PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters

 

Borislav Dimitrov            Georgi Nikolov               Pavel Andreev

 

assoc. prof. PhD eng. Borislav Dimitrov lecturer at  TU-Varna, department “Electrical Engineering and Electrîtechnologies”, bdimitrov@processmodeling.org

assist. prof PhD eng. Georgi Nikolov, lecturer at TU-Varna, department “Electronics and microelectronics”, georgi.nikolov@tu-varna.bg

eng. Pavel Andreev, master student in “Electrical Engineering and Electrîtechnologies”,

paliou@abv.bg

 

Abstract. A system for high temperature chamber resistive furnace (CRF) with silicon carbide (SiC) heaters is proposed. It is realized with programmable logic controller (PLC) and additional microcontroller board. The purpose of the system is to allow electrothermal apparatus control, while solving an important technological problem – temperature distribution inside the furnace. An experimental system is built and verified. It allows automatic control of separate SiC heaters, achieving more uniform distribution of the temperature field.

Keywords: Programmable logic controllers, regulation, silicon carbide heaters, microcontroller system.

 

². Introduction

 

Silicon carbide heaters that are used in CRF are with typical power 2kW and nominal supply voltage in the range 60¸100V, varying from manufacturers [8, 9, 10]. During their exploitation one major drawback appears – their resistance increase with aging. This leads to lower temperature and non uniform distribution inside the furnace. The change of the resistance (R%) with aging for two SiC heaters manufactured by Kanthal is shown in fig. 1 [7].

Additional investigations of this process are carried out with thermo imaging camera and are shown in Fig.2. The furnace that is investigated is old heaters, which lead to the non uniform temperature distribution. It should be noted that the two defective heaters cannot be compensated by the system, and they should be replaced.

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image001.jpg

 

Fig.1. Change of the resistance versus working hours for two SiC heaters.

 

Most of the CRF currently in use have step down transformers (Fig.3) and electromechanical contactors for control. Such systems cannot solve the problem with temperature distribution, and for this reason a separated PLC based power control for each heater is proposed.

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image002.jpg

 

Fig.2. Temperature examination with thermo imaging camera. À, B – two walls of the tested furnace with defective heaters and non uniform temperature

 

Such systems are proposed by some manufacturers, but in the information available, only the general requirements for powering the heaters are presented. Usually these systems are based on old technologies – triacs working in ON/OFF mode, powered by low frequency transformer. Besides this no information for temperature distribution inside the chamber is presented.

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image003.jpg

 

Fig.3. Low frequency step down transformer for CRF with SiC heaters.

 

The current paper is based on previous research [1, 2, 3], that allows control of CRF with spiral heaters and PLC systems.

The goal of this paper is to suggest a regulating system that allows independent control of each SiC heater. By this way uniform distribution of the power, respectively temperature inside the furnace should be obtained.

 

²². Analysis

 

The proposed system is realized with PLC, microcontroller and power converter. Block diagram is presented on Fig.4, where:

 

·   PLC – programmable logic controller. The experimental tests are carried out using PLC Easy and MFD, distributed by Moeller (Eaton) [11], but it can be any of the vast majority of PLCs with similar parameters. The last are used as temperature controllers with PID or ON/OFF mode.  The input output functions are as follows: 4 analog inputs, receiving signals from the temperature sensors and 4 digital outputs connected to the microcontroller systems (MS). The output frequency is in the range of 100Hz (minimum temperature) to 3000Hz (maximum temperature).

·   ÒC sensor unit. It consists of  thermocouples hat are located inside the chamber and signal conditioner, which amplifies the thermovoltage to the working range (0-10V) of the analog inputs of the PLC.

·   MS microcontroller system. Realized using PIC16F877 and TTL ICs and optically insolated transistor drivers. This block supplies the control signals for the output transistors. The signals is pulse width modulated square wave with variable duty ratio d=3¸97%. Low duty ratio (3%) corresponds to high temperature (and frequency 3kHz), which requires low power to the output transistors and vice versa. The temperature is converted by an external circuit which is not discussed here. It is also possible to use special Input/Output modules for the PLC working directly with thermo sensors.

·   EP – power converter. Experiments are carried out with full bridge, half bridge and push pull schemes, realized by MOSs, MOSFETs or IGBTs. The lower voltage is provided by high frequency ferrite transformer. Usually high power regulators require three phase power supply. However in this case several single phase converters are used. The balancing of the load is achieved by distribution of the converters among the phases.

·   SiCSiC heater with rated power 2kW.

·   PS – power supply for the PLC and MS.

The main blocks of the software for the PLC are shown in Fig.5.

Block À – Analog Value Comparator (Fig.5.À) is analog comparator that compares the analog input (I1) with voltage proportional to the measured temperature with set constant value (I2) – proportional to the desired value.

The parameters F1 and F2 are gain factors for I1 and I2 [6]:

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image004.gif

 

(1)

 

where: I1 – Comparison value 1; F1 – gain factor for I1, which can be constant or variable; Òtmp – current measured value, in this case temperature.

