DD219074A3 - METHOD FOR THE PREVENTION DETERMINATION AND CONTROL OF THE FIBROUS PROPERTIES OF SYNTHESIS FIBERS AND SILICONS BY IDENTIFICATION - Google Patents

Abstract

The invention relates to a method for the predetermination and control of the fibrous properties of synthetic fibers and silks by characteristic value determination by determining from tensile tests the 0th and / or 1st and / or 2nd derivative of the stress-strain characteristic, and this with the technological process adjusting properties of the material is compared. For setpoint deviations, correction options are specified for the subsequent process. The method is applicable in the textile fiber and silk industry, in particular the textile silk processing.

Description

Klaus PoIlL Roland Fajerski Albrecht Bauer Rolf-Dieter Schulze

Title of the invention -

Method for predetermining and controlling the fiber properties of synthetic fibers and silks by characteristic value determination

Field of application of the invention _:

The invention is applicable to the pulp and fiber converting industry for the quality control of synthetic fibers and silk as well as for the anticipation and control of the processing properties independent of the manufacturing stage of the pulps. This method is applicable in the textile industry, in particular textile silk processing.

Characteristic of the pray

The test methods used in conventional quality control of synthetic fibers are carried out and evaluated independently of each other. The implementation of several test methods at the same time is very expensive. The informative value of the method with separate evaluation is i. a. too small to derive quantitative conclusions for the following processing stages and / or the time available to translate such conclusions into effective tax measures. For example, for a staining test, the untextured silk must

* 1 * 714 7 G

knit, form-fixed, dyed and visually evaluated without the results on the subsequent process

Can influence.

¹ ν.

It is known from the literature that the stress-strain behavior provides essential information about the microstructure of the paser, in particular about the orientation in the amorphous regions. (Egbers, G .: Materialprüfungen JJ _- (1969) 3, pp 73-78; Meffert, B .: chemical fiber / textile industry 29/81 (1979) 2, pp. 132-135, pp. 279-282; .Schultze -Bebhardt, F .: To interpret the stress-strain characteristics: oriented polyamide threads, Faserforsch, and Textiltechnik 28 (1977) 9, pp 467-474) · Furthermore, it is known that the orientation determines the properties of the fiber before and after the textile processing , Except for a "method for determining the ink receptivity of texturises" DD-PS 154 850, these findings did not lead to quantitative methods due to the difficulties in determining characteristic points of the tension-elongation behavior. It describes a method that determines the 1st relative maximum of the first derivative d <5Vd £ of the stress-strain characteristic * * = <O (£ ) and uses it as a measure of the dye uptake of stretched silk. The application of the method to partially stretched or unstretched silk does not lead to the goal. This method is limited to the determination of dyeability and thus suitable only for the selection of the material.

Object of the invention

The object of the invention is the development of a rational and economical method for the early identification of properties of synthetic fibers and silks, which occur in the subsequent technology stages. In addition to the quality control and selection options, the predetermination of the characteristics should allow the influence in the succession processes to be achieved.

rather end products are guaranteed

Explanation of the essence of the invention

- Task

It is a method for the prediction and control of the Paserstoffeigenschaften of synthetic fibers and silks to develop by characteristic value determination. The process must be independent of the manufacturing stage, the condition and the chemical constitution of the material. The characteristic values to be determined should at least enable the selection of products with the same special properties but also form the basis for the control of the following technology so that products with specific final properties to be incorporated can be obtained from inconsistent starting material.

Features of the invention

According to the invention, the object is achieved by performing tensile tests performed on unstretched, partially stretched or stretched synthetic fibers or silks, the stress-strain characteristic S (.6 ) £ f 1 J and / or its first derivative do / d £ f2 J and / or second derivative d ² 67d £ ² £ .f 3 Ji in the 6 * stress and <5 strain, is used as a whole for the predetermination and / or targeted control of the properties from which the influence variables x ^ are determined. To determine the influence variables x. the stress \ strain characteristics are scanned, using the factor and regression analysis. The particular advantage of the method arises from the fact that the entire curve is independent of the presence of extreme values and thus the entire information content of the Spannimgs-elongation characteristic for the prediction and control of the subsequent properties is available, via the function, y = f (x _i ) / f4 J, in which x ^ is the variable resulting from £ f 1 J to £ f3 J.

Then, the fibrous properties y of the material to be tested are determined «Subsequently, the corrections of the control variables z. from the puncture ^ y = g (4ζ ^ C f5 Js in which the deviation of the Paser property from a given nominal value and Δ ζ mean the correction values of the control variables are determined, which are then forwarded to the control organs · ^x

In order to carry out the characteristic value determination, it is advantageous to proceed in such a way that the comparison values of the variables x 1 are once determined, against which the test items are compared, or for which the corresponding corrections are proposed.

