Potential methods of measuring the stirring intensity during secondary steel making in the ladle furnace

(1)

IN

DEGREE PROJECT TECHNOLOGY,

FIRST CYCLE, 15 CREDITS ,

STOCKHOLM SWEDEN 2016

Potential methods of measuring

the stirring intensity during

secondary steel making in the

ladle furnace

(2)

(3)

Abstrakt

Syftet med detta arbete var att hitta en potentiell mätmetod för att bestämma gasomrörningsintensiteten under skänkbehandling i ståltillverkning. Studien har till mestadels fokuserat på att hitta mätmetoder för att hitta den faktiska mängden gas som kommer in i skänken, då detta behövs för att kunna bestämma omrörningsintensiteten. Gasomrörningsintensiteten är i sin tur viktig eftersom stålverken vill ha en möjlighet att koppla kvaliteten och sammansättningen av stålet till gasomrörningsintensiteten.

Metoderna för att utföra denna studie baseras på en litteraturstudie och en fältövning, där fem olika Svenska stålverk besöktes. Dessa företag var SSAB, Uddeholms AB, Sandvik, Ovako och Outokumpu. Litteraturstudiens huvudsakliga avsikt var att titta på tidigare experiment som gjorts med liknande syften samt undersöka andra processindustrier där mätmetoder för gasflöden används. Eftersom de fysiska förutsättningarna på stålverk är relativt extrema (med avseende på temperatur och ljudnivå) i jämförelse med andra industrier, var flera av dessa studier ej applicerbara inom stålindustrin. Litteraturstudien visade även att tidigare forskning med identiska mål existerar i ett väldigt begränsat antal. De mätmetoder som hade störst relevans var kameramätnings- och vibrationsmätningstekniker. Fältstudiens syfte var att samla information om de nuvarande fysiska förutsättningarna på varje stålverk samt att förstå de praktiska problemen som uppkommer om en mätmetod skulle implementeras. Fältstudien omfattade också intervjuer och diskussioner med erfarna operatörer och ingenjörer. Från fältveckan kunde det sedan konstateras att de största svårigheterna för en eventuell implementering av ett mätinstrument är de fysiska förutsättningarna på ett stålverk, så som temperatur, ljudnivå och buller. Det är alltså inte bristen på forskning, modeller eller simuleringar som saknas för en implementering av ett mätinstrument. Resultaten av fältveckan indikerade att samtliga stålverk behöver en individuellt anpassad lösning. Emellertid har de undersökta stålverken i stor utsträckning liknande problem, såsom igentäppning av spolstenen som leder till felaktigheter i den uppmätta mängden argongas som går in i skänken. Det sistnämnda skulle möjligtvis kunna förbättras genom att placera flödesmätaren närmare skänken eller förbättra spolstenen.

(4)

Abstract

The main objectives of this research were to find a measuring technique for the gas stirring intensity and relate this to the argon gas stirring in the ladle furnace. However, the investigation mainly focused on methods of measuring the true amount of argon gas entering the ladle furnace, since this was needed to further determine the gas stirring intensity. The gas stirring intensity is of importance because steel plants want to relate the quality and composition of the steel to the effect of the stirring.

The methods for executing this research have been based on a literature study and a field trip, where five Swedish steel plants have been investigated. The steel companies included SSAB, Uddeholms AB, Sandvik, Ovako and Outokumpu. The literature study focused on previous experiments with similar objectives that have been carried out, and other processing industries where measurements of gas flow are used. Since the physical conditions at steel plants are rather extreme (with respect to temperature and noise) compared to most other industries, those researches were not applicable for this implementation. The literature study also showed that previous studies with the same aim only exist to a very limited extent. The measuring techniques found to have a certain level of relevance were camera and vibrational measuring techniques. The field trip focused on collecting information about the present physical conditions at each steel plant and understanding the practical problems if a measuring method were to be implemented. In addition to this, the field trip investigation also included interviews and discussions with experienced operators and engineers. From the field trip it can be concluded that the most severe limitations for implementing are the physical conditions at the steel plant, such as temperature, loudness and vibrations, and not the research and simulations being available of what they would show. The results of the field trip indicated that all steel plants need an individual solution. However, most investigated companies face problems such as clogging of porous plug and inaccuracy of the measurement of the argon gas entering the ladle furnace. The latter can be partially removed by placing the flowmeter closer to the porous plug.

(5)

Acknowledgements

We would like to give a special thanks to our supervisor Hans Kellner, who supported us with great ideas and initiated valuable discussions with us every week throughout this project.

Additionally, we would like to show our big appreciation to Professor Pär Jönsson who has not only been a motivator and believed in us through the course of this project, but also helped us to initiate the project through his extensive network within the steel making industry.

Furthermore, we would like to thank the following people, whom without the fieldtrip would not have been possible; Jens Sörlin and Sara Åslund from SSAB, Karin Steneholm and Ewa Persson from Uddeholms AB, Olle Sundqvist from Sandvik, Saman Mostafaee and Jörgen Bergroth from Ovako, and lastly Jesper Janis and Fredrik B. Larsson from Outokumpu whom also supported us with his encouragement, time and kindness throughout the project.

