Study of high-temperature chamber-type electric resistance furnace and attainment of uniform temperature field

 

Borislav Hristov Dimitrov,     Ivan Stoyanov Yachev

Borislav Dimitrov – Technical University of Varna, Varna, Bulgaria bdimitrov@processmodeling.org

Ivan Yachev – Technical University of Sofia, Sofia, Bulgaria yatchev@yahoo.com

 

 

Abstract— This paper presents an analysis of the temperature field distribution in the workspace of a high-temperature chamber-type electric resistance furnace operating on silit heaters. A methodology is suggested of how to implement a system of events ensuring the attainment of even distribution of the temperature field in the chamber. The methodology is based on numerical models and experimental studies to whose results the furnace controls are adjusted. Experimental studies were conducted under specific technological conditions, which confirmed the correctness of the methodology. The aim was to solve a major technological problem in the exploitation of similar facilities: the uneven field in the chamber due to some of its constructional specifics and the silit heaters aging.

 

                                                                                                                                             I.     Introduction

 

High-temperature chamber-type electric resistance furnaces (ERF), furnished with silit heaters, are designed to provide a wide range of heating processes. They have high reliability, long life cycle, and can be used both for special and general purposes. It is characteristic of these facilities that the temperature field in the chambers of many of them is not evenly distributed. This is basically due to two factors: constructional specifics of the furnace, which determine the configuration of the heaters and the load relative to each other as well as the alteration of the resistance of the heaters as a result of their aging. With the design of new furnaces [1,4], the constructional problem is solved by placing the heaters in such a configuration as to ensure a field of desired shape. For the existing equipment, however, this option is not applicable, because it would necessitate a thorough reconstruction. The problem with the increased resistance of the heaters results in uneven distribution of the power, and hence of the temperature, too, since they age in different ways in the process of operation.

This study suggests a way to solve these problems through a methodology, applied in the following order:

 

1.      Make a model of the furnace and conduct a simulation procedure by the End-Elements Method (EEM) [6,7];

2.      Test the model experimentally and correct it if need be;

3.      Attain an even  temperature field  by simulating the tested model, through manipulation of the power of the heaters;

4.      Set the furnace controls to the data obtained from the simulation;

5.      Conduct another experiment to verify the obtained results and ensure proper operation;

 

                                                                                                                                                          II.    Analysis

 

The studies were carried out by using a high-temperature chamber-type ERF operating on 10 silit heaters with a total installed power of 20kW. The technological process carried out by the furnace is firing of protective enamel coatings, which requires temperature within the limits of 800-830оС. The methodology suggested here was followed in the order given above.

 

1. Making a model of the furnace and simulation by EEM.

 

Fig.1 shows a model of the studied furnace, made by EEM and realized in the program Comsol Multiphysics [9], which shows unevenness of the temperature field. The experimental tests were conducted on several different silit ERF (fig.2.А), and the temperature of the heaters was measured by help of thermographic chamber (fig.2.B). The obtained results (unestablished mode) show uneven heating of the heaters, which is due to the factors mentioned above.

The workspace is outlined (fig.1) as a rectangle with sides A1-B1. According to the obtained results from the simulation, the temperature in the front part of the chamber (near the door) is lower than that in the back one (near the bottom). Because of the constructional specifics of the studied furnace, the larger part of space A1-B1 does not ensure even distribution of the desired temperature range. The desired range is obtained only in a significantly smaller area A2-B2, which would ensure quality production meeting the requirements in full. Additionally, the uneven distribution is also shown in fig.3 – the temperature on the surface of the charge, designated as points А-B in fig.1. The utilization only of workspace A2-B2 of the chamber would result in low productivity and inefficient operation of the equipment, therefore, this cannot be accepted for a solution.

The model was made by use of mathematical apparatus, realizing a transitional process of heating. The equation describing the process is [6,7]:

 

 


(1)

 

Fig. 1 Model of a high-temperature chamber-type electric resistance furnace with silit heaters.

1 – heaters; 2 – chamber; 3 – door; 4 – walls of the furnace; 5 – workspace

Fig.2. Experimental tests on a high-temperature ERF with silit heaters. A: photos of the experimented ERF;

B: Measuring the temperature of the heaters by thermographic analysis.

