre is typically very low and very complex to determine. Several possibilities to determine the bending stiffness have been used in the past, but they are very time-consuming and are not established on the market.
The new approach by Textechno is to insert the fibre into the vertical testing zone of a FAVIMAT+ as in a standard tensile test. This method can be automated and is therefore applicable to a large number of samples. The fibre is then clamped at both ends and axially compressed. Due to the stress the fibre is deflected and bent aside. The axial reaction force and the deflection are then measured and recorded continuously during this process. From the force-deflection diagram, data is taken in order to calculate the physically correct bending stiffness.
For this evaluation the mechanical behaviour of the fibre must be understood. The bending stiffness B is defined as the product of the Young's modulus E and the geometrical moment of inertia I.
The elastic modulus is a pure material dependent variable. The moment of inertia depends on the geometrical shape of the fibre cross section. For non-circular fibre cross-sections the area moment of inertia depends on the load direction of the bending force. For an oval fibre, the area moment of inertia is minimal in the direction of its minor axis. Given bending loads, the fibres rotate in the direction of the smallest moment of inertia. Therefore, the effective bending stiffness of a non-circular fibre is determined by its minimum moment of inertia [Reu00].
For the calculation of the bending modulus the mathematical analysis of buckling is used. According to Euler four cases of elastic 
buckling exist (Picture 1). The four cases are characterised by different boundary conditions. The boundary type determines a clamping factor for the free length L. The effective length L’ may be larger or smaller than the free length L. The technique
Of importance in this work is particularly the 4th Euler buckling mode. In this case the bar is clamped on both sides and can be displaced longitudinally in the axial direction. The boundary conditions of the fourth Euler's case correspond to the boundary conditions of the fibre bending in the FAVIMAT +.
In theory the bar is assumed to be straight and the load is perfectly centred. The cross section of the bar is constant over its length. Upon reaching a critical load in the axial direction the bar bends laterally. The critical value of the load is called buckling load Pcr. It can be calculated using the formula of Euler (2).
The formula is only applicable to bars or fibres with elastic deformation behaviour. For the determination of the buckling load with plastic deformation components other calculation methods must be used. These include the "tangent-modulus theory" and "reduced-modulus theory", two calculation methods that allow an estimation of the true buckling load [Tim85].
In the following example equation (3) is used to determine the bending stiffness of five different glass fibre specimens. The fibres were clamped in the FAVIMAT+ with gauge lengths Lg of 6 and 10 mm and compressed to 50% of the original gauge lengths (to 3 mm and 5 mm). Picture 2 shows the force-deflection diagrams of two example measurements. It is noticeable that no distinct maximum force occurs at the beginning of the deformation. This is due to imperfections such as slight pre-deformations of the fibres and a small eccentricity of the axial load. The exact buckling load Pcr, at which the lateral buckling first appears, can therefore not directly be determined. In studies of Shioya et al. on the bending behaviour of carbon fibres under axial load, this effect was observed as well [Shi99]. The authors propose a linear extrapolation of the curve into the zero point of the displacement (red lines in Picture 2) as a correction method. By this way a good approximation of the real buckling load can be determined. The results for all samples and trials are shown in table 1.
In order to validate the measured bending stiffness, the theoretical bending stiffness is calculated from the fibre diameter and the modulus of elasticity. For this purpose, equation (4) is used. As a simplification it is assumed that the fibres have a completely perfect circular cross-section, therefore the formula for circular cross-sections is used to calculate the moment of inertia of the fibre. The mean modulus of elasticity of the fibres is 75,75 GPa. It was determined by a tensile test of the same material on a FAVIMAT +. The real fibre diameter was determined by microscopic measurements of the top view of the fibres.
The results are summarised in Table 3. It is apparent that the measurement method can determine plausible values of the bending stiffness. The fibre diameter is the largest factor in the calculation of the bending stiffness, as it is included in the fourth power in the formula.
Small errors of the fibre circularity, diameter changes along the fibre axis and surface defects reduce the real bending stiffness. In reality the theoretical bending stiffness and resulting buckling load can typically not be achieved. Another factor that may affect the results is the clamping, because it can lead to a small change in fibre form and actual gauge length.
These effects decrease if a large clamping length can be used. For most applications, the actual bending stiffness is more relevant than the theoretical bending stiffness. The process is therefore very suitable to compare the stiffness of different fibres with each other. Also the method can be used to determine defects of the fibre cross-section, as a non-circular cross-section leads to significantly lower bending stiffness. The method has also been successfully applied to tapes and fabrics.
ITEMAT+ a new feeler-gauge interlace tester
The quantity and quality of interlaces (entanglement) in filament yarn is of high importance for the processing properties of the yarn as well as for the appearance of the fabric produced with these yarns.