(6)

1 INTRODUCTION

1.1 Background

Refining in a ladle furnace is one of the most significant steps in the processing of extracting metals. The purpose is to decrease the amounts of unwanted substances, such as sulphur, oxygen and hydrogen as well as to remove harmful non-metallic inclusions to the level that they increase the properties of the material. The purpose of the ladle furnace is therefore to refine liquid metal to produce high quality steel by which it raises the temperature of the molten steel and adjusts the chemical composition. The ladle furnace, also known as the ladle

refining furnace, is a proven technology for desulphurization, where the concentration of

sulphur is decreased to as low as 0.001 wt% [1]. The process also reduces levels of oxygen (deoxidization), hydrogen (degassing) and other undesirable non-metallic materials (micro-cleanliness). This is made to change the composition and remove inclusions for an improved microstructure to increase mechanical properties such as ductility, toughness and transverse properties [2] [3] [4].

In steelmaking, which is the primary metal production in Sweden, knowledge regarding the dynamic reoxidation between the slag and steel is very important to produce high quality steel. Stirring in the ladle furnace is used to improve the kinetics of refining operations in addition to achieve growth and separation of the non-metallic inclusions from the liquid steel to the slag. Normally, this is done through gas stirring, induction stirring or a combination of the two, to obtain thermal and chemical homogenization. Gas stirring is one of the most effective and important methods were argon gas often is used thorough a few so called porous plugs. This method combined with induction stirring, or electromagnetic stirring (EMS), is commonly used for many steel companies to minimize inclusions in the final steel. The stirring results in a turbulent flow in the ladle due to expansion of the gas bubbles as well as the rising of bubbles to the top of the slag. The rising gas bubbles impinge on the slag intermittently and break the slag layer to create a slag eye [5]. This will however be discussed in greater depth in the literary study. Yet, steel corporations have a lot of difficulties to understand the gas flow in microscopic size because of the heat and the opaque ladle furnace [6] [7] [8]. Hence, it is significant for steel corporations to understand the effect of the gas stirring intensity and how it relates to the steel quality regarding composition and inclusions.

(8)

(9)

1.2 Previous work at Swedish Steel Corporations

This investigation will consist of a literature study and a field trip where an investigation will take place. Before this is carried out, it is important to understand what has been done so far. From discussions with Fredrik B. Larsson at Outokumpu in Avesta, different techniques using a camera have been tried at several Swedish steel companies. Usually, one tries to monitor the deslagging by using a camera displaying different colours for the steel and the slag. Hence, this technique can also be used to watch the open eye to further correlate it to the stirring intensity.

As previously mentioned, one maintains varying slag due to the varying procedure in the converter depending on its state of input from the arc furnace as well as the assortment of steel being made. There will varying amounts of lime, dolomite and flux, the latter being known as the main slag producer, between the charges. In addition to that, the amount of reducing agent; silicon and aluminium, will vary depending on how much slag the batch contains. Furthermore, there is also a certain part of residual slag from the arc furnace, which composition will vary since one cannot make a complete deslagging before charging the converter. The ratio between SiO2/Al2O3 will alter between silicon reduced and aluminium

reduced steels.

After the reduction in the AOD converter, the charge is partly deslagged by carefully tipping the converter and letting the slag flow into a “slag butt”. How much slag that flows out depends on the controller executing this procedure. Thereafter, an artificial slag is added, consisting of lime (CaO) and flux (CaF2) to lower the sulphur levels before the stage at the

converter is finished. This is known as the two-slag practice and is used when low sulphur levels are prioritized. Moreover, the controller can remove further slag depending on the amount of slag the controller believes to have left, so that everything will fit in the casting ladle. All of these stages contribute to the reason for a varying slag. However, this may depend from company to company; the explanation above refers to the procedures at Outokumpu.

The composition of the slag can have a significant impact on the viscosity. This combined with the amount of slag that comes to the ladle furnace makes it difficult for companies such as Outokumpu to draw any conclusions about how the size of the open eye is related to the flow of gas. Despite these difficulties, the controllers still monitor the flow from the porous plug based on the size of the open eye, since a more vigorous flow is demanded in the beginning of the process to let it decrease towards the end. Additional problems faced in the steel industry is clogging of the porous plugs, making gas escape before entering the ladle or taking new paths without contributing to the stirring intensity.

(10)

might therefore have a less varying slag which may contribute to more accurate conditions of relating the size of the open eye to the gas flow.

1.3 Aims and Objectives

(11)

2 LITERATURE STUDY

2.1 Porous plugs

The primary use of the porous plug is to maximize the effectiveness of the stirring as well as to create the highest possible surface area of the bubbles entering the steel bath for increased yield. These are the main features of the plug when taking fluid mechanics into account, which is the most common calculation method used for steel simulations when estimating the rate of gas flow from the porous plugs.

Figure 1. This figure shows a) hybrid and b) slot purging plug designs [5]

(12)

Figure 2. This figure shows the slot purging plug a) cross section and b) the gas channel

configuration [7]

The hybrid purging plug is the most recently produced of the two and differentiates itself from the slot purging plug since it is made by two separate pieces, mounted together at a later state of production, see figure 3. One of the pieces is a protection for infiltration, which is placed at the top of the plug in the shape of the bottom of a pyramid, also made out of the grainy ceramic material; aluminium oxide. Between the ceramic part and the surrounding ceramic outer layer, which also acts as the component in which the aluminium oxide is installed, the gas is free to flow. The gas can either flow this way, or through the porous aluminium oxide in the centre of the plug, hence the name hybrid plug. According to RHI, the latter model has less risk of blockage of the plug, due to its configuration since infiltration of molten steel is decreased.

Figure 3. This figure shows a hybrid plug a) cross section and b) the gas channel configuration [7]

(13)

mounted together by separately produced pieces. The segment plug consists mainly of three parts; the protection for infiltration made out of porous aluminium oxide, ceramic segments and finally a ceramic outer layer. The porous aluminium oxide is designed so that the molten steel will not permeate the plug. However, permeation of the molten steel does occur, hence providing opportunities for improvements of this design. The ceramic segments form channels allowing the flow of gas. In this plug there are six paths for the gas flow. The ceramic outer layer is acts as the component in which the two prior parts are installed, similarly to the hybrid plug.

