TUBE MADE OF QUARTZ GLASS AND METHOD FOR MAKING THE SAME

Description

Technical area

[0001] The present invention relates to a method for producing a tube from quartz glass. In particular, the invention relates to a method for producing a tube from quartz glass by forming a hollow cylinder which has a hollow cylinder bore, an outer diameter C a , an inner diameter C i and a wall thickness of at least 20 mm, by rotating the cylinder continuously about a rotation axis at a relative feed rate VC and feeding it to a heating zone, softening it in certain areas therein, and radially expanding the softened area under the effect of a gas pressure applied in the hollow cylinder bore. A tube strand with a tube outer diameter T a , a tube inner diameter T i and a tube wall thickness is continuously formed from the softened area and drawn off at a draw-off rate VT.

[0002] Furthermore, the invention relates to a tube made of quartz glass having a tube outer diameter T a , a tube inner diameter T i , a tube wall thickness, an outer surface and an inner surface. State of the art

[0003] Using such processes and devices, hollow cylinders made of glass, particularly quartz glass, are formed into tubes in one or more hot-forming steps, whereby the radial tube dimensions are modified relative to the radial dimensions of the hollow cylinder. A starting hollow cylinder rotating around its longitudinal axis is softened zone by zone and, under the influence of a radially outward force, is either stretched against a forming tool arranged at a predetermined radial distance from the longitudinal axis of the tube, or it is formed without the use of a tool. The radially outward force is based on centrifugal force and / or an internal overpressure in the hollow cylinder's inner bore (also referred to as "gas pressure").

[0004] The terms "internal bore" or "bore" in connection with a tube or hollow cylinder do not imply that the internal bore or bore was created by a drilling process.

[0005] With regard to heating the initial hollow cylinder, a distinction can be made between a flame-based heating zone and an electrically heated furnace. Quartz glass is a good thermal insulator, so flame-based heating zones, in which the surface heating is primarily based on recombination reactions with a low penetration depth of the heating effect, are limited to the forming of initial hollow cylinders with thin walls, for example, wall thicknesses of less than 10 mm. In contrast, electrically heated furnaces allow for a higher and more homogeneous heat input over a longer distance, as the infrared radiation can penetrate deep into the glass.

[0006] Fundamentally, high dimensional accuracy and surface quality of the drawn pipe strand are desired. Tool-free forming ensures high surface quality.

[0007] Maintaining dimensional accuracy, however, is generally easier to achieve using a forming tool. In any case, the continuous measurement and control of a radial dimension of the pipe string, such as the outer diameter, inner diameter, or wall thickness, is essential. Common control variables for such a control system include the gas pressure, the relative feed rate between the hollow cylinder and the heating zone, and the temperature in the heating zone.

[0008] A process for producing quartz glass tubes of the type mentioned above is known from JP 2010-111557 A. A thick-walled hollow quartz glass cylinder with an outer diameter of 200 mm and a horizontally oriented longitudinal axis is continuously fed into a heating furnace while rotating at a feed rate of 4 cm / min, where it is softened zone by zone to a temperature of approximately 2100°C. Water-cooled graphite plates, which serve as forming tools, are arranged within the heating furnace at an adjustable radial distance from the longitudinal axis of the tube. Due to excess pressure within the hollow cylinder, the softened quartz glass is blown against the graphite plates, forming a tube strand that is withdrawn from the heating furnace at a withdrawal rate of 12 cm / min. During this process, the tube strand detaches from the graphite plates, which can result in changes in diameter depending on the process conditions.To achieve independent control of the pipe string's outer diameter to a target value of 340 mm, two cameras are provided. These cameras are positioned to the left and right of the pipe string, respectively, and generate an image of the left circumferential line of the pipe string and an image of the right circumferential line of the pipe string. Image processing continuously determines the outer diameter of the pipe string from these images as a control variable for the control system. The control variable is the radial distance between the water-cooled graphite plates and the pipe's longitudinal axis, which can be adjusted using a motorized mold jaw tracking system.

[0009] The documents US 2017 / 327403 A1 and JP 2007-001811 A describe further methods for producing large diameter quartz glass tubes using a forming tool, such as graphite plates. Technical task

[0010] As the hollow cylinder rotates around its longitudinal axis, the graphite plates spaced apart form a circular "quasi-opening" through which the softened quartz glass is forced during the forming process. The graphite plates lead to defects, particularly spiral-shaped streaks on the outside of the drawn quartz glass tube.

[0011] As a result of the internal overpressure of the gas, a circumferential bulge of soft quartz glass can build up in front of the graphite plates—as seen in the direction of the hollow cylinder's feed to the furnace. This volume is subject to a renewal mechanism in which new glass is constantly accumulated and drawn off at the same rate through the tube string, since the withdrawal speed of the tube string is significantly greater than the advance speed of the hollow cylinder. This speed ratio causes an "elongation" of the hollow cylinder and is inevitably accompanied by a reduction in the glass cross-sectional area of ​​the formed tube compared to that of the original hollow cylinder.

