RU1789908C - Method and device for determining viscosity of glass mass in the fiber-moulding zone - Google Patents

Description

create transverse vibrations of the material large enough for visual observation of the amplitude (and just such are required). The control is inaccurate when tensioning the jet, when changing the diameter and viscosity of the material; l; a; along the jets .; The method does not allow continuous monitoring.

It is also known to determine the longitudinal viscosity of the glass melt in the fiber forming zone, which, by its technical nature and the achieved result, is closest to the invention. This method consists in feeding the glass melt with a metering pump in the form of a jet from a horizontal capillary into the exhaust rolls, in changing (gradually increasing) and measuring their rotation speed, which is used to judge the deformation speed of the glass melt, in measuring the tension, and the diameter of the jet glass melt, the last of which judges the cross-sectional area of the glass melt, and in calculating the desired viscosity I by the formulas

(1) - F

about.

1 eSc

jrdc2

(3)

Sc

Ј fl (VB),

(2) (4)

where a is the tensile stress;

Ј is the strain strain rate (or, equivalently, the longitudinal velocity gradient);

Sc, dc, and F are, respectively, the cross-sectional area, diameter, and strain gauge of the molten glass measured by the sensor;

VB is the speed of rotation of the exhaust rolls.

The method is carried out by a device containing sensors for the rotation speed of the exhaust rolls and the tension of the molten glass jet, outputs connected to the inputs of the recorder, a jet diameter sensor and a control unit for the speed of rotation of the exhaust rolls. This unit includes a roll drive and provides a gradual increase in the speed of their rotation. Prototype objects do not allow continuous monitoring. They do not take into account the effect on Ј of the feed rate of the glass melt into the molding zone (speed of the ear rack). In installations for the production of optical fiber, the length of the forming zone is insignificant in comparison with the distance L from it to the exhaust rolls; therefore, in transient conditions, especially at high viscosity of the glass mass and small diameter de and elastic modulus E of the fiber, the velocity VB due to the elastic the deformation of the formed (hardened) fiber does not reflect the true speed

deformations Ј. As will be shown below, Ј by formula (4) must be found taking into account E, L, dB, Vui and F, i.e.

Ј f2 (VB, E, L, de, Ush, F). (5) The prototype does not take into account the change in the diameter of the jet both along its length and in time. In the stack method, the fiber and the stack are vertical, therefore, the horizontal arrangement of the jet proposed in the prototype cannot be used to compensate for its weight. The use of the analyzed objects for control of I in the processes of obtaining fibers from staffs leads to uncertainty also due to the proximity (2). More accurately

P (c)

0

(7

Sc

twenty

25

35

40

45

fifty

55

where FA is the strength of the inner friction.

At a constant drawing speed

FA F-Fn.H. + P, (7) where RP.N. - surface tension force;

P is the weight of the fiber portion above the tension sensor.

Glass melt in the molding zone is characterized by extremely large deformations, strain rate and intensity of mechanical impact (up to rupture). Therefore, the molten glass may (depending on its chemical composition) exhibit a viscosity anomaly. When controlling such an anomalously viscous glass melt, in which I tz (Ј), the prototype has a low sensitivity of changes in I when k is changed, since relation (1) is the tangent of the secant slope drawn from the origin through the given dependence point a (Ј ), and not the tangent of the slope of the tangent at this point of dependence a (Ј).

In the method for determining the viscosity of the glass melt in the fiber forming zone by measuring the diameter of the jet at the exit of the forming zone to calculate its cross-sectional area, measure the fiber tension and the speed of rotation of the exhaust rolls and determine the strain rate of the glass melt, measure the feed speed of the stack, excite vibrations the speed of rotation of the exhaust rolls, calculate the rate of exit of the fiber from the forming zone from the values of the cross-sectional area of the jet, fiber tension, rotation speed of the stretch x of the rolls and given the modulus of elasticity and the length of the fiber between the forming zone and the exhaust rolls, the strain speed of the glass melt is found from the calculated fiber exit speed and the feed speed of the staff, the oscillation amplitudes of the fiber tension and the product of the jet cross-sectional area and velocity are determined strains of the glass melt and viscosity are calculated from the ratio of the amplitude of the oscillations of the fiber tension to the amplitude of the product of the cross sectional area of the jet and the strain rate m voltage glass ..