Figure 4. This figure shows the segment plug a) cross section and b) the gas channel configuration [7] 2.2 Stirring in Ladle Furnace

The main point in the gas stirring operation is to identify procedures and equipment needed for achieving a minimum mixing time and a maximum yield of alloys at an optimal gas flow rate. In order to understand these phenomena, mathematical models have been invented to obtain detailed information for the rate of gas flow in the ladle furnace. These mathematical models are also important to understand the gas-plume behaviour and the interface between the slag-metal and melt mixing.

(14)

when using water models are to find corresponding materials to the steel and slag, with similar properties such as viscosity, interference and surface tension between the slag and the melt. Several tests have been made with different types of gas flow rates and the results have shown that the mixing time decreases with an increasing gas flow and numerous observations showing that the measured mixing time also depends on the point of tracer injection and the number of plugs. It has been reported that the mixing time decreases if the location of the plug became more off-centred [10]. On the other hand, alternative observations show that a shorter mixing was achieved with gas injection in the centre of the vessel [6]. The turbulent flow in the ladle therefore depends on where the plume is located in the bottom of the ladle.

Generally one off-centred plug is used to inject the argon gas into the steel bath, but dual plugged configurations have also been investigated by numerical simulations where the slag layer and gas flow is present [6]. The slag layer deformation given from the rising gas bubbles and slag formation of the open eye at different gas flow rates can be calculated from the mathematical models; Lagrangian discrete phase model and the Eulerian volume of fluid model. These models describe the bubble plume and the tracking the free surface of the melt.

Although a high gas flow rate made from the two-plug arrangement may improve the slag emulsification and increase the stirring intensity it also increases the interfacial velocity at steel-slag interface which leads to a thinner slag. Simultaneously, this high gas flow rate limits the slag layer to have enough time to absorb the inclusions. Therefore, one should be careful to use a gas flow rate which is too high for the most effective removal of inclusions in the molten steel [10].

(15)

Figure 5. This figure illustrates the ladle furnace while injection argon gas [11]

(16)

2.3 Vibrational Measurements

When argon gas is injected to the molten steel into the container, there is a certain extent to which the gas flow causes the container to vibrate. These vibrations can be measured through analogue signals that correspond to the rate of gas entering the molten steel. In 1998 during the Steelmaking Conference Proceedings held in Toronto, a paper by R.L. Minion was published, discussing that the flow rate of gas entering the steel containment vessel is not according to the flow rate of gas being delivered to the steel containment vessel. Minion et. al. came up with a method of stirring detection involving ladle vibrations to measure the energy transfer to the ladle [13]. In this proposed process, an accelerometer was attached to the ladle through a magnet, allowing vertical vibrations to be detected. The detected signals were filtered so that the frequency only measured the vibrations from the argon stirring. However, the results did not show a direct relation to the amount of gas delivered to the steel containment vessel. Due to Minion et. al. the incongruous results may have been caused by “bubble formation frequency at the porous plug”, yet there does not seem to be any empirical studies on this fact [13].

Similarly, as to an accelerometer, an Acoustic Doppler Velocimeter (ADV) can be used to measure sound waves according to the Doppler Effect. Sound waves, or vibrations, are created by the movement of the bubbles from the gas stirring. The Doppler Effect is formed by a frequency shift of the sound waves according to equation 1.

𝐹𝑑𝑜𝑝𝑝𝑙𝑒𝑟 = −𝐹𝑠𝑜𝑢𝑟𝑐𝑒

𝑉 𝐶

Equation 1. Describing the Doppler Effect.

Where 𝐹_{𝑑𝑜𝑝𝑝𝑙𝑒𝑟} is the change in frequency received, also known as the Doppler shift, where −𝐹𝑠𝑜𝑢𝑟𝑐𝑒 is the frequency of transmitted sound, V is the velocity of the source in relation to

(17)

Figure 6. Showing an Acoustic Doppler Velocimeter [14]

A promising company focusing on the design of innovative technology to solve problems unique to the metallurgical industry is Nupro Corporation seated in USA. They specify, supply and implement the latest technologies for companies in the steel industry. One of their applications, for measuring the gas flow rate in the ladle furnace is the Argon gas stirring &

Arc monitor. It can be applied for any multi-plug, two-plug or single plug lance and

electromagnetic system. Nupro Corporation have developed TrueStirTM to facilitate the argon gas stir rate, a vibration based argon stirring control system, see figure 7 [15].

Figure 7. This figure shows the application TruStirTM [15]

An accelerometer will sense the stirring intensity from vibrations when the argon gas causes the ladle to vibrate. The accelerometer in turn converts vibrational signals into digital signals and the stirring intensity can therefore be measured and compared to the amount of argon gas in the ladle shown by the flowmeter in the control room.

(18)

porous plugs. This technique results in immediate benefits due to the fully automatic system which senses the change in flow and compensates by adjusting the argon gas controller. The ladle vibrations will directly be linked to the stirring energy. At the website, Nupro Corporation briefly explains about the application including a list of immediate benefits that

TruStirTM contributes with at the steel plant, such as: lower argon consumption, decreased clogging of porous plug, cleaner steel due to improved desulphurisation and temperature homogenization etc. These benefits, of course, will be resulted with the obtained knowledge about the relationship between the gas amounts injected to the stirring intensity in the ladle.