 

Fig. 3. Distribution of the temperature on the surface of the heated charge,

situated between points А-В in fig.1

 

where:

k[W/(m.K)] – coefficient of conductive heat exchange;

r[kg/m3] – density;

Т[K] – temperature;

Cp[J/(kg.k)] – heat capacity at constant pressure;

qS[W/(m3.K)] – delivered or released thermal flow;

Q [W/m3] – installed power.

 

We used several boundary conditions. A condition for thermal impenetrability is assigned to the end boundary zone, including a sufficiently large air medium:

 


(2)

 

The equation of heat transfer through convective and radiant heat exchange from the furnace body toward the ambient medium:

 

 


(3)

 

and by analogy, with radiant heat exchange in the chamber of the furnace:

 

 


(4)

 

 

where:

q0[W/m2] – thermal flow;

h [W/(m2.K)] – coefficient of heat transfer;

Тinf[K] – outside temperature;

e - coefficient of blackness;

Тamb[K] – ambient temperature;

G [W/m2] – irradiation – total arriving radiative flux;

J0 [W/m2] – radiosity – outgoing radiative flux.

 

2. Test of the model through experiment.

 

The make of a proper model corresponding as much as possible to the processes running in the equipment is a basic requirement for further work. Therefore, its adjustment to the experimental data is of utmost importance. It should be noted that the simulation in fig.1 is based on an idealized model and does not present all geometric specifics of the furnace. This is done to economize computer resources and is another argument in favor of experimental studies.

The comparative analysis model (graph 1) – experiment (graph 2) is shown in fig.4. The experiment was not conducted under full installed power. The figure shows that at the beginning of the transitional process the error is over 100%, which is due to the following factors:

 

·                The experimental values were obtained from a thermal pair, which was not introduced into the model, whereas the data from the computation procedure were read directly from a point in the furnace space. The thermal pair was fitted in the chamber by means of a ceramic protective pipe and had its own thermal inertness. Its introduction into the model would have complicated the computation procedure to a large extent due to the great difference in the geometric sizes toward the other elements of the construction.

·                The convective heat exchange in the chamber of the furnace is described incompletely in the model. The conducted research [1,4] shows that convection in the closed  space between the heaters and the charge has a great impact at the beginning of the transitional process. The heat exchange after the zone of the mean temperatures is mostly radiant and therefore convection could be ignored.

·                The experimentally heated charge (А11 fig.1) is a metal stand with fixed enamel details to be subjected to thermal processing. The stand is introduced in the model as a monolithic body with averaged thermotechnical characteristics: coefficient of conduction, specific thermal coefficient and density.

·                The model of the furnace is two-dimensional and idealized, therefore some constructional elements, such as: specific geometry of the door, gaskets, defects and so on, were not introduced into it. These are taken into account by manipulating the thermotechnical characteristics of the insulating materials used in equations (1) – (4).

 

Despite the obtained error at the beginning of the transitional process, the model can be considered correct since the error in the desired temperature range is within admissible limits. In this case, the model is not applicable in operation at temperatures below 450оС, but silit furnaces are not commonly used for low-temperature processes.

It should be noted that the shown results are obtained after a multiple adjustment of the model to the experimental data. Although this approach is not strict enough, it is quite acceptable in view of the limited computer resources and in the context of the task being solves it can be considered admissible.

 

3. Attainment of even temperature field in the chamber of the furnace.

 

At this stage of the methodology, a specified model of the studied chamber-type ERF is used, and it is assumed that this model reflects the running thermal processes with sufficient preciseness. The attainment of even temperature field is achieved by varying of the power of the heaters. The result is presented in fig.5. – the temperature field in the workspace is even and within assigned limits. For the construction under consideration, this is achieved by the following distribution of powers: heater 1 – 1650W; 2,3 – 1850W; 4 – 1600W; 5 – 1500W. The powers of the oppositely arranged heaters are the same because the construction is symmetric. With these values is attained a distribution of temperature on the surface of the charge as shown in fig.6.

