the quality aspects of laminated TAMGLASS OY glass are also briefly discussed. Jorma Vitkala* NTRODUCTORY REMARKS

Transcription

1 107 The trend, especially of architectural glass, towards ever larger sizes means that glass processors and manufacturers of tempering furnaces need to address the issues related to large sizes and coated products. From recent tests at Tamglass R&D centre, the author reports on the options for limiting or eliminating optical defects, end curving and frame effects in particular. The effects of end and edge curving on Jorma Vitkala* the quality aspects of laminated TAMGLASS OY glass are also briefly discussed. NTRODUCTORY REMARKS The popularity and use of energy-efficient windows has increased significantly in Europe during the past decade. In Northern Europe, particularly in Scandinavia, energy-efficient windows have been used for more than 25 years. The obvious reason for this is the climatic conditions. The energy crises of the seventies of course also played a major role in the move towards new window solutions. In Finland and Sweden, especially, triple-glazed window structures were introduced to make energy economy as efficient as possible. In the nineties, the advances made in glass coating technologies opened up new opportunities and double-glazed window structures with coated glass became an even better alternative. From an energy saving point-of-view, they were also more cost-efficient because of their lower weight and simpler structure.

2 108 The groundbreaker in the use of coated glass has been Germany. New heat insulation regulations were redefined there in 1995, in a way which in practice meant standardization of the use of low-e glass in windows. The German regulations are based on energy balance considerations where the heat coming through the window is taken into consideration as well. Other European countries followed at a somewhat slower pace but now it is fair to say that the use and application of low-e glass has become standard in many European countries. Low-E glass applications have increased on other continents as well. Today more than 90 per cent of all windows sold in North America are insulating glass units and that high percentage is expected to rise even further. AFG s marketing research indicates that 20 per cent of all windows in the US today make use of low-e glass and that proportion is expected to grow by more than 50 per cent over the next five years. The window structures in the Asia-Pacific area are mainly single-glazed. This market, however, will probably witness an increased demand for insulating units as air conditioning becomes standard and building codes more strict. The market potential in this region is well illustrated by the fact that only 25 per cent of all windows produced in Japan are insulating glass units. Tempered low-e safety glass has so far been used mainly in public buildings in projects initiated by different authorities. This is largely due to the introduction of safety regulations and building standards which affect the public sector rather than private housing. The location of the windows and wall structures in private houses are usually such that safety glass is not really required. High-rise offices and other public buildings are different in the sense that they often use window structures for walls which require Glass bending in the furnace White haze and optical failure Large/uneven fragmentation that safety glass be used. The use of low-e glass has helped preserve the energy efficiency of window structures and it has thus sustained the trend in office construction, which moves towards larger glass surfaces and better daylighting properties. In this development, new requirements are placed on the quality of low-e glass, since potential flaws in the glass are more easily detectable in larger sheet sizes. The processing of low-e in itself requires special attention and care, which in turn emphasizes the responsibility and capability of the tempering processes used by the suppliers of tempered low-e glass. Manufacturers of tempering machinery are thus also faced with the challenge of developing new processes which are better and more easily controlled and allow flexible production of varying glass sizes and thicknesses. Mixed production, which is so important from a competitive point of view, and the trend towards shorter delivery times, place stringent demands on streamlined cooperation between glass manufacturers, glass processors and manufacturers of tempering machinery. In evaluating the capabilities of machinery suppliers it has become increasingly important to be aware of the fact that supplying good samples may be one thing, but supplying a well functioning system at an industrial scale can be something quite different. Some of the recent test results and the deductions made possible by these samples will be looked at here. COATING/TEMPERING SUITABILITY The first stage in any tempering process concerning any type of coated glass, low-e glass included, is the analysis of the suitability of the coating to the tempering process. Some coatings are less heat resist-