Over the years three function principles of interlace testers have been established: The needle method, as e.g. realised in Lenzing Instruments' RAPID 500, the optical method, as e.g. on Textechno's DYNAFIL ME or on Lenzing Instruments' PROMPT on-line-monitoring systems, and the feeler-gauge method as in the ITEMAT, developed by Enka Tecnica in the 1980s. While the needle method is slow (about 15 m/min yarn speed) it has been applied to flat yarns, the optical method in laboratory applications is recommended for texturized or BCF yarns on which the interlaces are visible to the human eye. The testing speed for optical testing can be as high as several hundred or thousands of metres per minute.
The feeler gauge method, however, is well suited for both yarn types and can be run at 100m/min for most applications. Textechno acquired the rights of the ITEMAT from Heberlein in 2007, after Heberlein had taken over Enka Tecnica.
The feeler gauge in principle measures the thickness of the yarn under a certain pressure (Picture 3). At places where all filaments are
free, they are pushed flat to lie in a few layers. At a good interlace point the detected thickness of the yarn thus is higher. A very similar feeler-gauge system builds the basis of TEXTECHNO's new ITEMAT+, which is a complete re-design with respect to the Enka Tecnica ITEMAT. ITEMAT+ is now available as a single-position or a multi-position unit with up to five positions. It is designed to accommodate either the feeler-gauge sensor or an optical sensor, but so far it is available with the mechanical feeler-gauge sensor. Advantages of the new feeler heads over the former version are a computer-selectable contact pressure and an in-built lift-off mechanism, which lifts the feeler off from the yarn for cleaning purposes or in case the sensor should not increase yarn tension for other applications. Another advantage of the ITEMAT+ over the ITEMAT is the fact that the new machine is equipped with a constant-tension feeder, which applies a programmable and constant tension to the yarn for better reproducibility of the test results.
Two versions are available: ITEMAT+ and ITEMAT+ TSI (Test for the Stability of Interlaces), the latter being equipped with a drawing-zone to stress the yarn prior to the interlace test.
The signal from the sensor is displayed and stored as micrometres of yarn thickness. A resolution and reproducibility of about one micrometre is required for sensitive and reproducible measurements.
When counting interlaces in a water bath, which is a manual 'reference' method, the user has to decide for each interlace knot, if this knot should be counted or not. That this is not always an easy decision can be seen from the following picture:
While this decision is very subjective, a fixed and reproducible threshold is set on the ITEMAT+. All interlaces exceeding a certain thickness will be counted. A so-called level curve helps to assess the best setting for this level threshold, since it displays the number of interlaces counted per metre as a function of the level. If all interlaces are to be counted, a low level setting is recommended.
Level curves from well and from badly entangled packages can be compared in order to find the level threshold which gives the best differentiation.
The steeper the decline of the level curve, the more regular are the interlaces. Insofar, the level curve can be more than a tool for selecting the level.
A great feature of the software is the fact that the level threshold can still be modified after testing. This certainly gives an advantage over the needle testing in which the threshold for the quality of an interlace knot is defined by the yarn tension at which the needle is shifted away. This setting obviously cannot be modified after testing, but the test has to be repeated with a different setting.
In the following, test results on Polyester FDY are presented. In a study the effect of spin finish content in percent by mass on the formation of interlaces was analysed. The results show that without spin finish nearly no interlaces are generated. Up to a spin finish level of 1 % the number of interlaces is increasing continuously, although at 0.6 % a good level of interlacing has already been reached.
A yarn with no interlaces, e.g. because of missing spin finish, cannot be tested for high lengths unless the yarn is pulled off from the package tangentially. The reason is that the twist which is generated by pulling off the yarn over head of the package adds up in front of the feeler gauge and passes through when a certain amount of twist is reached. This generates a signal similar to the one generated by an interlace point. In the case of a yarn with interlace points the twist does not strongly influence the signal, apparently it 'slips' through the feeler gauge at the interlace points.
Another example shows the behaviour of the interlaces in a texturized yarn after being drawn in the ITEMAT+ with increasing draw-ratio:
ITEMAT+ thus is a versatile and useful tool for interlace testing on flat yarn as well as on texturized yarn. The multi-position unit with up to five positions offers the same properties and is a very efficient tool for production control.
New possibilities for crimp and shrinkage testing
Measuring the crimp and shrinkage properties of filament yarns can be performed in both a static or a dynamic test. The static test is standardised according to EN 14621 and can be automated by means of Textechno's TEXTURMAT ME+.
Dynamic tests, i.e. tests on the running yarn, are possible with the DYNAFIL ME, Textechno's universal filament yarn tester. On a DYNAFIL ME (Picture 6) the yarn runs through a heater which is located between two godets. 
The yarn can be subjected to a certain draw-ratio or overfeed to measure the related tensile force - or be run under constant tension in order to measure its elongation or contraction (as e.g. shrinkage or crimp).