The parameter HY is switching hysteresis and is used to manipulate only input I2 [6]:

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image005.gif

(2)

(3)

 

 

where: I2HY – Comparison value 2, after calculation of HY; I2tmp – current value of I2; HY – value of the hysteresis, which can be constant or variable.

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image006.jpg

 

Fig.4. Block diagram of the CRF control with SiC heaters

 

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image007.gif

 

Fig.5. Main functional blocks of the PLC

 

Block DC – PID Controller (Fig.5.Â) Is PID controller with the following elements: I1, I2 – inputs; KP - Proportional gain, TN - Reset time, TV - Rate time, TC – scan time, MV – manual manipulated variable. The Easy800/MFD calculates and set corresponding output signal when there is difference between the set and actual values.

The controller equation is [6]:

 

 

(4)Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image008.gif

 

 

where: Y(n) – manipulated variable of the PID for current time (n); YP(n) – proportional component; YI(n) – integral component; YD(n) – differential component.

The value of YP is obtained by the gain proportional gain Kp and deviation of the equation. The last is the difference between the Set point at scan time (Xs) and the Actual value at scan time (Xi) for the current time. The following equation is used:

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image009.gif

(5)

 

 

Where: Õs – Set point at scan time; Õi – Actual value at scan time;

The integral component is calculated using equation (6):

 

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image010.gif

 

(6)

 

 

 

 

where: TC – scan time; Tn – reset time; Xs(n) – Set point at scan time n; Xi(n) – Actual value at scan time; YI(n-1) – Value of the integral component at scan time n-1;

 

For the differential block:

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image011.gif

(7)

 

 

 

Functional block PW – Pulse Width Modulation (Fig.5.C) generates square wave voltage and is normally used in conjunction with DC. Main settings of PW are: SV – Manipulated variable; PD – Period duration (0¸65535); ME – Minimum on duration (0¸65535).

One of the main functional blocks is PO-Pulse Output (Fig.5.D). It generates pulse output with maximum output frequency 3kHz or 5kHz depending on the used PLC. Its main application is in stepper motor control, but other uses are possible as well [3].

By using PO one can set exact number of pulses, corresponding to specific input value. The main settings are: I – input, sets the number of pulses; FS – starting frequency: (0¸5kHz); FO – operating frequency (0¸5kHz); RF, BF – coefficients that set the frequency change in acceleration phase and braking phase; P1 – Number of steps in Jog mode; QV – Actual number of steps completed; QF – Actual output frequency (0¸5kHz).

The number of pulses at turn on (nRRF) and turn off  (nRRF) are calculated as follows [6]:

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image012.gif

(8)

 

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image013.gif

 

(9)

 

 

 

By combining these blocks, the PLC can be set to work as PID or ON/OFF controller. When using the PO and PW functions, it should be noted that Easy/MFD system can control only two heaters. Uniform temperature distribution is obtained by regulating the supply voltage of the heaters, compensating their resistance change.

Usually the temperature control in CRF is done by using single thermo sensor, integrated in the furnace. The analog inputs of the PLC allow temperature sensor for each heater.

The schematic of the microcontroller block (MS) is shown on Fig.6, while the block diagram for the microcontroller is presented on Fig.7.

The experimental waveforms obtained during the tests are presented in Fig.9: À, Â – at high temperature, corresponding to low duty ratio d (waves 2 and 3); C – low temperature – high duty ratio d.

The improvement in the CRF is shown in Fig.8. One can see the more uniform temperature distribution of the SiC heaters, by implementing the individual control.

 

 

Description: Description: Description: image description

 

Fig.6. Electric diagram of the microcontroller system.

 

Description: Description: Description: image description

 

Fig.7. Block diagram of the microcontroller firmware

 

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image016.jpg

 

Fig.8. Temperature of the SiC heaters when using the proposed system.

 

Description: Description: Description: D:\Processmodeling.org\!!WebSite\article\fni\PLC based control system for high temperature chamber resistive furnaces with silicon carbide heaters_files\image017.gif

 

Fig.9. Waveforms from the designed system: 1 – output signal from the PLC; 2, 3 – signal for driving the output power transistors.

 

²²². Conclusion

 

By using the proposed system for high temperature chamber resistive furnace with SiC heaters is solved the problem for non uniform temperature distribution caused by the aging of the heaters. The difference can be seen on Fig.2 and Fig.8. This allows achieving the required parameters of the electro thermal apparatuses for the corresponding technological process. Besides this, it also extends the useful operation life of the SiC heaters and reduces the costs.

When using the modern PLCs, PID or ON/OFF control can be realized. The selected PLCs are low-end cheap alternative to the specialized controllers.

One of the major advantages of the proposed system converter-controller is the elimination of the low frequency, high power transform and the corresponding contactors. This greatly reduces the mass, volume and price of the system.