[0012] All attempts to produce quartz glass tubes with the same or higher glass cross-sectional area using this known method have so far failed. This can be attributed to the fact that this requires a reversal of the aforementioned speed ratio, i.e., a withdrawal speed of the tube strand that is lower than the advance speed of the hollow cylinder, which is also commonly referred to as "upsetting." During upsetting, the aforementioned renewal mechanism of the bead volume no longer functions, since more softened glass volume is permanently accumulated in front of the "quasi-opening" than is withdrawn. When forming a hollow cylinder, this "pushing" of the softened glass volume against the forming tool either requires the forming tool to be opened briefly, or the bead is pushed inward at irregular intervals, or it remains stuck to the forming tool.In the first case, there are fluctuations in the inner and outer diameters; in the second case, there are fluctuations in the inner diameter; and in the third case, the forming process is interrupted. In either case, the process is unstable.

[0013] Due to their short heating zone, compression of the initial hollow cylinder is possible to a small extent in flame-based heating zones, but - as already mentioned above - only with small wall thicknesses of the initial hollow cylinder to be formed of less than 10 mm.

[0014] There is therefore a need for large and at the same time thick-walled quartz glass tubes with good surface quality.

[0015] It is therefore an object of the invention to provide a corresponding tube made of quartz glass.

[0016] Furthermore, it is an object of the invention to provide a method which allows thick-walled starting hollow cylinders made of quartz glass with wall thicknesses of more than 20 mm to be formed into tubes with a larger diameter but approximately the same or larger glass cross-sectional area. General description of the invention

[0017] This object is achieved according to the invention with regard to the method starting from the method mentioned at the outset in that the gas pressure is used as a manipulated variable of a diameter regulation and / or a diameter control for the pipe outer diameter or for a geometric parameter correlated with the pipe outer diameter and that in a pressure build-up phase the gas pressure is gradually increased from a lower initial value to a higher final value and that the following applies to the ratio of VC and VT: VT = Vc ± 0.2 VC .

[0018] The initial hollow cylinder—or hollow cylinder for short—is made of quartz glass. This can be melted from naturally occurring raw materials or produced from synthetically produced SiO2. The quartz glass is undoped or contains one or more dopant substances. The hollow cylinder is thick-walled; that is, its wall thickness is, for example, more than 20 mm, more than 40 mm, more than 60 mm, or more than 80 mm.

[0019] During the forming process, the initial hollow cylinder is essentially oriented so that its longitudinal axis runs preferably horizontally. The result of the forming process is a tube strand from which one or more quartz glass tubes of the desired length are cut. The terms "tube" and "tube strand" are used synonymously here. The quartz glass tube has a larger outer diameter and a larger inner diameter than the initial hollow cylinder. Unless an explicit distinction needs to be made between inner diameter and outer diameter, the collective term "diameter" or "radial geometry" is used below for both dimensions. The "radial geometry" of the initial hollow cylinder is also referred to as the "starting geometry," and the "radial geometry" of the tube strand is referred to as the "target geometry."

[0020] The goal of the forming process is to produce a quartz glass tube with a glass cross-sectional area that is approximately equal to or even larger than the glass cross-sectional area of ​​the original hollow cylinder. The glass cross-sectional area is the area covered by glass in a section perpendicular to the tube's longitudinal axis.

[0021] The target geometry, particularly the outer diameter of the pipe string, is the target value for a closed-loop or open-loop control system. The closed-loop control is computer-controlled based on measured values ​​of the target value; the control can also be performed, at least partially, manually by adjusting parameters based on readouts. Combinations of closed-loop and open-loop control are also possible, for example, in closed-loop control with feedforward control. Unless expressly excluded, the terms "closed-loop control" and "regulate" are used collectively below to refer to "open-loop" and "control," respectively, and also to combinations of these measures.