A device for determining the viscosity of glass melt in the fiber forming zone, comprising sensors for the speed of rotation of the exhaust rolls, diameter and tension of the fiber, a unit for controlling the speed of rotation of the exhaust rolls and a recorder, is equipped with a head feed speed sensor, a probe of vibrations, a model of the speed of exit of the fiber from molding zone, a comparison element, a quadrator, a multiplier and two divisions of division, band-pass and smoothing filters and rectifiers, and the causative agent of probing vibrations is connected to input of the speed control unit of the exhaust rolls, the speed sensor of the exhaust rolls is connected to the first input of the model of the exit speed of the fiber from the forming zone, the tension sensor is connected to the second input of the model of the speed of exit of the fiber from the forming zone and through the series-connected first bandpass filter, rectifier ditch and smoothing filter - to the first input of the first division unit, the fiber diameter sensor is connected through a quadrator to the first input of the multiplier and the third input of the model of the fiber exit speed from For forming, the output of which is connected to the first input of the comparison element, the feed speed sensor of the staff is connected to the first input of the second division unit and the second input of the comparison element, the output of which is connected to the second input of the second division unit, the output of which is connected to the second input of the multiplier, the output of which through a series-connected second band-pass filter, a rectifier and a smoothing filter are connected to the second input of the first division unit, the output of which is connected to the recorder.

In order to carry out continuous determination and to increase sensitivity, it is proposed to find the desired viscosity by continuously determining the ratio

 (FA) m

I AM

(ScЈ) n

(8)

where (FA) m and (Sc Ј) m are the amplitudes of the sounding vibrations, respectively, of the internal friction force FA and the product of the cross-sectional area Sc and the strain rate E of the glass melt. The optical fiber forming zone is difficult to access for direct measurements and. observations, therefore, they express relation (8) in terms of easily defined parameters.

0 As is known, the most significant increments in the speed (102–103 times) and acceleration (104–105 times) of glass melt occur in the final section of the bulb (the viscosity of which is determined in this

5 case and for a very short period of time (- s). Therefore, the deformation of the molten glass in the area under consideration develops spasmodically. The magnitude of this strain jump can be estimated as

0 V4 - Vui (g)

g UV - Vu

Wush

It is proposed to consider the strain rate of the molten glass in the section under consideration proportional to this jump, 25 i.e.

. Ј KivVVuj- (s)

USCH

Since the cross-sectional areas of the final portion of the bulb and the formed fiber 30 differ insignificantly, it is also proposed to consider

Sc k2dB2. (11) In formulas (9-11)

UV and dB are the fiber speed and diameter at the 35th exit of the forming zone;

ki and k2 are proportionality coefficients.

To determine Uf, it is proposed to use a model that takes into account VB, E, L, 40. dB and F. The process of stretching a longitudinally moving absolutely elastic material is described in the most general form by the Yu.M. Winter equation, and in particular cases, by the Vasiliev N equations .A., Fainberg 45 Yu.M. et al. Glass fiber is linearly elastic until it breaks; therefore, it best satisfies these equations. The desired model for determining UV can be constructed as the outcome 50 from the Vasiliev NA equation, which is written as

TT7 - UV-L „RЈv. (12) 551 + Јv 0 + ev) 2

or in transformed form

Lp ЈВ - VB - Уф (1 + ЈВ) Х1 + Јв). (13) where ev is the relative elongation of the fiber between the forming zone and the exhaust rolls;

Phew

p is the differentiation operator.

Equation (13) can be written more simply, assuming the last factor 1 + Јv 1 .. Then we obtain the simplified Fainberg equation of Yu.M.

1r Јv Uv-Uf-UfЈv. (14)

Multiplying Equation (14) by ESB, where

jrdecn2 SB is the cross-sectional area of the fiber,

dBcp average fiber diameter between the forming zone and the draw rolls, and using the obvious

ESB ev F, (15) receive

 VBESB-LpF

. ESe + F (16).

A model of the rate of fiber exit from the molding zone in the proposed device is constructed in accordance with (16). Expression (16) can be realized in a variety of ways. One of the options for implementing the model is described further in the example.

