Device and method for cooling a component contacting a glass melt

Abstract

The present disclosure relates to a method for cooling a component of a glass melting plant that contacts a glass melt, the corresponding cooling device, as well as the system of the cooling device and the cooled component itself. The method provides that a pipe with an open pipe end at least on one pipe section is introduced into an open cavity in the component with the formation of a peripheral annular space, and a cooling medium is introduced through the pipe into the cavity and is deflected at the base of the cavity, flows back in the annular space, and flows out of the cavity. In its pipe section introduced into the cavity, the pipe has a constriction and has perforations through the pipe walls in the region of the constriction, whereby the cooling medium is accelerated in its passage through the constriction in the inside of the pipe, and a portion of the cooling medium flowing back from the annular space is aspirated into the inside of the pipe.

Claims

1. A method for the cooling of a component of a glass melting plant that contacts a glass melt, the component having at least one open cavity, the method comprising the steps of: introducing a pipe with an open pipe end into the cavity in the component so that a peripheral annular space is formed between the outer surface of the pipe and the inner surface of the cavity, while maintaining an axial distance S between the pipe end and the base of the cavity; and during the operation of the glass melting plant, introducing a cooling medium through the pipe into the cavity so that it is deflected at the base of the cavity, flows back in the annular space, and flows out of the cavity, wherein, in its pipe segment introduced into the cavity, the pipe has a constriction, and in a region of the constriction has perforations through the pipe walls, whereby the cooling medium is accelerated in its passage through the constriction in the inside of the pipe, and a portion of the cooling medium flowing back from the annular space is aspirated into the inside of the pipe.

2. The method according to claim 1, wherein, except in the region of the constriction, the pipe has an internal cross-sectional surface area A.sub.I and the cavity has a cross-sectional surface area A.sub.H perpendicular to the pipe, wherein A.sub.H corresponds to 1.5 times to 4 times A.sub.I.

3. The method according to claim 1, wherein, except in the region of the constriction, the pipe has an internal cross-sectional surface area A.sub.I and has an internal cross-sectional surface area A.sub.E at the narrowest place of the constriction, wherein the following condition is fulfilled: A.sub.E0.04.Math.A.sub.I.

4. The method according to claim 3, wherein the pipe has a circular cross section with a diameter D.sub.E=2.Math.{square root over (A.sub.E/)} at the narrowest place of the constriction.

5. The method according to claim 1, wherein the cavity has a cross-sectional surface area A.sub.H perpendicular to the pipe and a diameter D.sub.H=2.Math.{square root over (A.sub.H/)}, and in that the distance S corresponds to 0.1 times to 0.8 times the cavity diameter D.sub.H.

6. The method according to claim 1, wherein the cavity is formed by a blind hole.

7. The method according to claim 6, wherein, except in the region of the constriction, the pipe is configured cylindrically, at least in sections, and has an inner diameter D.sub.I in the cylindrical section, in that the blind hole has a diameter D.sub.H, and in that the following applies: D.sub.H={square root over (1.5)}.Math.D.sub.I to 2.Math.D.sub.I.

Description

SUMMARY OF THE FIGURES

(1) FIGS. 1a and 1b show a first embodiment of the system according to the present disclosure in the form of a cooled overflow wall.

(2) FIGS. 2a and 2b show a second embodiment of the system in the form of a cooled brick wall.

(3) FIG. 3 shows a view of a detail of the system having a first embodiment of the pipe.

(4) FIG. 4 shows the same view as FIG. 3 for explaining the pressure relationships in the pipe and in the cavity.

(5) FIG. 5 shows a view of a detail of the system having a second embodiment of the pipe.

(6) FIG. 6 shows a cross section through a test block for determining the cooling power.

(7) FIG. 7 shows a measurement plot of the temperature as a function of the blower throughput determined at a first position of the test block according to FIG. 6.

(8) FIG. 8 shows a measurement plot of the temperature as a function of the blower throughput determined at a second place of the test block according to FIG. 6.

(9) FIG. 9 shows a measurement plot of the temperature as a function of the blower throughput determined at a fourth place of the test block according to FIG. 6.