[0022] The forming process according to the invention differs from the previous procedure in at least three aspects. (i) The use of a forming tool is omitted for the reason explained above. This is because "pushing" the softened glass volume through the "quasi-opening" of the forming tool would lead to an unstable process in which increasing amounts of softened glass volume would accumulate in front of the forming tool, causing waviness in the tubing. This results in fluctuations in the cross-sectional area that cannot be determined by calculating the compression rate (ratio of feed rate to withdrawal rate) and the cross-section of the initial hollow cylinder. Eliminating the forming tool does offer advantages: the tube surface is smooth and bright. Chemical contamination of the glass by the forming tool material can be ruled out. However, in order to reliably and precisely produce the desired target geometry through free forming, a different control for the tubing diameter is required.In the method according to the invention, the gas pressure is therefore used directly or indirectly as a manipulated variable for a diameter regulation and / or a diameter control for the pipe string outer diameter or for a geometric parameter correlated with the pipe outer diameter. Indirect use in this sense occurs, for example, when the manipulated variable is not the gas pressure, but a parameter correlated with the gas pressure, such as the flow rate of a compressed gas. The inner and / or outer diameter of the drawn-off pipe string is measured or read. Geometric parameters correlated with the pipe outer diameter include, for example, the inner diameter and the wall thickness of the pipe. The pipe outer diameter can be detected optically. The inner diameter is determined from the cross-sectional area and the compression rate (ratio of feed rate to withdrawal rate).This requires a stable process in which the cross-sectional area does not change. The deformation of the hollow cylinder is based on the gas pressure or the centrifugal force during rotation (spinning), or on both effects. The contribution of the gas pressure to the tangential stress σT in the tube wall is described by the so-called boiler formula: . σT = Druck * Radius / Wanddicke where: Pressure = Gas pressure Radius = Tube radius in the deformation zone Wall thickness = Wall thickness in the deformation zone In the "deformation zone," wall thickness, diameter, and temperature (viscosity) change continuously. For example, the diameter of the softened glass strand increases continuously from the hollow cylinder to the tube, depending on the glass's resistance to deformation, which is determined primarily by the wall thickness and viscosity. In longitudinal section, the deformation zone has an overall funnel shape with a more or less pronounced S-shaped transition between the hollow cylinder and the tube, which is also referred to as a "shoulder" below. In practice, the deformation zone typically has a length of several hundred millimeters, for example, 200 to 800 mm, often around 500 mm. The tangential stress, in turn, determines the tube radius (r) in the deformation zone according to the differential equation: dRadius / dt = Spannung * Radius / Viskosität with: dRadius / dt = strain rate Stress = tangential stress Radius = pipe radius in the deformation zone Viscosity = local viscosity averaged over the wall thickness (ii) For each given set of parameters (radial geometry of the initial hollow cylinder and pipe string as well as process parameters such as feed and withdrawal speeds, gas pressure and temperature distribution in the deformation zone) there exists only one stable, stationary shoulder shape which results as a solution to the differential equation (2). Only the value for the gas pressure on which this calculation is based ideally leads to the target geometry starting from the starting geometry and depending on the process parameters. This value for the gas pressure can be used as the nominal initial value for the control or open-loop control of the radial geometry of the pipe string (provided that the gas pressure is used at least as an auxiliary control variable).However, it has been shown – and this also follows from equation (2) above – that every pressure change affects not just one point, but all points in the deformation area, and to varying degrees depending on the current local conditions. Therefore, every pressure change leads to fluctuations in the target geometry, which only become visible much later, i.e., after a very long dead time, and which can no longer be corrected. The control system can easily enter an unstable, oscillating state, resulting in a pipe string with a wavy structure. Due to manufacturing constraints, the long dead time is difficult to shorten. Therefore, it is recommended to avoid abrupt pressure changes, such as those that might be made at the beginning of a forming process to adjust the gas pressure to its nominal initial value.Instead, it is proposed to gradually increase the gas pressure from a lower initial value to a higher final value during a pressure adjustment phase. Gradually means that the gas pressure is increased steadily - in the smallest possible steps or, ideally, continuously. The initial value is a pressure value that is zero or significantly below the final gas pressure value, and the final value is a pressure value that lies within the range of the control value for the regulation / control of the target geometry. During the pressure build-up phase, the forming process is already taking place, i.e., quartz glass material of the initial hollow cylinder is continuously formed into quartz glass material of the pipe string, even though the pipe string has not yet reached its specified radial geometry. For each pair of initial and target geometries, a maximum possible ramp exists to avoid entering an unstable, oscillating state.This can be calculated using differential equation (2). This maximum possible ramp is an ideal value. Higher pressure increase ramps can lead to a deterioration in the constancy of the drawn-off outer diameter and / or to greater material loss. On the other hand, lower pressure increase ramps than the ideal value are harmless with regard to the constancy of the outer diameter. The opposite is the case; a lower limit in this regard could not be determined. However, a lower pressure increase ramp results in a long period of time until the specified gas pressure end value and thus the diameter target value are reached, which is accompanied by time and material losses, since the specified radial geometry of the pipe has not been established by then and the pipe material produced up to that point represents only material loss. Therefore, there is an economic motive to set the pressure increase ramp as large as possible.Although not the time loss, the material loss can be reduced by using a starting hollow cylinder in which the front end is made of dummy material of lower quality. (iii) One aim of the forming process is to produce a tube with a larger diameter than the starting hollow cylinder but with approximately the same or larger glass cross-sectional area. To achieve this, the ratio of the withdrawal speed VT of the tube strand and the feed speed Vc of the starting hollow cylinder must be set such that: VT = Vc ± 0.2 VC . In the forming process at hand, the withdrawal speed VT is therefore in the range of the feed speed Vc or even lower than it. In contrast, when elongating a starting cylinder to form a blank strand, the feed speed to the heating zone is significantly lower than the withdrawal speed of the tube strand from the heating zone.This results in a certain inherent stability of the drawing process during elongation, which facilitates the adjustment and control of the radial geometry of the tube strand. However, during elongation, the glass cross-sectional areas of the starting cylinder and the tube strand generally differ significantly. The inherent stability inherent in the elongation process cannot be achieved in the present forming process. The method according to the invention compensates for this disadvantage using the measures (i) and (ii) explained above.