3 ant and the coating layer may therefore be destroyed in the heating stage of the tempering process. Coatings in this category are mainly off-line coatings, sometimes called soft coatings. In the manufacturing of this type of glass, coating is always applied after the float process and that typically requires a low temperature. The most common method in use is vacuum coating, or sputtering. This method offers very flexible coating solutions in accordance with user requirements. Sputtered coating is usually applied for purposes of obtaining different colours and solar protection. Multi-functional coating alternatives and low-e coating are also typical applications. The other most common coating method is on-line, so called hard coating, where the coating becomes part of the float process. In this case the different coating layers are produced inside or just after the tin bath. The advantage is that the coating is created on the surface of the glass at the same temperature that applies to the tempering process and for that reason, there are no complications in making the coating withstand the tempering temperature. This kind of coating makes large production runs possible, since the properties of the coating are stable and where changes in the production runs are less desirable. The drawback is that this kind of coating gives fewer alternatives for variations of the coating. Different technologies may be used in attempting to achieve coatings which can withstand tempering. In the case of toughenable vacuum coating, however, a protective layer is designed in order to protect against the problems caused by heating. Recently, new materials/coatings which facilitate the tempering of low- E coated glass have been introduced. Sometimes it may be difficult for temperers as well From top to bottom: glass bending unstable, saddle and one direction From top to bottom: roller wave, end curving and coating burn as machinery suppliers to stay on top of what the new coatings really are good for and to take adequate time to run the required tests. SPECIAL OBSERVATIONS IN TEMPERING OF LOW-E GLASS If a clear glass is moved into a mere radiation furnace, the glass will bend strongly in the early heating stage. Bending is even stronger in the case of low-e glass. This bending is due to the different speed at which the glass surfaces heat up. The closeness of the rollers causes the bottom side of the clear glass to heat up quicker because heat transfer on the bottom side is faster in all forms of heating, radiation, conduction and free. In low-e glass this phenomenon is marked because the other side of the glass, i.e. the coated side, heats up more slowly because it cannot absorb heat as efficiently as the bottom since the coating reflects radiation. The extent of the reflection depends on the properties of the coating and the wave-length of the radiation. In the early stage at least two different bending patterns have to be taken into account in the bending of the glass. The bending of the glass focuses the total weight of the glass sheet on a very small area, which leads to white haze in the surface of the glass and possibly to viscoelastic changes. These can be observed on the surface of the glass in the form of exterior marks or optical flaws in the centre of the glass. 109 UNEVEN HEATING - A PROBLEM If the heating of the glass is uneven there are many potential problems. Some of them result in uneven breaking patterns and excessive granular particle size in the centre of the glass. This type of glass usually bends strongly and its bending is unstable, saddle-formed or curved. If the heating process is inadequately controlled, solutions are sometimes attempted by way of exces-

4 110 sive heating of the glass. If the glass temperature is too high it will cause optical flaws, a so-called roller wave, strong end curving. It may also change the properties of the coating and cause changes to the emissivity. When the temperature exceeds the settings, the coating may oxidize, thus leading to changes in both emissivity and colour. Most of the flaws analysed above are typical and result from reasonably well-known process problems which may be easily rectified. The end- and edge-bending of the glass are more serious issues. Problems may also multiply with the increase of glass sizes if the process is not entirely under control. Edge-effects of uneven heating One aspect which is seldom taken adequately into account in the heating of glass by different methods is the effect of radiation penetrating the interior of the glass through the glass edges. This adds to the effect of free in the furnace. The edge penetration leads to uneven temperatures in the glass and also to problems with the quality of the end product. Edge overheating causes excessive end and side curving to the glass. The extent of this effect depends on the furnace temperature and on the possibilities to control the radiation. The radiation heat effect is very significant since it multiplies to its fourth exponential in the heat transfer process. Typical end and edge curving Glass measuring rack The edge curving effect is a very complex problem. The best way found to approach it has been numerical modelling, since the optical temperature measurement instruments cannot give accurate values so close to the edge area. The optical flaws to the glass can be identified on the ends or edges of the glass sheet. The typical end curving effect is mm and the edge-bending is in the order of mm from the edge. These can be especially problematic in the glass lamination process. The elimination of these undesirable bending phenomena requires that heating is as even as possible, that is properly applied and that the amount of radiation is reduced. INDUSTRIAL TEST RUNS In order to obtain a proper analysis of the effects of different heating methods on the quality of coated glass, a number of test runs on an industrial scale were recently carried out at the Tamglass R & D centre. The objective of the tests was to analyse the difference between two tempering furnaces regarding the quality of the end product. The sizes of the glasses to be tested were chosen at 2,000 x 3,000 mm and the thickness was 6 mm with Pilkington K-coating. The number of test glasses was 100. The sample was this big because it enabled better loading of the furnace and deductions for industrial scale production to be drawn. The loading area of the furnace was the same in both cases, i.e. 2,440 x 5,000 mm. The first experiment was carried out in a radiation furnace where was added to facilitate even heating and the second test was run in a furnace with two chambers, the first one with low temperature and high and the second with a combination of radiation and. The edge work on the glass