DETAILED DESCRIPTION OF THE DISCLOSURE

(10) In two views, FIGS. 1a and 1b show a first embodiment of the system according to the present disclosure with a component 12 that guides or contacts a glass melt 10. FIG. 1a shows a sectioned lateral view, in which it can be recognized that the component 12 forms an overflow wall 100 that divides a melting tank or glass melting plant 13 into two sections 102 and 104. The melting tank 13 is bounded by a tank bottom 106 and side walls 108, which are in contact with the melt 10, as is overflow wall 100 also, and thus these are usually formed from refractory stone. From the bottom 106 of the tank, the overflow wall 100 projects up to just below the glass surface. In other words, the melt 10 covers the overflow wall on its upper side or top, the so-called wall crown. At this place, the melt has the highest flow rate within the melting tank, for which reason, the refractory material at the wall crown is subjected to the greatest wear. Therefore, one attempts to achieve a cooling that is as effective as possible at precisely this place.

(11) For this purpose, the component 12 has a total of five cavities 14 open from the bottom lying opposite to the wall crown. As can be recognized in the sectioned front view 1b, the cavities are distributed equidistantly over the entire width of the overflow wall 100. In each of the cavities 14 is introduced a separate pipe 16 (only shown in FIG. 1a by way of example), through which a cooling medium, in particular a gas, and more preferably air, is introduced into the cavity 14 during the operation of the glass melting plant 13. For this purpose, all pipes 16 are connected to a supply line 17, which is connected on the input side to a flow generator 18, preferably in the form of a blower. Furthermore, control elements that are not shown, such as, for example, a valve or throttles, or the like can be provided along the supply line 17 for control of the air flow that is introduced.

(12) FIGS. 2a and 2b show two sectioned views of another embodiment of the system according to the present disclosure. The component 12 here is a bridge wall, which once again divides the glass melting tank 13 into two sections 102 and 104. With its two side parts 112, 114 and the cross beam 116, together with the bottom 106 of the tank, the bridge wall 110 forms a window 118, also called a through-flow opening, which is immersed in the melt 10 completely below the glass surface. The window 118 forms a cross-sectional constriction opposite the tank cross section, which increases the flow rate in this region. For this reason, an intensified wear of the refractory material ensues on all four sides of the window opening, i.e., on both side parts 112, 114, the cross beam 116 and the tank bottom 106. Correspondingly, a cooling is provided here in all four sides of the component.

(13) In each of the two side parts 112 and 114 of the bridge wall 110 is found for this purpose a cavity 14 in the form of a blind hole, and a pipe 16 is introduced into each of these holes. In an entirely similar manner, two openings 14 in the form of blind holes, into each of which is inserted a pipe 16, are found in the bottom wall 106. Once again, the pipes 16 and 16 are supplied with air via a pipeline 17 and flow generator 18.

(14) The cross beam 116 has a different construction for the cooling in the form of two continuous boreholes 54 through which flows a cooling medium, preferably air, crosswise to the alignment of the tank, from one side to the other.

(15) Both embodiment examples discussed in connection with FIGS. 1A, 1B and 2A, 2B are considered to be schematic with respect to the configuration, number and arrangement of the cavities or blind holes and pipes. In particular, depending on the cooling requirement in each case, more or fewer cavities and pipes can be provided, also in different configuration and size and in another arrangement. The details of the cavity 14 and of the pipe 16 are explained below on the basis of FIGS. 3 to 5.

(16) FIG. 3 shows an excerpt from the system according to the present disclosure, comprising the component 12 in cross section through a cavity 14, which is configured as a blind hole 38 and into which is introduced a pipe 16 by way of a pipe section 21 with the formation of the annular space 22 between an outer surface 24 of the pipe 16 and an inner surface 26 of the cavity 14. Component 12 may be, for example, an overflow wall 100 according to FIGS. 1A, 1B, a bridge wall 110 or a tank bottom 106 according to FIGS. 2A, 2B, or another tank component that is in contact with the glass melt. The arrangement in FIG. 3 is shown only schematically and not to scale. In particular, the length-to-diameter ratio of the blind borehole and the pipe 16 are not to scale. Usually, the blind hole 38 and the pipe 16 are deeper or longer in relation to their diameter than they are represented in the illustration of the drawing. The depth or length may vary considerably. These depend in turn on the thickness of the component that is to be penetrated or on the orientation of the cavity in the latter.