At the same page on the website the company also explains the need of TruStirTM and explains the difficulties of measuring the stirring rate with respect of the argon gas injected to the ladle. These problems, as mentioned earlier in this work are:

 Clogging of porous plugs, resulting in lower than expected stir rate.  Leaks in the supply system.

 Variable back pressure due to clogging of porous plugs.

 Operator error in judging the stir rate due to variable slag thickness and consistency.  Lack of real time record of stir history on each ladle.

(19)

2.4 Camera Measurements

When stirring with argon gas, the bubbles created rise to the top of the ladle, dragging the melted metal with them usually resulting in a gas-liquid plume. If the velocity of the bubbles is big enough, the plume becomes large enough to create a bare metal surface. This is based on the assumption that the thickness of slag layer in relation to the gas flow does not exceed a certain level, which varies due to the viscosity of the slag. The region where the bare metal appears is called an open eye [16].

In secondary steel making, an open eye can be seen in the slag in the ladle furnace. Today, an instrument cannot specifically measure the open eye; it is simply judged by the controllers viewing the open eye, hence making their own judgement. This can then be seen as a human fault and irregularity, since different controllers judge the size of the open eye in slightly different ways, making this process different each time, depending on who is viewing the ladle furnace. Through the use of a camera placed above the ladle furnace, one could possibly measure the size, as in width or breadth, of the open eye and therefore establish standardized dimensions.

Previously, studies have been made on measuring the bubble flow in chemical reactions, among which one study was made by Buwa and Ranade where the effect of gas velocity and sparger design was studied through a rectangular bubble column. In addition, the pressure fluctuations on the wall were measured to identify the low frequency oscillations caused by the bubbles acting on their surroundings. A high speed camera can be used to measure bubble size and distribution [17].

(20)

Figure 8. This figure shows the set-up for the experiment by Buwa and Ranade [17]

(21)

Figure 9. This picture shows the set-up for the pair of optical scanners [18]

Another way of using a camera for measurements is by the usage of an infrared (IR) video camera. The German steelmaking company, Saarstahl AG [16], installed an IR camera by their ladle furnace to monitor the formed open eye. The images from the IR camera were analysed through a software package to accurately measure the size of the open eye. This could then be related to the flow of argon gas through a function. To determine the size of the open eye, the flow of argon gas needed to have reached a steady state at a predetermined rate of gas flow. This would usually take 2-3 minutes. In addition, the height of the top slag needed to be constant; this was measured to be in the range of 4-6 cm in a ladle of 170t with an inner diameter of 3 m and a liquid height of 3 m.

A limitation of the IR-camera is that it is not resistant to excess amounts of heat. Therefore, one needs to bear in mind where the camera will be placed for optimum measurements [1]. The camera could be placed in the lid of the ladle furnace, although it then needs to be resistant to the temperatures reached there. Another alternative is to place the camera on a mobile stand, which can be moved back and forth from the ladle furnace.

(22)

(23)

2.5 Semi-empirical model by Wu, Valentin and Sichen

Wu et. al. have developed models to predict the size of the open eye [16]. The empirical studies are based on water models and cold metal modelling, comparing the two to each other and applying them to real industrial practices to confirm their applicability. Through their experiments they tried to come up with a reliable method to estimate the true gas flow into the ladle furnace by measuring the size of the open eye. A mathematical model based on the results is shown below. This model is based on the velocity distribution of a plume, known as the Gaussian distribution, explaining the shape of a bell curve, followed by deriving this formula to come up with the model for the metal bath experiments (Ga-In-Sn) shown in equation 2. This equation shows the model for the Ga-In-Sn experiment to estimate the size of the open eye. This is believed be a good model to estimate the real gas flow rate, although there are uncertainties from blockage of the porous plug as well as leakage from the gas line. Wu et. al. thought of this model as a more accurate estimation for the true gas flow rather than the intended gas flow [16].

𝐴𝜖 = 2.082 (𝑈𝑝 𝑚𝑎𝑥− 0.5√ 2∆𝜌𝑔ℎ 𝜌_𝑙 ) 1.690 𝐻1.220 ℎ0.015

Equation 2. Semi-empirical model by Wu, Valentin and Sichen.

For equation 2; 𝐴_𝜖 is area of the open eye, 𝑈_{𝑝 𝑚𝑎𝑥} is the maximum velocity in the plume, ∆𝜌 is the difference in density between the lower and upper liquids (the supposed difference between steel melt and slag), 𝜌_𝑙 is the density of the liquid bath, H is the height of the liquid bath and h is the height of the modelled slag. If the gas flow is below a certain rate, the plume will not have the nature of a gas jet. By increasing the gas flow, a larger surface area of the diameter of the plume is formed. A higher rate of the gas flow also leads to increased velocity of the plume. Since the density difference between the upper liquid (the slag) and the lower liquid (molten steel) is not the same for the water model and the cold metal model, two equations were made to illustrate the experiments. A common model for the water model and the cold metal model appear not to be compatible due to the difference in physical properties. The model was relatively satisfactory with the results of the experiments used for the cold metal model, showing there may be a relationship between the size of the open eye and the rate of the gas flow.

(24)

(25)

3 METHODS

To find a measuring method for optimising gas stirring and quantifying the argon gas entering the ladle furnace a thorough investigation at a variety of steel plants in Sweden has been made through a field trip. The one week investigation consisted of a detailed mapping of the ladle furnace and the conditions concerning the argon gas injection at each investigated steel company. In addition to this, an extensive sampling of information from each of the steel plants has been carried out, followed by discussions regarding the methods and a possible solution with operators and engineers who are experts within their field. The main task was to determine and quantify the rate of argon gas which would improve the possibilities of identifying measuring method for the stirring intensity due to the gas flow rate.