The probe vibrations of the speed of the draw rolls VB cause fluctuations in the fiber tension F, the velocity UV, the internal friction force FA of the molten glass in the forming zone and the fiber diameter dB. Oscillations of the same strength RP.N. will be insignificant, since is determined primarily by the surface tension coefficient of the glass melt and the diameter of the staff. At an infralow frequency of probing oscillations, the force giving the fiber acceleration can be neglected, and the balance of forces will still be expressed by equation (7). Based on this and applying the rules of differentiation, from expression (7) find

(F $ m Fm, (T7) where Fm is the amplitude of the probe oscillations of the fiber tension. Substituting expressions (10), (11), (17) into formula (8), we finally obtain

Fm

AND -

kl (Sc

. Phew - Vu

Vu

%

m

kl k2 (dB 2 V) 1

The upper boundary of the excited probing vibrations is determined by the inertia of the exhaust rolls and their speed control unit. This boundary can be described as follows. The draw roll speed control unit is typically electric. The most inertial element in this unit is the actuator

an engine whose electromechanical time constant is 10-150ms, which is equivalent to a cutoff frequency of 16-1 Hz. This frequency determines the maximum frequency of the pathogen that will still be processed by the exhaust rolls. This frequency belongs to the infra-low-frequency range and corresponds to the restriction imposed on the frequency of probing oscillations when deriving relation (17).

A preliminary determination of the length L is required because this parameter appears in the algorithm for determining A (see relation (16), and it can change with

15 changes in the process plant caused, for example, by changes in protective coating operations.

In relation (16), the module E also appears.

20 technological conditions, characterized by high homogeneity of the chemical composition and the stability of the physical properties of each individual staff along its length, obtaining a fiber of constant diameter 25 mm, which leads to small variability of the E fiber obtained from one staff, this method provides for not continuous, but one-time determination E fibers for each staff.

30 In this case, E can be determined (in order to improve accuracy):

a) reference data for a product similar to the controlled one;

b) according to the measurement data of fiber samples made earlier and similar to controlled;

c) according to the measurement of fiber samples obtained at the beginning of this controlled spinning process.

40 As follows from the foregoing, in the proposed method, the ratios (16) and (18) are continuously determined. Therefore, the method allows viscosity determination to be carried out continuously.

45 The expressions (8) and (18) used are differential type relationships. Therefore, the method increases the sensitivity of the viscosity determination. In determining e, E, I, dB, Ush, F are additionally taken into account50 (equations 5, 10, 16); the influence of FH.H strength control results is excluded. and weights P (equations 6, 7, 17). Consequently, the accuracy of determining the longitudinal viscosity of the glass melt is increased.

In FIG. 1 is a block diagram of an apparatus for implementing the method; in FIG. 2, 3 - fragments of the oscillograms of the parameters of the process of obtaining fibers from quartz

staff when testing the declared objects.

Example. The elastic modulus E and the fiber length L between the forming zone and the exhaust rolls are preliminarily determined, probing infrared-frequency oscillations of the glass-fiber-fiber system parameters are excited by changing the speed of the exhaust rolls, the speeds VB of the exhaust rolls and the stacker are measured, and the fiber diameter dB the exit of the molding zone and the tension of the F fiber. The dB indicates the cross-sectional area Sc of the glass melt. From E, L, VB, de and F, the strain rate Ј of the glass melt is determined. The amplitudes Fm and (Sc Ј) m of the sounding vibrations are determined, respectively, of the tension F and the product Sc Ј and the ratio Fm / (Sc e) m are evaluated the longitudinal viscosity of the molten glass in the molding zone.

The device for determining the viscosity of glass melt contains a series-connected pathogen of probing vibrations 1 and a control unit 2 of the speed of rotation of the exhaust rolls 3, a sensor A of the speed of rotation of the exhaust rolls 3, model 5 of the speed of exit of the fiber from the forming zone, the sensor 6 of the fiber tension 7 , the first band-pass filter 8, the rectifier 9 and the smoothing filter 10, the first dividing block 11, the recorder 12, the fiber diameter sensor 13 at the outlet of the furnace 14, the quadrator 15, the multiplier 16, the second band-pass filter 17, the rectifier 18 and the smoothing filter p 19, comparing member 20, a sensor 21, feed speed of bars 22, the second unit 23 dividing.