[0023] In this way, for the first time, tubes with wall thicknesses of more than 20 mm can be produced from thick-walled quartz glass hollow cylinders by free forming, with a glass cross-sectional area equal to or larger than that of the original hollow cylinder.

[0024] A conventional control to the target geometry of the pipe string would, starting from the low initial gas pressure value, result in the fastest possible adjustment of the final gas pressure value, which, however, can lead to process instabilities and an oscillating state for the reason explained above. The gradual increase of the gas pressure in the pressure build-up phase is therefore preferably carried out independently of the diameter regulation or independently of the diameter control.

[0025] In the simplest case, the gas pressure is increased manually in small increments until the target gas pressure is reached. Alternatively, the gas pressure is increased linearly or based on a non-linear function until the target gas pressure is reached. Independent of regular diameter control, the gradual increase in gas pressure during the pressure build-up phase is preferably also achieved through closed-loop control.

[0026] The regular diameter regulation or control is assigned a diameter target value for the pipe outer diameter and / or for the pipe inner diameter, with a target gas pressure being assigned to the diameter target value. It is advantageous if the lower initial value is in the range of 0 to 50% of the target gas pressure, preferably in the range of 10 to 30% of the target gas pressure, and the higher final value is in the range of 70 to 110% of the target gas pressure, preferably in the range of 90 to 100% of the target gas pressure.

[0027] The pressure build-up phase precedes the actual forming process. At the beginning of the pressure build-up phase, the hollow cylinder has reached its softening temperature, and the process parameters, such as the heating zone temperature, the feed rate, and the withdrawal rate, are set as expected in the actual forming process. During the pressure build-up phase, the gas pressure in the hollow cylinder's inner bore (and thus also in the tubing's inner bore) is slowly increased, starting from the lower initial pressure value, to the final value. Only when the target gas pressure is reached is a tubing with the target geometry produced. The tubing material formed up to this point is wasted. This material loss can be reduced if the initial section of the initial hollow cylinder is made of inferior quartz glass.Regardless, the pressure build-up phase should be as short as possible and only as long as necessary, both to minimize material loss and to achieve a stable state of the forming process through a flat pressure increase ramp. To keep the pressure build-up phase as short as possible despite a flat pressure increase ramp, a certain initial gas pressure value above zero up to approximately 50% of the target gas pressure, preferably up to a maximum of 30% of the target gas pressure, is advantageous.

[0028] The target gas pressure is, for example, in the range of 2 to 20 mbar, preferably in the range of 3 to 15 mbar, and particularly preferably in the range of 4 to 10 mbar. A particularly high gas pressure of more than 20 mbar can impair process stability.

[0029] On the other hand, it is not absolutely necessary that the pressure increase during the pressure build-up phase exactly reaches the target gas pressure. The regular diameter regulation or control can take over the pressure adjustment when at least 70%, preferably at least 90%, of the target gas pressure has been reached, or even when the pressure is already slightly higher than the target gas pressure.

[0030] A balance must be struck between the shortest possible pressure build-up phase, thus minimizing material loss, on the one hand, and the flattest possible pressure increase ramp, thus enabling precise adjustment of the target geometry with minimal diameter fluctuations, on the other. A particularly suitable compromise has been found for the pressure build-up phase to last between 1 and 120 minutes, preferably between 5 and 100 minutes, particularly preferably between 10 and 80 minutes, and especially between 15 and 60 minutes.

[0031] These considerations also determine the temporal progression of the pressure increase during the pressure build-up phase. The pressure increase can, for example, occur with a constant gradient (linear) or it can have a non-linear progression with a decreasing or increasing gradient at the end. It has proven advantageous if the gas pressure during the pressure build-up phase is increased at least temporarily with a temporal ramp Δp, for which the following applies: 0.01 mbar / min < Δp < 0.8 mbar / min, preferably: Δp < 0.5 mbar / min and in particular Δp < 0.2 mbar / min.

[0032] Regardless of the temporal progression of the pressure increase during the pressure build-up phase, the pressure increase ramp can also be assigned an average value over the entire pressure build-up phase. In this case, the gas pressure is preferably increased during the entire pressure build-up phase with an average temporal ramp Δp m , for which the following applies: 0.01 mbar / min < Δp m < 0.5 mbar / min, preferably: Δp m < 0.1 mbar / min and in particular Δp m < 0.06 mbar / min.

[0033] The forming process is particularly suitable for the industrial production of large-volume and thick-walled quartz glass tubes. Therefore, the hollow cylinder (C) and the tube (T) preferably have the following radial geometry: 150 mm < C a < 300 mm , 30 mm < C i < 180 mm , 40 mm < C a − C i / 2 < 100 mm 300 mm < T a < 500 mm , 250 mm < T i < 450 mm , 20 mm < T a − T i / 2 < 60 mm .