The comparison value determination is made as follows. A representative sample of the circumference Έ is taken from each subject and the tension-strain characteristic is determined on a tensile testing machine. The characteristic is η-malaligned until the Paser break. This yields a data matrix of the extent η χ U with η variables χ. per subject and Έ value sets. From each subject, the properties of interest y are detected in the present production stage or in the further processed state. >. · ....

The η variables x, serve as regressors (independent influencing variables), the characteristic y as regression edge (independent target variable) for a regression analysis. With the H χ η data matrix or a selection thereof, as well as the feature 'y, a regression analysis is performed with a suitable conventional regression method. It is advantageous, in particular with strong statistical dependence of the influence variables x. and / or insufficient accuracy of the comparison values (coefficient of determination, residual standard deviation) to use the first or second derivative d6Vd £ or d 6Vd £ of the stress-strain characteristic instead of the non-derivative characteristic and to proceed analogously as described above, another advantageous one Variant consists of

stress-strain characteristic or their 1. or .2. Derivation of a factor analysis. Instead of the η influence variables, then m factors (m <η), which are statistically independent »form the new influence variables. The regression analysis is then carried out with them. For the routine control of the production on the basis of the comparison values and the values of the current sample, a random sample of the circumference K is taken from each production unit or delivery, the stress-strain characteristic is recorded, the values of the significant influence variables are calculated as £ f1 J; £ f2 J '; ff} J, the probable values of the characteristic y C f4 J and the control value recommendations for the following. Steps £ f5 J calculated (characteristic value determination).

For this purpose, the values are always to be taken directly from the 6th and / or "1st and / or 2nd derivative, or the factor values to be determined via the factor analysis, then via the function £ f4 J into the characteristic values to be expected For example, the remainder values of R. are to be calculated in the texturized and modified state for a factor and regression analysis on an EDVA for about 100 minutes for 100 coils to be inspected For the control method without and with factor analysis, the prerequisites exist for "providing the control results even before the further processing of the delivery items." The entire data acquisition, characteristic value determination and control can run fully automatically as a closed program can for more properties y at the gl be sampled.

Particular advantages of the method are that beyond the selection and control options already shortly after the goods receipt the quality parameters

can be defined and thus there is a quick control over the delivery. From the deviation of the precalculating properties can be further drawn conclusions on the causes up to'Rawstoff- or pulp production process.

Since the entire information content of the stress-strain characteristic is used to determine the characteristic value, the influence on other passer properties is possible. Such properties are, for example, the crimp contraction, the breaking strength and the elongation at break of the textured and modified silk. The method is also independent of the form of the characteristic and is therefore not limited to a synthesis material.

The invention will be further described by the following examples. , <. , -

Example 1 (comparison value determination)

From several deliveries of stretched PES (plain silk) intended for production, a random sample of a total of 1 = 100 bobbins is taken. From each coil, the stress-strain characteristic curve is determined on the basis of 10 individual tests and subsequent averaging. * The calculation of the mean values and their subsequent storage for further calculations is expediently carried out electronically.

This can z. B. a microcomputer directly (on-line) are coupled to the Zugprüfgerät. The output voltage of the force measuring head of the tensile testing machine is amplified to the necessary input level of the subsequent analog-to-digital converter (ADC).

The force measurement signal to be digitized within 1 ms with an accuracy of 0.2% of the final value is switched to the corresponding input of a microcomputer. The strain measurement is performed according to the incremental method. An opto-electronic initiator detects Dennungsande- by means of a slot-typewriter connected to the expansion drive.

0.5 mm. Me electronic pulses whose number of strain change is proportional, are summed with a counter, which gives the result to the microcomputer. With a gripping length of the thread of 1 = 500 mm, this corresponds to an accuracy of 0.1% relative elongation. The strain measurement is thus independent of the set strain rate and fluctuations of the same. Under the conditions mentioned above, it is possible to precisely record and store the force (stress) values at intervals of 0.1 % strain each time until the sample breaks.

From the voltage-strain values stored in digital form, the 1st and 2nd derivative, d6Vd <£ and / or d S * / d £ , can be digitally formed in the digital manner and also stored at intervals of 0.1 % strain. In the present example, the 1 · derivative n .. = 17 values from the strain range 0.4 % <S < 4 % (at intervals of Δ C * 0.2 %) and η «= 5 values from the range- 4 % έ. ζ < 10 % (at intervals of Δ S - 1%). The η = η + n ₂ = 22 values per coil results in a data matrix η χ m = 22 χ 100 values which is provided internally or on an external data medium (perforated tape, magnetic tape) for the regression analysis. In the present case, the property y of the silk of interest is the dyeability of the PES in the textured and modified state.