(17) FIG. 4 shows the same view of the same arrangement as FIG. 3; it contains only additional technical data. Among other things, a distance S is depicted therein, which is maintained between pipe end 20 on the end side of the introduced pipe 16 and the base 28 of the cavity 14 or of the blind hole 38 on the end side. In order to assure in a simple way that the distance is maintained with the required precision, a spacer 29 is preferably provided between the pipe end 20 and the base 28 of the cavity 14 on the end side. More preferably, this spacer can be held in place by the pipe 16.

(18) Adjacent to the pipe end 20, the pipe 16 has a peripheral constriction 30. The reference number 30 characterizes the entire region of the constriction 30, which is formed by a section 31 tapering in the flow direction, downstream of a cylindrical section 32 and an end section 33 widening in the flow direction further downstream. The directional data in this regard refer to the flow direction of the cooling medium inside pipe 16. With this geometry, the cylindrical section 32 forms the narrowest place 34 of the constriction 30. In the cylindrical section 32, the pipe 16 has perforations or boreholes 35 through the pipe walls 36, which fluidically connect the inside 37 of the pipe to the annular space 22.

(19) In the following, the flow of air through the arrangement will be explained, which is characterized by arrows in the figures. The cooling medium (usually air) 40 that is introduced into the pipe 16 flows upward in the pipe in the axial direction, as is characterized by the arrow 41. Within the constriction 30, the flow rate of the air is increased based on the cross-sectional constriction, until it reaches its maximum at the narrowest place 34. At the same time, the stationary pressure within the pipe is minimal at this place. Up to the pipe end 20, the air has again attained the initial rate, since here the pipe diameter corresponds to the diameter of the rest of the pipe before the constriction 30. At the pipe end 20, the air exits on the end side and is deflected from the opposite-lying base 28 of the blind hole 38, as arrows 42 illustrate. The air flows back in the annular space 22 and first passes through a widened region of the annular space 22, which coincides with the region of the constriction 30 and is designed to be complementary to this, since the pipe walls 36 are of equal thickness over the entire length of the pipe 16 and the cavity 14 is a cylindrical blind hole 38. In the widest region of the annular space 22, the flow of air further slows down, whereby the stationary pressure increases in the annular space until it reaches a maximum at the radially narrowest place 34 of the pipe 16, which is adjacent to the widest place of the annular space 22. Here, a difference in pressure arises between the inside 37 of the pipe and the annular space 22, which is responsible for the fact that a portion of the air flowing back from the annular space 22 is aspirated through the perforations or boreholes 35 into the inside 37 of the pipe, as indicated by the arrows 44. The circulated flow of air that is intensified in the region of the pipe end 20 leads to an increase in the local heat transfer coefficient and thus to a stronger cooling effect.

(20) Furthermore, the air flows back again in the annular space 22, which is symbolized by the arrows 46, and flows out of the cavity at the lower end of the component 12, as the arrows 47 illustrate.

(21) The pressure relationships inside and outside the pipe are qualitatively considered in the following, on the basis of FIG. 4. Since the exemplary embodiment involves the ideal case of a rotationally symmetric geometry, thus a pipe 16 and a blind hole 38 which are each present with circular-round cross sections, the calculations can be referred back directly to the real diameters indicated in the figure. If rotational symmetry were not present, instead of the real diameters, equivalent diameters D would be used, which can be calculated from the real cross-sectional surface areas A as D=2.Math.{square root over (A/)}.

(22) The pipe 16 is configured as cylindrical, except in the region of the constriction 30, and has an inner diameter D.sub.I in the cylindrical section. The blind hole has a diameter D.sub.H. The most favorable relationship of the cross sections or diameters as a compromise between cooling power and pressure loss was found in a region D.sub.H={square root over (1.5)}.Math.D.sub.I to 2.Math.D.sub.I with an optimum at D.sub.H={square root over (2)}.Math.D.sub.I. As described above, for the pressure loss, the matching of the hydraulic diameters of the pipe 16 and of the annular space 22 would be advantageous, and for the heat transfer, a flow rate in the annular gap 22 that is as high as possible would also be advantageous, but high preliminary pressures are associated therewith and thus high capital and operating costs. An approximate matching of the flow cross sections of the pipe 16 and of the annular space 22 has turned out to be advantageous. For circular cross sections, this describes the above optimum. while disregarding the wall thickness of the pipe.