Filters 8 and 17, 10 and 19, respectively, are made with the same amplitude-frequency characteristic and tuned to the frequency of the probing vibrations. The filter parameters are determined by the interference spectrum and the required speed of the converter.

A model 5 implementing relationship (16) may consist of series-connected differentiating element 24, a comparison element 25, and a division unit 26, the output of which is the output of the model. The model also contains a multiplier 27 and an adder 28, the first inputs of which are connected to the output of the filter 29, the second inputs are the first and second inputs of the model, respectively, and the outputs are connected to the second inputs of the blocks 25 and 26, respectively, with afoM the second input of the block 28 is also connected to the input of link 24, and the input of the filter 29 is the third input of the model. The transmission coefficients of the link 24 (proportional to L) and the filter 29 (proportional to E) are tuned in accordance with the change in the structural dimensions of the processing unit and the elastic modulus of the resulting fibers. The filter 29 is designed to suppress fluctuations in its output signal caused by sounding, and thereby obtain the necessary correspondence of its output signal not immediately

total de, a dBcp The filter can be in the form of a low-pass filter or a blocking filter tuned to the sounding frequency.

The device operates as follows. A periodic signal is supplied from the driver 1 of the probe oscillations to the input of the control unit 2. Unit 2, responding to a signal from the pathogen, causes sounding fluctuations in the speed of the exhaust rolls 3, which leads to sounding oscillations in the tension of the fiber 7, as well as the speed and diameter of the fiber at the exit of the forming zone. Sensor signals 4, 6 (directly) and 13 (via quadrator 15) containing sounding vibrations are fed to the inputs of model 5. The output signal of the model on the fiber speed / f at the output of the forming zone is converted using blocks 20, 23 and sensor signal 21 according to the algorithm ( 10) a signal about the strain rate Ј of the glass melt, which is supplied to the second input of the block 16. At the same time, a signal from the output of the square 15 of the cross-sectional area Sc of the glass melt 15 is received at the first input of the block 16. From the output of block 16, directly proportional to Sc Ј, a variable component caused by sounding is extracted by filter 17. The sinusoidal signal from the output of the filter 17 is converted using a rectifier 18 and a smoothing filter 19 into a direct current signal, directly proportional to (Sc Ј) m. Similarly to this signal, the signal of the sensor 6 is converted by a chain of blocks 8, 9 and 10 into a signal directly proportional to the value Fm of the probe oscillations of the fiber tension. The output signals of the filters 10 and 19 are fed to the inputs of block 11, at the output of which, in accordance with expression (18), a signal is obtained about the desired viscosity recorded by block 12.

When tested, the pathogen 1 was made in the form of a generator of sinusoidal oscillations with a frequency of 3 Hz. The amplitude of the sounding oscillations of the speed VB of the exhaust rolls is 1.5–3%. Bandpass filters 8 and 17 are implemented by cascading two active low-pass filters and two active low-pass filters of the second order with a quality factor of 0.5. The maximum adjustable filter gain in the passband of 1.8-4.4 Hz was 3200. Rectifiers 9 and 18 are made half-wave, and the smoothing filters 10 and 19 are in the form of active low-pass filters of the first order with a constant time 0.56 s and a gain of 1.6. The signal-to-noise ratio at the output of the filter 10, defined as the ratio of the values of this signal with and without sounding, was 4-7.3 depending on the drawing mode. The sensor 4 is made in the form of a sensor for the angular velocity of the exhaust rolls. Tension was measured in front of the exhaust rolls by the resistance force exerted by the fiber when it curved its path. This force was perceived by the mechatronic transducer, made on the mechanotron 6 MX 7C and used as a dynamometer. The output of the sensor 6 for suppressing high-frequency interference is equipped with a first-order low-pass filter with a time constant of 0.13 s. In addition, the fiber tension could be measured by an electronic bridge at the point of suspension of the staff, taking into account the weight of the staff or by resonance.

The sensor 13 was an industrial laser measuring device. The sensor 21 is made in the form of an angular velocity sensor (EMF) of a DC motor of independent excitation of the feed feed drive.