[0034] The forming process results in a quartz glass tube with a larger outer diameter and larger inner diameter than the original hollow cylinder. The following relationships preferably arise for the hollow cylinder and the tube strand: 4 ⋅ C i < T i < 8 ⋅ C i , C a + 100 mm < T a < C a + 300 mm .

[0035] Particularly preferably, the tube outer diameter T a is in the range of 1.5 times to 2.2 times the hollow cylinder outer diameter C a .

[0036] The forming process is also particularly suitable for producing quartz glass tubes with a large diameter and approximately the same or larger glass cross-sectional area as the initial hollow cylinder. The following advantageous relationship exists between the glass cross-sectional area AC of the initial hollow cylinder and the glass cross-sectional area AT of the tube strand: AT = AC ± 0.15·AC , preferably: 0.90·AC ≤ AT ≤ 1.15·AC , in particular: AC ≤ AT ≤ 1.10·AC .

[0037] The glass cross-sectional areas Ac and AT are preferably in the range of 250 to 1000 cm 2< .

[0038] When elongating a starting cylinder into a raw strand, its feed rate to the heating zone is significantly lower than the withdrawal rate of the tube strand from the heating zone. This results in a certain inherent stability of the drawing process during elongation, which facilitates the adjustment and control of the radial geometry of the tube strand. However, the glass cross-sectional areas of the starting cylinder and the tube strand differ significantly. In contrast, in the forming process at hand, the withdrawal rate of the tube strand is in the range of the feed rate or even lower. As a result, the inherent stability inherent in the elongation process cannot occur. The method according to the invention avoids this disadvantage, with the ratio of Vc and VT preferably being: 0 , 8 ⋅ V C ≤ V T ≤ 1 , 05 ⋅ V C , in particular: 0 , 9 ⋅ V C ≤ V T ≤ 0 , 99 ⋅ V C .

[0039] The wall thickness of the pipe string is determined by the ratio of the velocities Vc and VT. If this velocity ratio is less than 1, the pipe string is not elongated but rather upset.

[0040] It has been shown that process instabilities are easier to avoid with a comparatively low throughput of glass mass. In this regard, it has proven effective to adjust Vc so that a throughput of quartz glass mass of at least 30 kg / h but less than 60 kg / h, and preferably less than 45 kg / h, is achieved.

[0041] With regard to the quartz glass tube, the above-mentioned object is achieved according to the invention in that it has a fire-polished outer surface produced by tool-free forming in the molten metal flow and in that the following apply to the tube outer diameter T a , the tube inner diameter T i and the tube wall thickness: 300 mm < T a < 550 mm , 250 mm < T i < 450 mm , 20 mm < T a − T i / 2 < 60 mm .

[0042] The quartz glass tube has a large outer diameter and is nevertheless thick-walled, meaning its wall thickness is greater than 20 mm. It has an outer surface produced without the use of a mold, i.e., fire-polished in the molten state, which is smooth and free of streaks. It can be produced by the method according to the invention and is preferably produced using this method. The tube is preferably made of synthetically produced quartz glass.

[0043] Advantageously, the following apply to the pipe outer diameter T a , the pipe inner diameter T i and the pipe wall thickness (T a -T i ) / 2): 330 mm < T a < 500 mm , 275 mm < T i < 400 mm , 25 mm < T a − T i / 2 < 50 mm . Example

[0044] The invention is explained in more detail below using an exemplary embodiment and a drawing. In detail, Figure 1a device for forming a hollow cylinder made of quartz glass into a quartz glass tube in a side view and in a schematic representation, Figure 2 a photograph of a quartz glass tube section emerging from a heating furnace during a pressure setting phase of a forming process according to a first comparative example, Figure 3 a diagram with profiles of outer diameter and wall thickness of the formed quartz glass tube section of Figure 2 , Figure 4 a diagram showing the time course of the gas pressure during the start phase of the forming process and afterwards in the first comparison example, Figure 5 a photograph of a quartz glass tube section emerging from a heating furnace during a forming process according to a second comparative example, Figure 6 a photograph of a quartz glass tube section emerging from a heating furnace during a forming process according to a first embodiment of the invention, Figure 7a diagram with profiles of outer diameter and wall thickness of the formed quartz glass tube section of Figure 6 , Figure 8 a diagram showing the time course of the gas pressure during the start phase of the forming process and afterwards in the first embodiment, Figure 9 a diagram with profiles of outer diameter and wall thickness for the reference example, and Figure 10 a diagram showing the time course of the gas pressure during the forming process in the reference example.

[0045] Figure 1shows a schematic of a device for forming a hollow quartz glass cylinder 2 into a tube 22. The hollow cylinder 2 consists of synthetically produced, undoped quartz glass. It is manufactured using the so-called OVD (Outside Vapor Deposition) process in a soot deposition process. In this process, a liquid or gaseous starting substance is subjected to a chemical reaction (hydrolysis or pyrolysis) and deposited from the gas phase as solid SiO 2 on a rotating mandrel. The starting substance is, for example, silicon tetrachloride (SiCl 4 ) or a chlorine-free silicon compound. The intermediate product obtained is a porous SiO 2 soot body in the shape of a hollow cylinder, the inner bore of which has been formed by the mandrel that is later removed. The hollow cylinder 2 made of synthetic quartz glass is obtained by vitrifying the soot body.