(23) The distance S between the pipe end 20 and the base 28 of the blind hole 38 advantageously is from 0.1.Math.D.sub.H to 0.8.Math.D.sub.H, wherein under the assumption D.sub.H={square root over (2)}.Math.D the optimum lies at approximately S=0.35.Math.D.sub.H. These proportions have been demonstrated to be optimal, in particular with respect to a pressure loss that is as small as possible in the region of the flow deflection, thus between pipe end and blind-hole base. As stated above, the measurement results from the selection of similar flow cross sections in the region of the flow deflection. If the gap or distance S is selected too large, the heat transfer at the end surface area is poorer; if the gap is selected too narrow, the pressure loss increases and so does the required preliminary pressure, which drives up the costs of the cooling.

(24) The following considerations are relevant for the dimensioning of the constriction 30: The circled points 1 to 6 depicted in FIG. 4 mark the places where the pressure values with corresponding indices that are drawn on in the following are assumed. p1 thus stands for the pressure at the position 1, etc. Pressure differences between two adjacent points are indicated with appropriate double indices. p12 accordingly stands for the pressure difference between point 1 and point 2, i.e., p12=p1p2, etc. If a pressure loss is present from 1 to 2, then p12 is positive.

(25) Thus, for the effect according to the present disclosure to occur at all, according to the above considerations, first the following must be valid:
p5>p2.
This relation can also be represented by means of the pressure differences as:
p1p12p23p34p45>p1p12
Equivalently rearranged, one obtains:
p23p34p45>0.

(26) In this case, p34 is the pressure loss from 3 to 4 based on the flow deflection. This can be determined empirically by tests or flow simulations and is also designated below as A.sub.pu. Accordingly, the following is thus valid:
p23p45>pu.(1)

(27) The pressure differences p23 and p45 can be converted to the following based on the Bernoulli equation:

(28) $\begin{array}{cc}p23=p2-p3=\frac{}{2}.Math.\left(v_3^2-v_2^2\right)& \left(2\right)\end{array}$
or

(29) $\begin{array}{cc}p45=p4-p5=\frac{}{2}.Math.\left(v_5^2-v_4^2\right).& \left(3\right)\end{array}$

(30) Here, .sub.x designates the rate at the position x, which, as is known, is proportional to the volumetric flow {dot over (V)} divided by the flow cross section A.sub.x at the corresponding position: .sub.x={dot over (V)}/A.sub.x. Here, if the latter is inserted in Eqs. (2) and (3) and this in turn is inserted in Eq. (1) and rewritten, the result is:

(31) $\begin{array}{cc}-\frac{1}{A_3^2}+\frac{1}{A_2^2}-\frac{1}{A_5^2}+\frac{1}{A_4^2}>\frac{2}{{\stackrel{.}{V}}^2}\mathrm{pu}.& \left(4\right)\end{array}$

(32) This relation has general validity for the entire subject of the present disclosure and, in particular, for all preferred surface area relationships of A.sub.H=1.5.Math.A.sub.I to A.sub.H=4.Math.A.sub.I and permits determining, with given pipe cross section A.sub.I=A.sub.3, given cavity cross section A.sub.H=A.sub.3+A.sub.4+A.sub.W and taking into consideration the wall thickness A.sub.W of the pipe at that axial position, the maximum cross-sectional surface area of the pipe at the narrowest position of the constriction A.sub.E=A.sub.2 or the minimum cross-sectional surface area of the annular space at the corresponding axial position A.sub.5.

(33) Relation (4) is always fulfilled for any small values of A.sub.E=A.sub.2, which also corresponds to theoretical expectations, yet the Bernoulli effect continually increases with the cross section A.sub.E becoming smaller. Of course, counter to this is the fact that any narrow constriction in practice is accompanied by a no longer tolerable pressure loss at any time over the entire flow path, since the cooling then could only still be driven with a very high initial pressure, and therefore would not be economical. For this reason, for practically relevant pressure relationships, a minimum cross-sectional surface area of the pipe at the narrowest position of the constriction can be indicated by A.sub.E=0.04.Math.A.sub.I or a corresponding diameter ratio D.sub.E=0.2.Math.D.sub.I as the lower limit.