Each sample spool is used to texture a sample, modify it, knit it and tie it in a tube. Thereafter, the test fabric is in a common bail with a dye which strongly marked differences in structure of the material, dyed (Formulation 0.5% Resolinmarineblau GLS = Disperse Blue 136) heating from 60 ⁰ C to 125 ⁰ C in 40 min, 30 min dyeing time, without carrier).

From each of the 100 samples, the spectral reflectance is measured in the remission minimum with a spectrophotometer, which is a measure of visually discernible color depth differences between the m samples and is used as a regression.

sand (dependent variable) for the regression analysis. With the now complete values of the first derivation of the stress-strain characteristic on the plain silk on the one hand and the associated remission values of the modified text silkides on the other hand, the regression calculation is carried out according to a known method and supplies the puncture f f4 J y = f (x _i ), where ^ is a selection from the η values made by the regression method:

I = 20.12 + 2.41 M (1.24) + 9.24 M (3.55) - 11.74 M (9.87)

/ "Gin 1 /

for which M (8 i) == x expresses the value of the 1 "derivative at the strain value S ^ in % .

'. ..- 'ι' ·. - '·.

Example 2 '

For routine non-uniformity control on 40 bobbins in a batch B with 1800 bobbins of stretched smooth silk, the bobbins tested (16.7 tex polyester silk) are the first derivative (modulus) values M (1,24); M (3.55); M (9.87), used in Equation 1

I = 20, 10 + 2.41. M (1.24) + 9.24 M (3.55) -11.74 M (9, .87) TGIn 1 J

and the anticipated remission of the material after texturing and modification predicted. This process is automatic * Only the tensile tester is to be used, if an automatic bobbin change is possible, the tensile test is also carried out automatically after a single loading of the automaton «The modulus values measured on the 40 coils and ^v the according to GIn. 1 calculated remission value shows Table 1.

Table 1

Spools no.

Batch B

measured values of the 1st derivative M (1,24) M (3,55) M (9,87)

ET /%. Tex U / ^. Tex H /%. Tex

Calculated value of the remission

1 2 3

4 5 6

7 8 9

10 11 12 13 14 15 16 17 18

19 20 21 22 23 24 25 26 27 28

29 30 31 32 33 34 35 36 37 38

39 40

7.39 7.29 7.71 7.24 7.31 6.90 7.52 7.25 7.38 7.35 7.60 7.27 7.36

7,75 7,51 7,51 7,45 7,42 7,50 7,19 7,62 7,70 7,86

7.49 7.38 7.83 7.15 7.61 7.72 7.52 7.55 7.57 7.85 7.57 7.76 7.40 7.86 7.56 7.81 7, 76

5.31

5.22

4.77

5.12

5.27

5.38

5.03

5.22

5.31

5.16

5.69

5.40

5.50

5.28 _:

5.09

5.40

5.39

5.39

5.30

5.28

5.54

5.55

5.58

5.49

5.52

5.26

5.66

5.45

5.77

5.39

5.40

5.38

5.49 5.32 5.36

5.27 5.40 5.58 5.53 5.41 0.86 0 _v 83 0.83 0.85 0.87 0.87 0.84 0.87 0.88 0.86 0.89 0, 89 0.88 0.85 0.83 0.88 .0.88 0.87 0.86 0.87 0.87 0.86 0.95 0.85 0.86 0.85 0.84 0.82 0.88 0.86 0.88 0.89 0.91 0.78 0.87 0.87 0.89 0.89 0.87 0.86

76.9 76.2 73.0 74.9 76.2 76.2

74.9 75.6

76.7

75.5

80.5

77.1

78.4

77.6

75.5

77.8

77.6

77.6

77.0

76.0

79.5

79.9

79.5

78.9

78.8

77.6

79.8

79.2

81.8

78.0

77.9

77.6

79.1

78.4

78.2

76.5

78.5

79.4

79.9

78.7

Arithmetic mean =

Standardabv / erro

scattering

77.74 1.81 3.26

Since an individual correction for each presentation unit is not technically possible (no individual scattering of the jobs), the entire batch B is compared with the previously supplied batch A. The mean remission value as a measure of the staining level of batch B is R _B = 77.74 % (see Table 1), the value of batch A was previously determined as R 1 = 77.0.

Charge / setpoint / actual value. B 77.0 77.7

The difference / R ^ - R ^ / »0.74 % is statistically tested with the t-test (99 % confidence level) and shown as unsecured.

Conclusion: Charge A and B have the same staining level. Both batches can be processed together in a ready-made piece without increasing the risk of color gradation in the final product. The comparison is also part of the automatic calculation.