[0046] Welded to the end faces of the hollow quartz glass cylinder 2 to be formed are holding tubes 4, 5, which are clamped in the chuck 6 of a horizontal glass lathe 8 and rotate synchronously around the rotation axis 10. The front holding tube 4 in the feed direction 32 is closed by a plug 34. A compressed gas can be introduced into the other holding tube 5.

[0047] The hollow cylinder 2 is continuously pushed by the glass lathe 8, rotating around its rotational and longitudinal axis 10 at a predetermined feed rate, into a resistance furnace 12 that surrounds the hollow cylinder 2 in a ring shape and is heated zone by zone to a temperature of approximately 2100 °C. The resistance furnace is connected to a computer 24 via a controllable power source 26.

[0048] A gas inlet 14 allows a compressed gas to be introduced into the inner bore 16 of the hollow cylinder 2 and the tube 22, and a defined gas pressure can be set. The gas inlet 14 is connected to a compressed gas container 20 and the computer 24 via an adjustable valve 18.

[0049] Driven by centrifugal force and gas pressure, the low-viscosity mass of the quartz glass hollow cylinder 2 is inflated to the desired diameter of the quartz glass tube 22. To measure the resulting outer diameter and to regulate the outer diameter, a measuring and control device is provided. This device comprises a high-resolution CCD camera 30 directed at the pipe string 22. The measurement data acquired by the camera 30 is continuously transmitted to the computer 24, on which an image processing program is installed. This program determines the current outer diameter of the pipe string 22. This is fed into the control system as the actual value of the controlled variable (pipe string outer diameter), whose manipulated variable is the gas pressure. Gas lines are schematically marked by double lines, power lines by single lines, and data lines by dashed lines.

[0050] In the following, Figure 1Examples of the forming process are explained. Comparison example 1

[0051] C a 220mm 43mm Feed VC 7.5 mm / min Deduction VT 7.4 mm / min mold tool Without Throughput 36kg / h Target T a 363mm Target T i 290mm Target wall thickness 36.5 mm Heating output approx. 740kW speed 10 rpm Glass surface AC 365 cm 2< . Glass surface AT 374 cm2<

[0052] In this comparative example, the initial hollow cylinder 2 was inserted into the heating furnace 12 while rotating around its longitudinal axis 10, and the length of the initial hollow cylinder 2 was evenly heated in the heating furnace 12. The gas pressure was then set to the value calculated using equation (2) to achieve the specified pipe string outer diameter (target gas pressure). In this case, the target gas pressure is 5.3 mbar. At the same time, the feed rate was set to 7.5 mm / min, and the free pipe string end was withdrawn at the specified withdrawal speed of 7.4 mm / min.

[0053] After some time, this led to an excessively rapid increase in the outer diameter of the tubing, so the gas pressure had to be reduced to 1 mbar to prevent the glass from coming into contact with the heating tube of the resistance furnace 12. This led to a rapid reduction in the outer diameter of the tubing after some time, so the pressure had to be increased again to 4 mbar. However, a stable condition was not achieved, and the target geometry could not be achieved by the end of the forming process.

[0054] Figure 2 shows a photo of the pipe string 22 exiting the heating furnace 12. The irregularity of the outer diameter is immediately visible.

[0055] The diagram of Figure 3shows the corresponding curves of outer diameter and wall thickness. The y-axis represents the diameter Ta and the wall thickness d in millimeters, and the x-axis represents the length L of the drawn pipe string in meters. Curve A represents the outer diameter, and curve B represents the wall thickness (which could not be measured consistently due to the significant diameter fluctuations).

[0056] The diagram of Figure 4 shows the gas pressure curve explained above. The y-axis represents pressure p in mbar, and the x-axis represents time t in minutes. Comparison example 2

[0057] C a 220mm C i 43mm Feed VC 7.5 mm / min Deduction VT 7.4 mm / min mold tool Contact at 363mm outer diameter Throughput 36kg / h Target T a 363mm Target T i 290mm Target wall thickness 36.5 mm Performance approx. 740kW speed 10 rpm Glass surface AC 365 cm2< Glass surface AT 374 cm2<

[0058] In this comparative example, a forming tool was installed in the heating furnace 12 to determine the outer diameter of the drawn-off tubing 22. The initial hollow cylinder 2 was inserted into the heating furnace 12 while rotating around its longitudinal axis 10, and the length of the initial hollow cylinder 2 was evenly heated in the heating furnace 12. The feed rate was set to 7.5 mm / min, and the free tubing end was drawn off at the specified draw-off speed of 7.4 mm / min. The gas pressure was initially set to a fixed 4 mbar. An attempt was made to form against the forming tool, which was set to an outer diameter of 363 mm. However, upon initial contact with the forming tool, the quartz glass began to twist. The pressure was then reduced to 2 mbar and then increased again with a gentle ramp of 1 mbar / h (approximately 0.017 mbar / min).Despite several attempts to stabilize the forming process, the specified target dimension was not achieved and the mold ultimately had to be opened to release the quartz glass mass.