The investigation included the five steel making companies with facilities located in Sweden:

 SSAB - Oxelösund

 Uddeholms AB - Hagfors

 Sandvik - Sandviken

 Ovako - Hofors

 Outokumpu - Avesta

In the results, a description of each investigated steel plant is given. It includes a description of the ladle furnace and its primary function together with settings and the physical conditions for each company if a future implementation is possible. To gather similar and comparable information from each of the steel plants, six major questions were asked for each of the companies, as seen below:

 What is the main purpose of the argon gas stirring?

 Which kind of flowmeter does the steel plant use today and where is this located in relation to the ladle furnace? As in, how many meters of pipeline is it from the flowmeter to the ladle furnace?

 Is it possible to relocate the flowmeter closer to the ladle and the porous plug?

 How do the operators determine the gas flow and/or stirring intensity today?

 What previous work has been done regarding measuring methods of the gas flow and/or the stirring intensity and how is it analysed today?

(26)

4 RESULTS AND DISCUSSION

4.1 Results

The investigation at each steel plant greatly enhanced the knowledge and understanding regarding the steel plant itself with its advanced technology and complex logistics. The field trip was of great importance for the conformity of the results from the literature study to further be able to comprehend the similarities and differences of theory and practice, and therefore the possibilities and challenges of implementing a solution in practice.

As a starting point based on the literature study the camera and vibrational measuring methods showed to be possible solutions for implementations of measuring the stirring intensity in the ladle furnace. After the investigation a comparison from the steel plants through collected information was made and a greater knowledge about the advanced technology and physical conditions was obtained. At each steel plant, the main overall use of argon injection and its applications differentiates from company to company, which is an important aspect to take into consideration during the discussion. The results have been achieved based on interviews and discussions with the operators, process developers and engineers. Even though the results of the field trip showed to bring a number of limitations to reach the complete aim of this study, there is still a great interest among the operators and engineers to find a method of measuring the gas stirring intensity and hence measuring the amount of gas entering the ladle furnace.

(27)

4.2 Companies

4.2.1 SSAB – Oxelösund

The steel plant in Oxelösund was founded in 1913 and has since 1961 been Sweden´s only fully integrated steel industry. SSAB is a leading producer for high-strength steels and high performance steel. The company is based in Oxelösund, Sweden, with production facilities in Luleå, Borlänge and Oxelsöund where they develop, manufacture and market cold steel, heavy plate and quenched steels. The products are used for manufacturing and maintaining construction machinery, cisterns, bridges, ships, and offshore equipment.

The main purpose for the argon blowing in the ladle furnace is to refine the melted steel from inclusions, homogenize the composition and temperature of the melt. The argon gas flow injection is at the bottom of the ladle furnace to achieve the most effective stirring, according to the operators. The injection of gas is injected from two of centred purge plugs in the bottom of the furnace which in turn results in two open eyes in the slag. The porous plugs have a tendency to clog after a while, but since SSAB has changed the supplier of the porous plugs to RHI, the clogging has decreased.

SSAB measures the gas flow through the use of a flowmeter from Brooks Instrument placed fifty meters from the ladle furnace. To explain; this is 50 meters of pipeline which is the distance the gas has to travel since the most recent measurement by the flowmeter. From the control room, the operators control the gas flow based on the figures from the flowmeter and are then able to visually analyse the stirring intensity through the slag and the open eye. The open eye gives an approximation of how intense the gas stirring is. SSAB also uses EMS stirring. If the stirring is too turbulent and the melted steel splashes at the electrodes, signals are sent to the control room where an automatic system lifts the electrodes further from the surface. By this, operators know when the stirring intensity is too high.

(28)

4.2.2 Uddeholms AB – Hagfors

The design of the processing at Uddeholms AB is focused on production of high quality tool steels. Uddeholms AB uses recycled steel as raw material for its production. For the basic melting and metal refining an electric arc and ladle furnace are used. The vacuum degassing station where argon gas stirring is used is not a bottleneck in the production and is therefore in no need of being further optimized in a time aspect. Despite this, it is of great interest to determine and correlate the gas flow rate to the quality of the steel. It is worth mentioning that there was a minor breakdown in the steel melting shop when this investigation took place. Similarly to the manual deslagging process at SSAB, Uddeholms AB’s process of deslagging is vital for their steel quality. The quality of deslagging leads to a variation of the composition of the new synthetic slag added in the ladle station. For refinery, homogenisation of composition and temperature EMS stirring is used throughout the process (both EAF, LF and VD). During the heating process in the ladle furnace, the molten steel is homogenized using EMS. After the first treatment in the ladle furnace is done the molten steel is moved to the degassing station where argon gas is injected by two porous plugs at the bottom of the ladle. A vacuum lid is then placed on the top of the ladle. To clarify – they do not have a separate vacuum tank like other steel plants may have. In order to improve the fluid flow the EMS stirring is also used which simultaneously improves the removal of inclusions. The usage of argon gas stirring is used to a limited extent at Uddeholms AB, it is only used in the vacuum degassing station. The main purposes of argon gas stirring are for the removal of sulphur, nitrogen, hydrogen and inclusions. Sulphur is removed when the turbulent flow makes the slag mix with the steel.

(29)

4.2.3 Sandvik – Sandviken

Sandvik Materials Technology AB, is a world leading company with operations based on unique expertise in materials engineering and technology. A global market-leading manufacturer of tools and tooling system together with products with advanced stainless steels and special alloys for the most demanding industries.