In FIG. 2, 3 the following designations are adopted: 30 - output signal (electric voltage) of unit 10; 31 - sensor signal 6; 32 - consumption of upper argon; 33 - electric current of the furnace heater. The levels of the remaining parameters of the fiber manufacturing process for recording waveforms are constant: dB 146 microns; VB 0.1 m / s; Wush

 5-10 m / s; lower argon flow rate 50 l / h; heater current (for FIG. 2) 89 A; the flow rate of upper argon (for FIG. 3) is 290 l / h.

By active changes in the argon flow rate and heater current, a change in the cooling and heating conditions of the glass melt in the fiber formation zone was achieved; as a result, the longitudinal viscosity A of the molten glass changed. As the argon flow rate increases (see Fig. 2), A increases, as evidenced by the increase in fiber tension (curve 31). From FIG. 2, the signal of 30 ° Fm is also increased. With an increase in the heater current (see Fig. 3), A decreases, o, due to the decrease in the fiber tension (curve 31). From FIG. 3 it can be seen that a signal of 30 ° Fm is thereby reduced. Thus increase or

a decrease in A also corresponds to an increase or decrease in the numerator Fm of relation (18). This, as well as the shown possibility in principle of exciting infrared frequency oscillations of the fiber tension by the exhaust rolls, prove the feasibility of the claimed objects in% of real conditions.

..-...-. about

Using the invention, it is possible to achieve continuity in determining the longitudinally oblique glass melt, which makes it possible to build systems for automatically controlling the physicomechanical and optical properties of the glass melt and the resulting fiber; increase in accuracy; increasing the sensitivity of determining changes in the longitudinal viscosity of abnormally viscous glass melt with changes in its strain rate.

Claims (2)

SUMMARY OF THE INVENTION 1. A method for determining the viscosity of glass melt in a fiber forming zone by measuring the diameter of the jet at the exit of the forming zone to calculate its cross-sectional area, measure the fiber tension and rotation speed of the exhaust rolls, and determine the strain rate of the glass melt, characterized in that , in order to increase the accuracy in the fiber production by the stauch method, measure the feed speed of the stile, excite fluctuations in the speed of rotation of the exhaust rolls, calculate the fiber exit speed From the forming zone from the values of the cross-sectional area of the jet, the fiber tension, the rotation speed of the exhaust rolls and the given modulus of elasticity and the length of the fiber between the molding zone and the exhaust rolls, the strain rate of the glass melt is determined from the calculated fiber exit speed and the feed speed of the staff, determine the amplitudes of the oscillations of the fiber tension and the product of the cross-sectional area of the jet and the strain rate of the tensile strain of the glass melt, and the viscosity is calculated from the ratio s oscillation fiber tension to the oscillation amplitude product of the cross sectional area of the jet to the strain rate tensile glass. 2. A device for determining the viscosity of the glass melt in the fiber forming zone, comprising sensors for the speed of rotation of the exhaust rolls, diameter and tension of the fiber, a unit for controlling the speed of rotation of the exhaust rolls and a recorder, characterized in that, in order to increase accuracy in the production of fiber - by the method, it is equipped with a bobbin feed speed sensor, a probe oscillation exciter, a model of the fiber exit speed from the forming zone, a comparison element, a quadrator, a multiplier, and two division blocks, strip and smoothing filters and rectifiers, the probe exciter being connected to the input of the exhaust roll rotation speed control unit, the exhaust roll rotation speed sensor connected to the first input of the fiber exit speed from the forming zone, the tension sensor connected to the second input of the model the speed of exit of the fiber from the forming zone and through a series-connected first bandpass filter, a rectifier and a smoothing filter to the first input of the first division unit, the fiber diameter sensor through square the adator is connected to the first input of the multiplier and the third input of the model of the fiber exit speed from the forming zone, the output of which is connected to the first input of the comparison element, the feed speed sensor of the staff is connected to the first input of the second division unit and the second input of the comparison element, the output of which is connected to the second input of the second the dividing unit / output of which is connected to the second input of the multiplier, the output of which is connected through the second series pass-through filter, the rectifier and the smoothing filter to the second input dividing in the first block, the output of which is connected with the registrar. Fig.Z ofwe