(34) In the exemplary embodiment with the geometric condition A.sub.H=2.Math.A.sub.I, Equation (4) is further simplified. The condition A.sub.H=2.Math.A.sub.I rewritten to the indices 3 and 4 with A.sub.I=A.sub.1=A.sub.3 and with A.sub.H=A.sub.3+A.sub.4, neglecting the wall thickness of the pipe, simply means A.sub.3=A.sub.4. This is inserted in Eq. (4) and rewritten for the indices E and H, and taking into consideration the wall thickness A.sub.WE of the pipe in the region of the narrowest constriction with A.sub.2=A.sub.E and A.sub.5=A.sub.HA.sub.EA.sub.WE, one obtains:

(35) $\begin{array}{cc}\frac{1}{A_E^2}-\frac{1}{{\left(A_H-A_E-A_{\mathrm{WE}}\right)}^2}>\frac{2}{{\stackrel{.}{V}}^2}\mathrm{pu}.& \left(5\right)\end{array}$

(36) For the pressure loss due to the deflection at the base of the cavity, in the case of the indicated geometric conditions including a setting depth S=0.35.Math.D.sub.H, an estimate of a doubled 90 deflection of a pipe flow has been demonstrated to be sufficiently accurate, from which a value of 50 mbar results for the term pu. Thus, based on the relation (5), in this example, an estimate for the maximum cross section A.sub.E at the narrowest position 2 can be indicated.

(37) All of the above considerations do not take into account dissipative effects and in this regard represent an approximation for the design of the constriction. However, they show the boundary conditions under which the basic principle operates. If one wishes to optimize the geometry of the constriction, i.e., to further increase the efficiency, then it is recommended that a flow simulation be carried out, which can take into consideration the real geometric parameters and also surface effects.

(38) In the exemplary embodiment shown in FIGS. 3 and 4, the region of the constriction 30 is configured in such a way that the conically tapering region 31 and the conically widening region 33 are equally long and mirror-symmetrical in the axial direction. This geometry is just as compelling as very generally the proportions of the constriction, the course of the wall of the pipe, or, for example, the arrangement of the perforations. It is crucial that the geometry can give rise to a pressure difference of p5>p2 and thus the effect based thereon. Nevertheless, by a variation in the detail geometry of the pipe and/or of the cavity inside and outside the constricted region 30, such as, for example, that of the length of the constricted region 30 and/or that of one or both angles of inclination of the tapers or widenings, an optimizing of cooling power and pressure loss could be effected.

(39) The width variation that will be given for the concrete configuration of the pipe shall be explained on the basis of FIG. 5, which is discussed in the following. This figure shows an alternative embodiment of the pipe 16, which is differentiated from the pipe 16 of the configuration described with reference to FIGS. 3 and 4 by a modified region of the constriction 30. This region 30 has a section 31 that tapers in the flow direction, subsequently a cylindrical section 32, and subsequently thereto an abruptly expanded cylindrical section 50, which again has the initial cross section of the pipe. The cylindrical section 32 forms the region of the narrowest constriction 34, in which the flow rate is again maximal and the stationary pressure is minimal. Slot-shaped perforations 35 through the walls 36 of the pipe 16 that are oriented in the flow direction are provided in the cylindrical section 32. In the transition to the cylindrical section 50, the flowing gas does not expand abruptly, but rather equally uniformly, as in the above-discussed exemplary embodiment. For this purpose, these slots are open further away at the front end of the cylindrical section 32.

(40) Only the part of the cylindrical section 32 that is not surrounded radially by the additional section 50 functionally forms the region of the narrowest constriction 34. The slots begin in this region of the narrowest constriction 34 in the flow direction in front of the cylindrical section 50 and at the same time form here the perforations 35, through which the cooling medium can flow radially into the inside 37 of the pipe from the annular space 22 due to the stationary pressure difference, as described previously.