Example 3

To determine the color unevenness of coils as in Example 2 a batch C the measured values with the aid of the factor analysis in factor values F. to F ₁ - and in the equivalent of the general function f f4 J y = f (x ^) underlying Equation 2

I = 77.43-1.6 F ₁ -0.53 F ₂ -0.73 F ₃ + O, 27 F ₄ -0.23 F ₅ /".GIn.2 J

used and the expected remission determined. Equation 2 is the result of the known ones. Procedure of factor and regression analysis. The mean remission value of charge C is 'R' = 81.6 %. The difference to the previous batch / R _c - R _ß / = 4.8 % is statistically significant.

The material shows deviations in the setpoint comparison (Table 2), which are noticeable in the final product as Parbunterschiede. Control units are proposed to correct the setpoints (Table 2, column 4-6). Than that of the general puncture 5. f f5 / underlying equation was developed especially for texturing as a correction variable z.

'- ^ΛΒ = -1.5

κ. ,

determined.

Pur a reduction of the staining level of the charge C um

4 R »- 4.8 % becomes 4Τ _φήτ = Ξτ4γ = 3 K.

This correction is sufficient to match Charge C to the other batches. The other control variables, with which a correction is alternatively possible, can be kept constant.

Charge setpoint actual value Correction door values

  '. , · Texturi'er- Δ T.Span- Modifizie

temp. T _Tex ~ tion planning

C 77.0 81.6 + 3'K -; - +01 + 0K

Example 4: '' ^·: · '· -. ^{; :} ·. "

A sample of 30 bobbins of pre-oriented (SSW) silk from lot D is tested. The 30 remission values, again determined after the general puncture [ f'4 J , give an average staining level Rt. = 77 »3 % and does not differ significantly from batch A. The scattering of remission values as a measure of the unequal

^; 2 · ·

Temperance between the coils is S ₇ . = 8.25. Scattering greater than s _Q = 4.5 increases the risk of color fringing.

The test with the F-test shows that s _{D is} secured above the critical value s _Q. From this follows the measure to select batch D and "by no means to use for strip sensitive products.

Example 5 '·

On Charge E, with the values x ^ of the non-derivative stress-strain characteristic (o (S ) on 40 bobbins of untextured silk on the basis of the general function fiA'J, the expected initial strain ^ Tex (3 »8) = f (xj) of the de-crimped, textured thread.

The arithmetic mean of the sample is 0.23 N / tex. "This amount is 12 % below the target value." This is to be expected in this batch, a modified tensile force of the textlien fabric and a deterioration of the deformation behavior of the Konfektionsstückes. By correcting the texturing tension by the corrective variable lead (gear ratio at the texturing machine) by one step, a correction of the corresponding texture silk property is carried out. ,

Example. 6

A sample of partially-stretched plain silk should be checked for color nonuniformity. For this type of material, Equation 1 of Example 2 is not usable. As a result, a suitable equation must be determined for partially-feathered silk. This is also done from sample material. The basis is the 0th, 1st and 2nd derivation of the stress-strain characteristic, from which values are taken at intervals of 0.5% strain. The result of the following calculation is the equation

* 49.2 - 10.4 6 * (47.5). + 28.0 6 (60.0)

where O (<£ *.) = _ s ^. the value of the O. derivative at the strain value S ^ expressed in % corresponds to.

On the basis of this equation, the prediction of the nonuniformity of the above and further samples is carried out analogously to Example 2. '

Claims (2)

invention claim 1. A method for predicting and controlling the passer properties of synthetic fibers and silk by identifying 'Valuation from the stress-strain characteristic, characterized in that from tensile tests on the 0 and / or 1 and / or 2 » derivative of the stress-strain characteristic with the general functions f 1 - 3 J and / or d6Vd <£ and / or d ² 6Vd £ ² , in which 6 "V = Mean tension and £ - stretch, influence variable χ. are determined as the function value from / 1 to 3 J at defined strain values C , and the general function £ 4 J y = f (x ^) predetermines the fibrous and sown properties y resulting after processing, which are used as characteristic values for the pulp and silk properties are valid, wherein, the concrete form of the general function. / "4 J and location and number of strain values defined by evaluation of the entire information content of the stress-strain characteristics of a finite number of initial sample in a known manner by regression or factor analysis can be determined by experimentally determining the actual pulp and silk properties for these samples, as well as from the characteristics thus determined for the pulp and silk properties via the general function / 5 / Iy = g ( Δz _i ), in the Δ γ the deviation of the pulp and silk property from a predetermined desired value and Δζ · the corr ectures of the control variables z. mean that the control values for the Bach subsequent processes for the production of the end product are determined with setpoint properties, wherein the determination of the function f5J again takes place in a known manner on a finite number of initial samples and the experimentally determined influence of varied control parameters. 2. The method according to item 1, characterized in that the control of the properties according to the determination of the general function 5 by the Texturiertempe- temperature and / or texturing thread tension and / or modification tempering. Method nach'Punkt 1 and 2, characterized in that the characteristic values for the control of the actuators manually and / or fully automatically fed.