5 was the same in both test runs. The test runs were carried out during continuous loading, i.e. there was no empty time in the furnace. 111 Quality control/fragmentation test At the beginning of the test some samples were broken in order to determine that the stress level was set high enough. During the continuous run some glasses were also taken out and broken for the fragmentation test (2,000 x 3,000 mm). This was done to ensure that there were no changes in the glass temperature and heating uniformity in the furnace during the test run. The fragmentation test was carried out by breaking the glass in the middle of its long side and even fragmentation was checked optically (there was no time for actual counting of the fragments because the test glasses were loaded in continuous operation). Glass bending/end-curving/roller-wave The overall bending of the glass and its repeatability was checked optically for each glass. The end-bending and roller-wave were quickly measured during the test run and quality was thus noted between each loading. After the test run the glasses were sent for further external quality check in order to obtain reliable and accurate measurement results. After the test run the bendings were measured. During measurement the glass was supported and kept in a vertical position. The bending of the glass was carefully measured at several points as shown in the figure. The roller-wave was also measured. Two-stage Combi system One-chamber radiation furnace with forced system Test results/radiation furnace The first test was carried out in a one-chamber radiation furnace including a forced system to ensure even heating. At the beginning of the test, a couple of test glasses were used to find the right settingvalues. After that the glasses were inserted in continuous production. The test results show how the bending of the glass, the end-bending and the roller-wave changed during the production run. The end-bending varied from mm and the overall bending, lengthwise from 0 to 3 mm and width-wise from 0-3 mm. The average was around 2 mm and the roller wave was within mm. The test results show that a good quality level can be achieved with a one-chamber radiation/ furnace, but that the settings and quality control have to be well controlled, particularly in the case of large low-e glass. Changes are also required in the system inside the furnace. Combi system The second test was done in a double-chamber furnace. Here, the temperature of the first chamber was 430 C and heat transfer was by high. The second chamber was a radiation/ furnace. At the beginning of the test run the correct temperature settings had to be tested on some glasses, and the amount of also needed to be controlled on the top as well as on the bottom side. The test glasses immediately showed the effect of improved quality and particularly a major reduction of end-bending. In practical terms, the end-bending was eliminated during the test run and it could no longer be detected visually either.

6 112 The typical test results were an end-bending of less than 0.05 mm and not really measurable at all in any of the test glasses, an overall glass bending length-wise of mm for an average of around 1 mm, a width-wise bending of mm for an average of around 2.8 mm and a roller-wave of maximum mm. The quality and repeatability of the glass remained steady and there was no need to change the settings of the furnace during the production run. Quality aspects on laminated glass The test results referred to above raised a number of important considerations, especially in the case of laminated glass. Lamination of tempered or heat-strengthened (HS) glass has become more common in recent years because it makes laminated glass more resistant to heat shocks, impact and bending. This enables designers to use thinner glass structures particularly in applications where the glass needs to stay in its frame if it breaks. If the tempered or HS glass needs to be laminated, special attention must be paid to optical quality, bending and especially to the end-curving of the glass. Identical duplication of the process is also essential if both glasses are tempered. The end-bending can vary a lot in different types of coated glass and the thickness of the glass also has an influence. The tempering process and the end quality of the products are very much influenced by the quality of the edge-grinding carried out in the pre-processing stage. Glass edge overheating by radiation and free Direction of end curving in the case of laminating THE IMPORTANCE OF CONSISTENCY Laminated tempered glasses should always be from the same production run to ensure that the quality of the glass is consistent. Simultaneous tempering also ensures that the conditions in heating and cooling are as identical as possible, which makes the multiplication of production runs easier. The same objective is facilitated if the same operator, with a professional and experienced view of the tempering stage, takes care of that phase of the process as well. It is also important that the loading directions and position are the same for both glasses. The glass should also come from the same manufacturer and the same production batch, because glass, even if it comes from the same manufacturer, may vary during production. In fact it is quite normal that the float glass and coatings can vary in different production runs. It should be observed, however, that the differences arising from manufacturing technology are minimal and that the biggest influence on the glass comes from the operator s parameters, set-points, heating control and from the pre-processing stage (edge grinding, seaming and grinding etc.). This places certain requirements on the process. In lamination, it is vital to ensure that the end-bending of the glass goes in the same direction. If there is end-bending of the glass, it may easily result in air bubble problems in the end section of the glass, unless the glasses are positioned in lamination so that the