With its origin in the 1970s, the first DYNAFIL model was developed in order to characterise texturized yarns with respect to their crimp and shrinkage properties, helping machine producers and their customers to optimise the texturizing process.
As a starting point to test unknown yarns, a so-called speed-curve is established, which shows the contractive force of the yarn at constant overfeed as a function of yarn speed. Since the temperature which the yarn reaches when passing through the heater is inversely related to the yarn speed, the speed curve can be read as an inverse temperature curve.
In the case of polyester the optimum crimp development takes places at a yarn temperature of about 100°C, while shrinkage in general occurs at much higher temperatures (lower yarn speed) only. This fact allows to separate the crimp 'signal' and the shrinkage 'signal' by measuring the contraction or the contractive force at different yarn speeds, i.e. lower speed for shrinkage and higher speed for crimp.
This testing strategy has been successfully applied to a variety of different polymers and texturizing processes including the BCF-process.
However, in cases where the temperature ranges of optimum crimp development and shrinkage release over-lap, the strategy may fail. This can happen for some polymers or for very coarse yarn at high heater temperatures. At these higher temperatures a gradient exists between the outer and the inner filaments of a yarn.
For these cases Textechno now offers an optional third godet which allows users to pass the yarn through the heater at a very low tension (e.g. 0.01 cN/tex) between the first two godets, so that it can contract nearly freely. Between the second and the third godet a higher tension (e.g. 1 cN/tex) is applied which withdraws the crimp. The speed difference between the first and the third godet gives the shrinkage information, while the speed difference between the second and the third godet corresponds to the crimp.
The following example shows results from six polyamide BCF packages with different crimp (bulk) levels. The static crimp test according to EN 14621 performed on a TEXTURMAT ME+ was used as a reference to characterise the crimp, and the correlation of different tests on a DYNAFIL ME+ with this reference was studied. The results are plotted relative to the average value of all six packages for the respective test.A very good correlation and great differentiation is found between crimp force (CF) and the static reference test at 180°C (R²=0.997). If temperature is increased to 240°C (e.g. because shrinkage is to be measured, too), the correlation becomes weaker (R²=0.9043). In this case the crimp%-test using the third godet (CR%) helps to achieve a better correlation (R²0.9947).
As another example the results on five polyester BCF packages with different godet temperature settings (low, high, and standard) in production are shown below. At 180°C heater temperature a rather good differentiation could be achieved. At higher heater temperatures the effects of shrinkage and crimp interfere and the differentiation worsens.
At 30 m/min and 240°C e.g. the force value of the standard package is lower than the one of all other packages. Increasing the speed to 50 m/min, i.e. testing at a lower effective yarn temperature, leads to somewhat better results, but the differentiation between the package 'High2' and the standard is very week.
With the mechanical separation the crimp percentage reading clearly differentiates the packages with the different temperature settings, even at 30 m/min.
The DYNAFIL ME+ with an optional third godet thus allows users to distinguish between shrinkage and crimp signals by mechanical means and by different speed, i.e. different effective yarn temperature. In addition it remains an excellent tool for draw-force testing on POY or shrinkage, respectively. Further options like the automatic package changer for 20 positions, the on-line-connected COVAFIL+ evenness tester, the COMCOUNT linear-density tester (with sample preparation for spin finish analysis) and sensors for interlace and broken-filament counting make the DYNAFIL ME+ an effective tool for both research and production control on filament yarn.
Summary and outlook
A new method for single-fibre stiffness testing by means of a compression/bending method has been explained. The method will soon be available for Textechno's FAVIMAT+ single-fibre fineness-, crimp-, and tensile tester. The method can be automated with Textechno's AIROBOT2-system.
ITEMAT+, a new feeler-gauge type interlace tester, provides valuable information on the interlace quantity and quality in flat and texturized filament yarns.
A new third godet as an optional extension to the existing set-up makes the DYNAFIL ME+ even better suited for crimp- and shrinkage testing on texturized yarns and BCF from certain polymers.
In addition, Textechno and their subsidiary Lenzing Instruments will continue their efforts to develop new and easy-to-use testing methods and instruments with the highest possible degree of automation, precision, and reliability.
References
[Reu00] Reumann, R. D.: Prüfverfahren in der Textil- und Bekleidungstechnik. 1. Auflage Berlin, Heidelberg, New York: Springer, 2000, S.190ff.
[Sch08] Schmidt, R.: Mechanik Festigkeitslehre. 2. Aufl. Aachen, Rheinisch-Westfälische Technische Hochschule, Skriptum, 2008
[Shi99] Shioya, M.; Nakatani, M.; Nakao K.: Axial compression bending tests on carbon films and carbon fiber composites. Journal of Material Science 34 (1999), S. 1301-1311
[Tim85] Timoshenko, S. P.; Gere, J. M.: Theory of elastic stability. 2. Auflage. New York: McGraw-Hill International Book Company, 1985