(41) For purposes of mechanical fastening of the additional cylindrical section 50 to the narrower cylindrical section 32, several spacer elements 52 (three here) that join together the two sections 32 and 50 are arranged at a distance radially and are peripherally interposed therebetween.

(42) FIG. 6 shows a sectional view of a test block 12, which has a cylindrical shape. Centrally, thus coinciding with the cylinder axis, a cavity 14 in the form of a blind hole 38 is found in the test block 12. A plurality of measurement boreholes 61, 62, 63, 64, 65 and 66 are found both on the outer periphery as well as on the end side opposite the base 28 of the blind hole 38.

(43) The test block has an outer diameter of 240 mm; the blind hole 38 has a diameter of 65 mm. The measurement boreholes 61, 62 and 64 are deep boreholes that are close to the blind hole 38, except for a distance of 10 mm in each case, and two of these are radial and one is axial. The measurement boreholes 63, 65 and 66 are surface boreholes, each of which only projects 10 mm from the outer side into the test block. Two of these are also arranged radially and one is on the end side. For the practical determination of the cooling power, a thermocouple was inserted into each of the test boreholes 61 to 66 for the temperature measurement. The thermocouples assigned to the test boreholes 61 to 66 bear designations in the same sequence: TE_1, TE_2, TE_3, TE_4, TE_5 and TE_6. Another thermocouple TE_A was arranged below the blind hole 38 at the position 70 for the measurement of the temperature of the exhaust air.

(44) For purposes of the test operation, a pipe according to FIG. 3 was inserted from below into the blind hole 38 of the test block 12 and held centrally therein. The setting depth, i.e., the distance S between the upper end 20 of the pipe and the base 28 was freely adjustable and was varied during the tests. The pipe was connected to a blower and was acted on with varying volumetric flows of the cooling medium, air, during the tests.

(45) For purposes of comparison, a straight cylindrical pipe without constriction, but otherwise with the same dimensions and geometry was used. This pipe is called a standard pipe in the following. It was inserted in the same way from below into the blind hole 38 of the test block 12, held centrally therein, and operated with the identical setting depths and volumetric flows.

(46) All of the following diagrams contain measurements with both pipes, in each case for three different setting depths: 15 mm, 25 mm and 35 mm.

(47) The furnace was held constant at 1480 C. during the test time. The pipe was provided with cooling air by a blower having a variable volumetric flow. In the diagrams discussed below, measurement values are reproduced with volumetric flows of the blower of 100 m.sup.3/h, 150 m.sup.3/h, 200 m.sup.3/h and 250 m.sup.3/h. The measured temperature values given in the following were determined after reaching a temperature equilibrium.

(48) The results of the measurements with the pipe according to the present disclosure according to the representation in FIGS. 3 and 4 are denoted as Venturi in the legends of the diagrams discussed in the following. The comparative measurements with the pipe are denoted as standard in the legends of the diagrams. The setting depths are also indicated in the legends of the diagrams, with the numerical notation 15, 25 and 35 at the end.

(49) The diagram according to FIG. 7 reproduces the temperature curve in C. as a function of the blower throughput in m.sup.3/h, measured by the thermocouple TE_3 in the surface borehole 63 on the end side of the test block 12. This measurement point represents the temperature of the component next to the exposed end side that is subjected to the most wear.

(50) From the curves of the diagram, it can be discerned that a cooling effect that increases both with increasing air throughput and decreasing setting depth can be obtained even with the standard pipe. Independent of the setting depth and the blower throughput, however, the cooling of the outer surface of the end side of the test block 12 is on average about 5 C. greater when it is brought about by the use of the pipe according to the present disclosure having a peripheral constriction when compared with the otherwise identical arrangement with the cylindrical standard pipe. At the same time, the principle of the constriction, unlike an additional reduction in the setting depth or increase in the air throughput, however, does not bring about an essentially additional pressure loss, and thus has a significantly improved efficiency. Since it is precisely at this place that the thermal and mechanical stresses of the component are the greatest, the cooling precisely contributes to the increase in the service life of the component.