It is worth mentioning that during the field trip to Sandvik only the visitors’ pathway was available. This meant that there was no opportunity to neither speak to operators nor get an in depth investigation of how the gas tubes were placed near the furnace.

At Sandvik the main purpose for the gas stirring is to remove sulphur from the melt. They use EMS for stirring at other stations which, according to Sandvik, results in a less vigorous stirring than the contribution of the argon gas stirring. Previously, Sandvik had a top-blown converter but changed this in 2008 to a bottom-blown set up. After having done this, they notified a significant change in the EMS stirring. Sandvik mainly uses the ladle to remove inclusions and for temperature control. Similarly to other steel plants, Sandvik has a varying slag. It is therefore difficult to determine the gas flow rate by analysing the slag eye from an IR-camera.

The flowmeter used at Sandvik is of brand TBR Engineering and is located five to ten meters from the ladle.

(30)

4.2.4 Ovako - Hofors

Ovako is a leading European company of engineering steels with a focus on bar, tube, ring and pre-components in low-alloy steels. The steel production is based on scrap. Ovako uses a vacuum tank while injecting argon gas into the ladle. The vacuum process streamlines the process and contributes to a better control of the molten steel. The main purpose of gas stirring is for homogenization of temperature and composition during the ladle treatment.

The flowmeter, placed fifteen meters from the argon injection ingot, displays the measured flow in the control room. Together with a pressure gauge, the flow rate of argon gas and visual observations of the open eye the operators can determine whether there is a leakage before the porous plug or not. This is determined mostly through the pressure gauge; if the pressure is too low, there is no reaction force of the gas flow, meaning that there is a leakage in the pipes. However, small leakages cannot be discovered through this method, only major leakages can be found this way. Also, an exact amount of leakage of the gas cannot yet be measured this way. To conclude how the operators determine the gas flow and stirring intensity, this is done through visual observations combined with the indications from the flowmeter. Like other steel plants, Ovako determines the stirring from the control room by observing the open eye from a camera placed right under the lid to the ladle furnace. This is also where one observes how intense the stirring is. To draw conclusions from only visually observing the stirring, the operators need to have experience regarding the size of the open eye; big contra small open eye.

When connecting the pipeline to the ladle furnace, the operators work close to the ladle and connect it manually. Previous automatic trials have been made, but caused increased gas leakage since the matching of the pipeline and the nozzle was inaccurate.

(31)

4.2.5 Outokumpu - Avesta

Outokumpu is a company focusing of stainless steel production and the head office is seated in Espoo, Finland. Outokumpu’s stainless steel plant in Avesta is a scrap based steel plant, where steel is produced from raw materials. The scrap is melted in the electric arc furnace by three graphite electrodes before the melt moves to the AOD-converter where the main purpose is to lower the carbon content in the stainless steel and to remove sulphur. These can be measured by theoretical models or through dynamic models based on analyses.

(32)

4.2.6 Summary of companies

Table 1. The table above compiles the major differences of the five companies investigated during the

field trip.

Table 1 gives an insight of the similarities and differences between the investigated steel plants and includes which kinds of previous trials have been made regarding camera and vibrational measurements as well as the notable differences between the findings and ideas that the field trip resulted in. The first finding, important to take into consideration, is the purpose of gas stirring for the steel plants.

Before any further work is done, it is important to investigate the purpose of the processes mentioned in the table above in combination with the importance of support from each steel plant. The table gives a distinct comparison of the similarities and differences and above all that each steel plant needs a tailor-made implementation if a measuring technique was to be implemented.

(33)

4.3 Results related to the literature study 4.3.1 The porous plugs

The porous plugs play an important role for achieving the most effective stirring in the ladle. The disadvantage of the plugs, common for every steel plant, is that they have the tendency to clog after some time. All steel plants use a hybrid plug model, which contributes to the most effective and even flowrate as according to the literature study [7]. The challenge of the clogging of the porous plugs may simply have to be a fact to accept. If one is not able to prevent the clogging, the major issue would then be how it will be possible to measure the true argon gas amount in the ladle even though the porous plugs might clog after a while.

Through discussions with operators and experienced engineers at SSAB, it was concluded that the stirring intensity is determined through visual observations combined with knowledge and experience about the back pressure from in the pipelines, which is an indicator of the clogging of the porous plugs. By visually determining the open eye, the numbers displayed from the flowmeter and the warning of vigorous flow by the electrodes, the operators know if clogging of the porous plugs occurs. At SSAB, experts have looked at the inside of porous plugs to determine what the clogging is a result of; which usually is found to be a result of minor cracks from the centre of the porous plug. These become pressurised by the argon gas, hence making the cracks grow. The optimal condition would be to have a porous plug that does not crack and has no possible leakage of gas. If this optimum condition occurred, engineers would be able to trust that all the gas entering the opening of the porous plug also enters the steel melt. However, the discussion also focused on whether the quantification of argon gas is of use for the steel quality. Today, the composition between different steel batches are similar enough, but not identical to each other. This is because the process differentiates slightly due to the operators’ experiences. Therefore, with a quantified number of the amount of gas entering the ladle furnace the operators would know how much gas to use for stirring - which would lead to a more consistent quality of the steel.

(34)

Ovako has recently changed the supplier of porous plugs to RHI from which there is no experience of clogging of the porous plugs. From discussions with an operator with over twenty years of experience, it can be clarified that the porous plugs are changed frequently enough to avoid clogging. As explained in the methods, the operators notice if there is a gas leakage due to the back pressure in the pipeline. However, it would be of interest to standardise the process so that the same time and intensity of stirring is used, which in turn would be simplified by identifying the exact amount of gas entering the ladle.