[0059] The photo of Figure 5 shows the twists and irregularities in the outer diameter of the pipe string that has been removed so far. Example 1

[0060] C a 220mm C i 43mm Feed VC 7.5 mm / min Deduction VT 7.4 mm / min mold tool without Throughput 36kg / h Target T a 363mm Target T i 290mm Target wall thickness 36.5 mm Performance approx. 740kW speed 10 rpm Glass surface AC 365 cm2< Glass surface AT 374 cm2<

[0061] In Example 1, the initial hollow cylinder 2 was inserted into the heating furnace 12 while rotating around its longitudinal axis 10, and the initial gas pressure was set to 1 mbar and kept constant during heating. After the length of the initial hollow cylinder 2 had been uniformly heated in the heating furnace 12, the feed rate was switched on to 7.5 mm / min, and the free end of the tubing was withdrawn at the specified withdrawal speed of 7.4 mm / min. At the same time, the initial gas pressure was increased with a controlled ramp of 0.06 mbar / min until the gas pressure reached its final value of 5.3 mbar. This final value corresponds exactly to the target gas pressure calculated using equation (2). The pressure build-up phase was thus completed, and the tubing 22 assumed the desired outer diameter of 363 mm. Until then—during the pressure build-up phase—the actual diameter control via the computer 24 was not activated.Subsequently, the outer diameter control was activated to maintain the geometry within the desired range. The gas pressure served as the control variable. By inflating the originally smaller volume of the hollow cylinder's inner bore to the larger volume of the tube's inner bore, a continuous gas supply is required to maintain the gas pressure, which leads to the target outer diameter T a of the tube.

[0062] Figure 6 shows a photo of the pipe string 22 exiting the furnace. This has an essentially constant outer diameter and a smooth wall with a defect-free surface.

[0063] The diagram of Figure 7shows the corresponding curves of outer diameter and wall thickness. The y-axis represents the diameter Ta and the wall thickness d in millimeters, and the x-axis represents the length L of the drawn pipe string in meters. Curve A represents the outer diameter, and curve B represents the wall thickness.

[0064] The diagram of Figure 8 shows the gas pressure curve explained above. On the y-axis, the pressure p in mbar is plotted against time t in minutes. The pressure build-up phase is defined by period a with the initial constant pressure value of 1 mbar and the subsequent period b with a slow increase in the gas pressure to the final value of 5.3 mbar. During the pressure build-up phase, the diameter control is inactive.

[0065] Thick-walled synthetic quartz glass tubes with a large average outer diameter of 363 mm, a large average inner diameter of 290 mm, and a large average wall thickness of 36.5 mm were cut from the drawn-off tube string 22. These tubes are characterized by smooth, largely defect-free outer and inner surfaces produced by the melt flow process. Reference example

[0066] C a 220mm C i 45mm Feed VC 12 mm / min Deduction VT 66 mm / min mold tool set to 343mm Throughput 60 kg / h Target T a 343mm Target T i 330mm Target wall thickness 6.5 mm Performance approx. 800kW speed 20 rpm Glass surface AC 364 cm2< Glass surface AT 69 cm2<

[0067] The reference example represents the state of the art. The hollow cylinder used as the starting cylinder consists of undoped quartz glass melted from naturally occurring quartz raw materials using the Verneuil method. This involves melting crystalline quartz powder using an oxyhydrogen flame to create a large, elongated block. A core drill coated with diamond abrasive particles creates a central hole, and the inner and outer walls are then mechanically smoothed.

[0068] In this forming process, a forming tool is used, and the tube strand is drawn (and not compressed) from the heating zone and the softened quartz glass mass while rotating around its longitudinal axis. After the longitudinal section of the initial hollow cylinder 2 was evenly heated in the heating furnace 12, the feed rate was switched to 12 mm / min and the free end of the tube strand was set at the specified withdrawal speed of 66 mm / min. At the same time, the gas pressure was very rapidly increased at a ramp of 1 mbar / min and manually controlled. The glass is blown against the mold jaws, which are set to an outer diameter of 343 mm, using the gas pressure.

[0069] The diagram of Figure 9 shows the temporal progression of the outer diameter and the wall thickness (as Figure 7 ). It can be seen that due to the use of the forming tool, the outer diameter is very constant, despite fluctuations in the gas pressure, as can be seen from Figure 10The fluctuations in gas pressure during this forming process have no significant effect on the outer diameter and wall thickness. This is due to the fact that once the target outer diameter is reached, the gas pressure is only very roughly adjusted manually to keep the outer diameter within the desired range.