7 FINAL HEATING OF GLASS THICKNESS 3 mm end-bendings do not match one another. A thicker film should also be used for lamination. The quality of laminated glass is also strongly influenced by the operations of the pre-pressing machinery. It is very important that the glass and the PVB-film are heated to the same temperature, in order that excessive heating of the edges and end section be avoided. There are different methods available for the heating of film and glass. The first is infrared long-wave radiation. This heats the sandwich formed by the laminated glasses and the film so that the glass surface is first heated up and the heat is transferred through the glass and film by heat conductivity (k). In this case, random radiation and free may contribute to overheating the edge areas of the glass. Another is short-wave radiation, which is used to heat the film between the glasses directly, since the short-wave radiation passes through the glass and the major part of the energy is absorbed by the film. The problem here may be the overall temperature in the pre-pressing furnace, which becomes too high because of random radiation and reflection from the glass surface. As the furnace temperature rises, the share of long-wave radiation from the walls of the furnace, the rollers etc. increases. In combination with free, this causes over-heating of the edges. Furthermore, the high- heating methods have proven to be the most efficient way to heat up the whole sandwich. Glass heating speed at the end of the heating cycle Glass sandwich heating by low temperature and high The specific advantage of heating in the pre-pressing unit is that the whole sandwich bundle to be laminated, glasses as well as film, are heated up to the same temperature. This is done by applying high- systems with air jets, which cause the heat from the glass surface to penetrate the whole sandwich by conduction. Since the overall temperature is very low (70-90 C), heat transfer from radiation plays a minor role. Its share is reduced to the fourth exponential of the temperature and thus there is no overheating of the edges. Finally, suction in the autoclave is also one of the best methods for heating up the sandwich. The problem here may be that this method slows down the process quite a lot. CONCLUSIONS AND SOURCES FOR QUALITY IMPROVEMENT As is evident from the test results, in the twochamber furnace, end curving and roller wave particularly were significantly reduced. In fact they were so small that they could not even be recognized. This quality improvement was due to the fact that, in the combination furnace of the testing procedure described above, the temperature of the first furnace was only 430 C and the heating method was high-. The heat transfer from becomes decisive because the temperature of the furnace was low, and in addition, the materials within the furnace had fairly weak radiation prop- 113

8 114 erties. The minimization of radiation in the first furnace proved to be crucial to end-product quality. The transfer of radiation heating was minimized, because the temperature was lower and the share of radiation heat transfer will be reduced to the fourth exponential of the temperature. When the beginning of the heating stage is well controlled and the glass heats up as evenly as possibly from the beginning, no major temperature differences between the edges and the centre of the glass can form. Because of this even heating, the glass may be taken out of the furnace at a lower temperature, which improves its overall quality. Naturally, the quality of the edge grinding also has a major influence on the temperature at which the glass may be removed from the furnace. In a two-stage furnace the temperature is higher in the second chamber than it would be in a high temperature/ furnace. This higher temperature contributes to quicker heating of the glass to the final tempering temperature. When low-e glass has been heated up in the first furnace, the emissivity of the surface also changes, and it becomes easier to apply radiation heating. The edge effect, i.e. over-heating, may be efficiently controlled by air jets and heating profiles. The reduction in radiation had a decisive impact in reducing the overheating of the glass edge and thus on the quality of the endproduct. Radiation enters the glass via the edge which becomes overheated. The higher the furnace temperature, the more marked this phenomenon will be. The effect of radiation is much greater in high-temperature furnaces and there are no control facilities available for this. Radiation heat transfer is, of course, even higher in radiation-type furnaces than in hightemperature furnaces. In a radiation furnace with a special system, however, it is possible to achieve very good product quality. This is due to the fact that the heat transfer in the furnace can be controlled. Tamglass HTF ProE tempering furnace with new and heating profile systems In older types of radiation furnaces, the edge effect, i.e. edge overheating, can be compensated by heater/temperature profiles. The latest methods, with the Tamglass HTF ProE tempering system allow the control of temperature both with heating and profiles, since the channels are placed in the direction of the heating elements. This new way of combining radiation with heating and profiles and movement control facilitates the advent of a newer and better quality standard for the tempered glass market. REFERENCES 1. Jorma Vitkala, Low-E glass tempering, Glass Processing Days Proceedings Book Roger Kennedy, The History and Future of the Flat Glass Industry, Glass Processing Days Proceedings Book Jorma Vitkala, What Kind of Coated Glass Can Be Tempered?, Glass Processing Days Proceedings Book This paper will be presented at the Sixth International Glass Processing Days Conference in Tampere, Finland, June *Chairman of the GPD 99 Organizing Committee TAMGLASS - FINLAND Information Service no. 19 See coupon on the last page