(35)

4.3.3 Vibrations

The vibration technique shows a great capability to be implemented when referring to the accelerometer TruStirTM, the product of Nupro Corporation. The question which then follows is; how effective is this instrument in reality and is it possible to use this method in the loud environment at a steel plant? This is a question that remains to be answered through research and decisions by the steel plants. The high technology instrument from Nupro Corporation will have the main purpose of measuring the stirring intensity of which all steel plants do not find equally important to invest in. However, this solution might not only solve the problem of identifying the gas stirring intensity but also it allows control of the entire manufacturing process. The beneficial opportunities of improved manufacturing process are therefore likely to be of interest for all companies and hence an investment in an instrument likes TruStirTM.

According to the investigated steel plants there are some possibilities to determine and get information of the stirring intensity in the vessel by using vibrational measurements. The application was presented for each steel plant followed by a possible solution. For all steel plants there is a common use of argon gas stirring in the ladle, although all corporations have different intensions of usage. Therefore, vibrational measurements may not suit every steel plant even though it seems to be the most effective way of quantifying the gas flow in the ladle furnace.

At Uddeholms AB this was tested in 2005, when Jeremy Jones implemented the test on a ladle furnace to measure the stirring intensity in the ladle but unfortunately without any success. This was because surrounding noise from nearby applications was detected by the accelerometer. However, Uddeholms AB should be one of the companies with more suitable conditions for an implementation similar to TruStirTM, due to the reason that the argon gas injection is placed relatively isolated compared to other steel plants, unlike Sandvik and SSAB. At SSAB vibrational measuring techniques were also brought up for discussion. Although vibrational measuring techniques in theory appear to be a decent solution for identifying the gas stirring intensity, SSAB does not believe it will work in practice. The reason for this is equivalent of the one of Uddeholms AB; disturbing noise from surrounding applications will be recorded on the accelerometer and interfere with the wanted frequencies.

(36)

4.3.4 Camera

The use of a camera is important at each steel plant while the effect of using it may vary. The main purpose of using a camera for all of the steel plants is to observe the open eye and relate the size of the open eye to the stirring intensity in the ladle. If the open eye is too big the injection of argon gas is too big and if the open eye is too small the injection of argon gas is not enough, or clogging of the porous plug has occurred. However, not all steel plants have the type of slag that allows an open eye; sometimes the slag is too thick or not viscous enough to allow an open eye to show. Nevertheless, most operators at the investigated steel plants judge the flow of argon gas by visually determining the size of the open eye.

At the steel plant in Oxelösund, SSAB has made several trials with regular cameras as well as IR-cameras. The company faces two major problems with cameras; for the regular camera, the lens collects dust and other particles from the steel making process which harms the picture. Furthermore, the lens is generally placed at non-accessible locations; either very close to the ladle where temperatures are very high, or at a high altitude, therefore being hard for the operators to reach and clean. The problem encountered with the IR-camera is that the identified heat is too extensive. The displayed heat does not only consist of the wanted area (e.g. the open eye) but also consist of surrounding heat from smoke and rising gases, suddenly making the use of this type of IR-camera irrelevant for the operators.

Previous trials at Uddeholms AB with IR-cameras from Metsol showed that the camera was placed too close to the melt, therefore not allowing the camera to display the full picture of the cross section of the surface area. This could not be changed or improved since the camera needed to be placed inside the lid of the tank, which only allowed a specific maximum distance. Moreover, the picture was not clear enough due to the disturbance of smoke, which also was the major identified problem for Ovako.

(37)

4.3.6 Additional ideas

An idea that developed throughout the field trip and was brought up and discussed with all operators was the relocation of the flowmeter. In most cases, the flowmeter is located between fifteen to thirty metres of pipeline from the nozzle to the ladle. If the flowmeter however was to be placed as close to the porous plug as possible, preferably after the last connection to the porous plug, hence positioned inside the porous plug, this would increase the accuracy of the measured gas flow to a significant amount. Nevertheless, one opinion at SSAB stated that the relocation of the flowmeter would not solve the initial problem, which is the clogging and cracking of the porous plugs.

An interesting aspect discussed at Sandvik, was the use of a vacuum tank to measure a specific gas flow out of the tank. Other discussions at SSAB brought up the aspect of relating the pressure in the vacuum tank to the argon pressure. Ideas about relating the exiting argon gas pressure was also brought up as an interesting area for discussion. However, the problem still remains of how one is to collect the exiting argon gas, since leakage within the brick walls of the ladle still occurs.

At Ovako, the hydrogen content’s connection to gas flow were discussed. This would however not quantify the rate of argon gas which the operators would find interesting to know. Yet, one could ask whether the time, cost and effort to invent a measuring technique would increase the steel quality to a significant extent. Having asked this throughout the week to a varying group of people from operators with twenty years of experience to engineers who have a done a lot of research, there is an interest of relating the gas flow to the stirring intensity. Due to this, it would be of great interest to quantify the gas amount contributing to the stirring intensity.

(38)

5 CONCLUSION

A major deduction that can be made from the results is that every steel plant needs an individual solution if a measuring method is found to be suitable. This is because of the different conditions at each plant in combination with its differing need of a measuring application. Another finding was that the purpose of gas stirring is not identical for all companies since their manufacturing processes differentiate from each other as they produce dissimilar steel products.