(51) The diagram according to FIG. 8 reproduces the temperature curve in C. as a function of the blower throughput in m.sup.3/h, measured by the thermocouple TE_4 in the deep borehole 64 on the end side of the test block 12. This measurement point represents the temperature of the component near the cooling air flow inside the block. The measured temperature once again was approximately 50 to 75 Kelvin higher when the standard pipe was used, as compared to when the pipe according to the present disclosure was used, independent of the setting depth and the blower throughput. The diagram in FIG. 8 is thus suitable for confirming the finding from the previous diagram, wherein the absolute greater difference in temperature at this place, of course, is to be attributed to the fact that this measurement point is close to the flow of cooling air.

(52) The diagram in FIG. 9 represents the temperature measurement in the deep measurement borehole 61 near the blind borehole 38 by means of the thermocouple TE_1. The measurement shows that at this measurement place that is the furthest down axially, a significantly improved cooling cannot be achieved with the use of the pipe according to the present disclosure. This confirms the assumption that an improved cooling occurs only in the region of the pipe end 20 or the base 28 of the blind hole, since an increased circulation of the cooling air is brought about there with the use of the pipe according to the present disclosure The cooling is therefore only increased locally where it is also required most in practice. This particularly confirms the effectiveness of the cooling according to the present disclosure.

(53) Also, the temperature of the discharged air at the measurement point 70 was monitored by means of thermocouple TE_A, whereby it could be confirmed that an increased cooling resulted with the use of the pipe according to the present disclosure. The exhaust air temperatures when the pipe according to the present disclosure was used were found to be reproducibly greater than the exhaust air temperatures of the standard pipe with otherwise equal conditions, which can be concluded to be due to a greater removal of heat.

(54) From all of the measurement data, it can be further discerned that a lesser setting depth is basically better with respect to the cooling power than a greater depth. This is valid in any case down to a certain minimum distance in which in reality, the pressure loss of the entire system begins to increase over-proportionally. If one considers the distances of the curves with different setting depths, it is seen that with the present system (D.sub.H=65), a reduction from 35 mm to 25 mm can still cause a significant increase in the cooling power; however, a change to 15 mm no longer brings about this effect. A good compromise between removal of heat and pressure loss is accordingly a setting depth of <35 mm. If one wants to further increase the cooling power, taking into the bargain a higher pressure loss, a setting depth of <25 mm is to be preferred, but which then may make necessary an increased power of the blower under certain circumstances. Furthermore, it should be taken into consideration that tolerance deviations in the assembly and manufacture of the pipe and of the blind hole will be made more noticeable in the case of small setting depths.

(55) While the present disclosure has been described with reference to one or more particular embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope thereof. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the disclosure without departing from the scope thereof. Therefore, it is intended that the disclosure not be limited to the particular embodiment(s) disclosed as the best mode contemplated for carrying out this disclosure.

LIST OF REFERENCE CHARACTERS

(56) 10 Glass melt 12 Component 12 Test block 13, 13 Glass melting plant 14, 14, 14 Cavity 16, 16, 16,16 Pipe 17, 17 Supply line 18, 18 Flow generator/blower 20, 20 Open end of pipe 21 Pipe section that can be introduced/is introduced 22, 22 Annular space 24 Outer surface of the pipe 26 Inner surface of the cavity 28 Base of the cavity 29 Spacer 30, 30 Constriction 31, 31 Tapering section of the pipe 32, 32 Cylindrical section of the pipe 33 Widening section of the pipe 34, 34 Narrowest place of the constriction 35, 35 Perforation 36, 36 Pipe walls 37, 37 Inside of the pipe 38 Blind hole 40 Cooling medium 41 Flow arrow 42 Flow arrow 44 Flow arrow 46 Flow arrow 47 Flow arrow 50 Additional cylindrical section 52 Spacer element 54 Borehole 61-66 Measurement borehole 70 Temperature measurement point 100 Overflow wall 102, 102 Section of the melting tank 104, 104 Section of the melting tank 106, 106 Tank bottom 108 Side wall of the melting tank 110 Bridge wall 112 Side part 114 Side part 116 Cross beam 118 Window S Setting measurement D.sub.I Inner diameter of the pipe D.sub.H Diameter of the blind hole D.sub.E Diameter of the pipe at the narrowest place TE_1-TE_6 Thermocouple TE_A Thermocouple