Claims

1. A method for manufacturing a tube (22) from quartz glass by forming a hollow cylinder (2) having a hollow cylinder bore (16), an outer diameter Ca, an inner diameter Ci, and a wall thickness which is at least 20 mm, by said hollow cylinder, continuously rotating about an axis of rotation (10), being fed at a relative feed rate VC to a heating zone (12), being softened therein in regions, and the softened region being radially expanded under the action of a gas pressure applied in the hollow cylinder bore (16), and a tube strand (22) having a tube outer diameter Ta, a tube inner diameter Ti, and a tube wall thickness being formed continuously from the softened region and being withdrawn at a withdrawal rate VT, characterized in that the gas pressure is used as a manipulated variable of a diameter closed-loop control (24) and / or a diameter open-loop control for the tube outer diameter or for a geometrical parameter correlated with the tube outer diameter, and in that, in a pressure build-up phase, the gas pressure is gradually increased from a lower starting value to a higher final value, and in that the following applies to the ratio of VC and VT: VT = VC ± 0.2·VC.

2. The method according to claim 1, characterized in that the gradual increase of the gas pressure in the pressure build-up phase occurs independently of the diameter closed-loop control (24) or independently of the diameter open-loop control.

3. The method according to claim 1 or 2, characterized in that the diameter closed-loop control or open-loop control is assigned a diameter target value for the tube outer diameter and / or for the tube inner diameter, and in that a target gas pressure is assigned to the diameter target value, and in that the lower starting value is in the range from 0 to 50% of the target gas pressure, preferably in the range from 10 to 30% of the target gas pressure, and the higher final value is in the range from 70 to 110% of the target gas pressure, preferably in the range from 90 to 100% of the target gas pressure.

4. The method according to claim 3, characterized in that the target gas pressure is set in the range of 2 to 20 mbar, preferably in the range of 3 to 15 mbar, and particularly preferably in the range of 4 to 10 mbar.

5. The method according to one or more of the preceding claims, characterized in that the pressure build-up phase has a duration between 1 and 120 minutes, preferably a duration between 5 and 100 minutes, particularly preferably a duration between 10 and 80 minutes, and in particular a duration between 15 and 60 minutes.

6. The method according to one or more of the preceding claims, characterized in that the gas pressure is increased in the pressure build-up phase at least temporarily with a time ramp Δp, for which the following applies: 0.01 mbar / min < Δp < 0.8 mbar / min, preferably: Δp < 0.5 mbar / min and in particular Δp < 0.2 mbar / min.

7. The method according to claim 6, characterized in that the gas pressure is increased during the entire pressure build-up phase with an average ramp Δpm over time, for which the following applies: 0.01 mbar / min < Δpm < 0.5 mbar / min, preferably: Δpm < 0.1 mbar / min and in particular Δpm < 0.06 mbar / min.

8. The method according to one or more of the preceding claims, characterized in that the following applies for the hollow cylinder (2) and the tube strand (22): 150 mm < C a < 300 mm , 30 mm < C i < 180 mm , 40 mm < C a − C i / 2 < 100 mm 300 mm < T a < 550 mm , 250 mm < T i < 450 mm , 20 mm < T a − T i / 2 < 60 mm .

9. The method according to one or more of the preceding claims, characterized in that the following applies for the hollow cylinder (2) and the tube strand (22): 4 ⋅ C i < T i < 8 ⋅ C i , C a + 100 mm < T a < C a + 300 mm .

10. The method according to one or more of the preceding claims, characterized in that the tube outer diameter Ta is in the range of 1.5 to 2.2 times the hollow cylinder outer diameter Ca.

11. The method according to one or more of the preceding claims, characterized in that the starting hollow cylinder (2) has a glass cross-sectional area AC and the tube strand (22) has a glass cross-sectional area AT, wherein the following applies: AT = AC ± 0.15·AC, preferably: 0.90·AC ≤ AT ≤ 1.15·AC, in particular: AC ≤ AT ≤ 1.10·AC.

12. The method according to claim 11, characterized in that the glass cross-sectional areas AC and AT are in the range of 250 to 1000 cm2.

13. The method according to one or more of the preceding claims, characterized in that the following applies for the ratio of VC and VT: 0.8·VC ≤ VT ≤ 1.05·VC, in particular: 0.9 ⋅ V C ≤ V T ≤ 0.99 ⋅ V C .

14. The method according to one or more of the preceding claims, characterized in that VC is set to result in a throughput of quartz glass mass of at least 30 kg / h but less than 60 kg / h and preferably less than 45 kg / h.

15. A quartz glass tube having a tube outer diameter Ta, a tube inner diameter Ti, a tube wall thickness, an outer lateral surface, and an inner lateral surface, characterized in that it has a fire-polished outer lateral surface produced by tool-free forming in the melt flow, and in that the following apply for the tube outer diameter Ta, the tube inner diameter Ti, and the tube wall thickness: 300 mm < T a < 550 mm , 250 mm < T i < 450 mm , 20 mm < T a − T i / 2 < 60 mm .