 

4. Setting the furnace controls to the data obtained from the simulation.

 

The suggested methodology is feasible provided that independent control is used for each silit heater of the furnace. The operation is ensured by replacing the furnace transformers and circuit closers with electronic converters feeding the heaters independently [2,3,5]. Here we suggest a principally new approach for control of silit ERF.

In the case, the apparatus part is provided by a system of programmable logical controllers (PLC) MFD-CP8NT, operating with a specialized entry-exit module for control of thermal processes MFD-TP12-PT-A [8]. The control over the converters is carried out by an additional microprocessor module. We used bridge and semi-bridge circuits as well as circuits with a common point, made with MOS, MOSFET or IGBT transistors [5].

The setting of the controls is carried out by introduction of the powers and temperatures mentioned above into the program software of PLC.

 

5. Conducting of another experiment.

 

The second experiment is conducted to confirm the obtained results, which is of special importance when using the equipment for industrial purposes.

 

Fig. 4. Comparative analysis model (1) experiment (2) in transitional process of heating the studied furnace

 

Fig.5 Even distribution of the temperature field in the chamber of the furnace,

attained through the system of control

 

Fig.6. Distribution of the temperature on the surface of the heated charge,

situated between points А-В in fig.1

 

The assessment of the obtained results is within the competence of experts on thermal processes, technologists and others. The end result of the conducted multiple experiments shows that from a technological point of view it is recommendable that to each of the heaters be introduced an independent thermal pair, situated in the chamber in immediate proximity to the heater [2]. This is of special importance when operating old and amortized ERF.

 

                                                                                                                                             III.   Conclusions

 

The relevance of the suggested methodology is experimentally confirmed by testing several types of ERF with silit heaters (fig.2). The obtained results entail the following conclusions:

 

·         The combines usage of mathematical models, experimental tests and the system of control developed as above [2], enables the attainment of even temperature field in the chamber of a silit furnace. The suggested order of operating solves a significant problem in high-temperature silit ERF by eliminating the impact of the construction and the effect of aging of the heaters.

·         The application of the controls described above which facilitate the independent control of the heaters [2,3,5], considerably broadens the possibilities for temperature control in ERF. It should  be borne in mind that operation of furnace transformers feeding in parallel all silit heaters is a morally outdated and inefficient practice.

·         The possibility to attain the temperature field shown in fig.5 compared to that in fig.1. considerably broadens the operating zone under the specified temperature requirements. This results in increased productivity, and hence, greater energy efficiency of ERF.

·         The experiments carried out repetitively validate the conclusion that the data from the simulation can be used for setting of the furnace controls, whereas the computation procedures facilitate the determination of the power required for each heater. Nonetheless, it is recommendable that thermal pairs be added to the heaters in order to guarantee the attainment of even temperature field.

·         The models of the studied ERF thus developed and specified can be included in the passport of the furnace. This will allow the methodology to be used for a wide range of heating processes, charges with a variety of characteristics and so forth.

 
Acknowledgment

 

This publication was made possible thanks to project МУ-03/163: “Promotion of energy efficiency and optimization of electrоtechnological processes and devices”, financed by the Fund „Scientific Research” within the Ministry of Education, Youth and Science in Bulgaria.

 

References

 

[1]     Dimitrov B. Analysis and modeling of electrоtechnological processes and devices /electric resistance furnaces wth periodic action/. Doctoral dissertation, TU-Varna 2008.

[2]     Dimitrov B., Nikolov G., Andreev P. A system for control of high-temperature chamber-type electric resistance furnaces with silit heaters by programmable logical controllers. The 50th anniversary congress of TU-Varna 2012, under print.

[3]     Yordanova M. V. Dimitrov. Accounting for the impact of indicators of electrical safety during operation of high-temperature chamber-type electric resistance furnaces. The 50th anniversary congress of TU-Varna 2012, under print.

[4]     Tahrilov H, B. Dimitrov. Analysis, modeling and optimization of chamber-type electric resistance furnaces. Second edition TU – Varna 2010.

[5]     Dimitrov B. A. Marinov, V. Valchev Experimentally based selection of a structure of a power electronic converter for control of silicon carbide electric heater. Siela 2012, 82-89 pages.

[6]      

[7]     /

[8]     http://moeller.com/