Furthermore, it is important to emphasise that the quantification of argon gas does not only lead to opportunities of identifying stirring intensity but also opens doors for further research which is dependent on the gas stirring intensity. Throughout the thesis this was an area of confusion since some operators argue that a need for quantifying the gas amount exists, while others argue the opposite. However, the conclusion made is that to reach the aim of this project; to identify methods for measuring the gas stirring intensity, one first needs to identify the amount of gas affecting the stirring. By this, the need of quantifying the gas arose, hence the shift of focus from a stirring intensity focused research to a gas quantification focused field trip. Moreover, the major shift in focus occurred during the field trip since one does not fully understand the physical conditions and its implications before experiencing it.

The problem identified at all investigated steel plants was the clogging of the porous plug. If an optimal porous plug was to be developed, it would lead to a chain of overcoming problems for steel companies. By optimal means the prevention of cracking in the material of the porous plug since it leads to prevention of gas escapes and therefore clogging. Companies like RHI may find that an improvement of the materials and functioning of the porous plugs may be of interest.

Another identified solution for improvement of accuracy of the gas flow is to move the location of the flow meter closer to the porous plug since leakages would not be detected by the flowmeter; hence, the flowmeter would show a more accurate figure. This may be an interesting idea that could be further developed by the suppliers of flowmeters together with the suppliers of porous plugs.

(39)

(40)

6 FURTHER RESEARCH

Although a measuring technique for the gas stirring intensity was not implemented during this project, there is a great demand of further research towards a final solution. After this project, further areas of research are recommended on the:

 The design, cracking and clogging of porous plugs; through e.g. RHI

 Implementation of vibrational measurements; through e.g. Nupro Corporation

 Implementation of IR-camera measurements; through e.g. Agellis

 Further trials on empirical methods; such as the model by Wu et al.

 Improvement of gas leakage to the ladle furnace

In addition to this it is believed that for a technique to be implemented at any of the investigated companies, the steel plants need to have individual projects to reach a possible solution.

(41)

7 WORKS CITED

[1] _{H. Kellner, Interviewee, Sandvik. [Interview]. 22 February 2016.}

[2] _{I. Corp., “Inductotherm Corp.,” Inductotherm Corp., [Online]. Available:}

http://inductotherm.com/blog/products/ladle-refining-furnaces-lrf/. [Accessed 19 February 2016].

[3] _{M. E. Ltd., “Megatherm - Ladle Refining Furnace,” Megatherm Electronics(P) Ltd., [Online]. Available:}

http://www.megatherm.com/index.php?option=com_content&view=article&id=6&Itemid=476&lang=en. [Accessed 20 February 2016].

[4] _{A. I. S. GmbH, ABP Induction Systems GmbH, [Online]. Available:}

http://www.abpinduction.com/en/steel-plants/steel-refining/ladle-furnace/. [Accessed 19 February 2016].

[5] _{W. F. A. V. L. K. a. G. H. Bernd Trummer, “Purgning Plugs for Soft Gas bubbling: A Water Modelling}

Comparison of Hybrid and Slot Designs,” RHI Bulletin, pp. 29-33, 2014.

[6] _{R. I. L. G. S. Joo, Modeling flows and mixing in steelmaking ladles designed for single- and dual-plug}

bubbling operations, Issue 6 ed., Montreal: Springer-Verlag, 1992.

[7] _{H. Bäckström, “Jämförande Studie av Spolstenar,” Luleå Tekniska Universitet, Instutitionen för}

tillämpad fysik, maskin- och materialteknik, Avdelningen för Strömningslära, Luleå, 2008.

[8] _{P. J. a. L. J. Jonas ALEXIS, “Heating and Electromagnetic Stirring in a Ladle Furnace–A Simulation}

Model,” ISIJ International, 2000.

[9] _{P. Jönsson, Interviewee, Professor. [Interview]. 1 February 2016.}

[10] _{Z. Q. a. M. X. Heping Liu, “Numerical Simulation of Fluid Flow and Interfacial Behavior in}

Three-phaseArgon-Stirred Ladles with One Plug and Dual Plugs,” Steel Research International, Weinheim, 2011.

[11] _{J. W. E. Dipak Mazumdar, Modeling of Steelmaking Processes, New York: Taylor & Francis Group,}

2010.

[12] _{S. E. a. G. G. Dirk Räbiger, “Application of the Ultrasound Doppler method for velocity measurements}

in an electromagnetically-stirred liquid metal,” ResearchGate, pp. 143-146, 2016.

[13] _{D. I. W. J. A. T. J. Francis L. Kemeny, “Process for controlling the stirring energy delivered by a gas}

flowing through a liquid,” 2001.

[14] _{Palmer, “Acoustic Doppler Velocimetry,” 2002.}

[15] _{nuprocorp, “http://www.nuprocorp.com/,” Nupro Corporation, [Online]. Available:}

http://www.nuprocorp.com/pdf_file/TruStir_Brochure_0505.pdf. [Accessed 24 February 2016].

[16] _{P. V. a. D. S. L. Wu, “Study of Open Eye Formation in an Argon Stirred Ladle,” steel research} international, vol. 81, no. 7, pp. 508-515, 2010.

[17] _{V. V. R. Vivek V. Buwa, “Dynamics of gas–liquid flow in a rectangular bubble column: experiments and}

single/multi-group CFD simulations,” Chemical Engineering Science, vol. 57, no. 22-23, p. 4715–4736, 2002.

[18] _{T. H. a. J. Sakakibara, “High-speed scanning stereoscopic PIV for 3D vorticity measurement in liquids,”} Measurement Science and Technology, vol. 15, no. 6, 2004.

[19] _{C. Medioni, “Influence of stirring on the inclusion characteristics during vacuum degassing in a ladle,”}