Home About Us News Statistics Specifications Trade Shows Publications Stockwatch Research Papers



ADVANTAGES OF PULSE FIRING IN FUEL-FIRED FURNACES

FOR PRECISE LOW-TEMPERATURE CONTROL
(May 2002)

ANIL VASWANI, Senior Manager (Corporate Planning And Development)
DEBAJIT DAS, Senior Management Executive (Electrical and Instrumentation)

THE WESMAN ENGINEERING CO LIMITED-  WESMAN CENTER, 8 MAYFAIR ROAD, KOLKATA 700019, INDIA

 

The Wesman Group is among Indias leading manufacturers of industrial furnaces and combustion systems. Wesman has identified a market need for combustion systems capable of precise control even at very low temperatures. Such systems are typically required in specialized high-precision applications like heat treatment of forged rolls and vitrification of ceramic insulators. These applications require very low rates of temperature rise, and the total cycle time may run into two to three weeks or more. This type of temperature control is extremely difficult using conventional PID control, and the controllability range is only effective above 3000C. Because of the extremely long cycle time, the throughput of the furnace is very low. Consequently the cost of production as well as the cost of rejection is extremely high, further emphasizing the need for precise control to reduce the risk of rejection.

FIGURE 1: Profile of typical heat treatment cycle showing low rates of temperature rise, rapid cooling, and critical control at low temperatures.
Conventional fuel-fired furnace combustion systems designed for high temperatures cannot maintain these low temperatures effectively. This is because at the design stage, selection of burners is based on the maximum furnace load and the highest temperature to be attained. Conversely, conventional systems designed for low-temperature control cannot provide the necessary heat input required to attain higher temperatures. As these performance shortcomings are intrinsic to the inherent design principles of conventional combustion control systems, it would be instructive at this point to provide a brief overview of how a conventional system operates. In these systems the air/fuel mixture to burners is regulated through flow control valves operated by modulating motors (or alternatively by electro-pneumatic actuators), which are in turn driven by external PID controllers. The controller compares the actual furnace temperature against the setpoint and adjusts the percentage opening of the modulating motor to regulate the air/fuel flow rate to meet demand. Flame length and thermal energy input are therefore modulated on a continuous scale between the burners full-fire and highest turndown states on the basis of the controllers heat demand signal.

Why do conventional systems designed for high temperatures exhibit poor control at low temperatures? One reason for this comes from positioning errors inherent in the operation of modulating motors. A numerical example will illustrate this point with lucidity. Suppose our furnace has a total energy input of 1200 kcal, and the modulating motors have a positioning error of 1%. This error translates into 12 kcal over the entire operating range. However, for a lower temperature requiring only say 400 kcal of energy, this same 12 BTU error amounts to 3% of the required energy input. This small error therefore now becomes amplified by a factor of three. For lower temperatures, this control error assumes even larger magnitudes for conventional combustion systems. This leads to furnace temperature overshoots at low temperatures.

Conventional systems also have many mechanical linkages in the control loop between the PID instrumentation and the final control elements. In addition to delayed electromechanical reaction arising from the modulating motors, a major source of feedback delay is the air/fuel ratio regulator used to control the combustion ratio. This control delay leads to difficulties in controlling furnace temperature during heating or cooling cycles, especially when the rate of rise or fall is very slow.

Another shortcoming of conventional control systems is the restriction on burner turndown ratio, which has serious consequences for low-temperature control. Typically, for oil-fired burners the turndown ratio is limited to 1:5. This means that even when the burner is operated with its modulating motor in fully-closed position (when the PID controller decides that the process requires no additional heat input), the burner still provides the furnace with 20% of its rated energy capacity. This unwanted heat input will lead to temperature overshoots. For gas-fired burners this situation improves only marginally, as the turndown ratio is limited to 1:10 or a minimum energy input of 10%.

TABLE 1: Comparison between conventional combustion control systems and pulse firing systems.

CONVENTIONAL FIRING PULSE FIRING

Amplitude modulation

Turndown ratio is typically 1:5

Excess air absorbs heat energy, increasing fuel consumption.

In turned-down states, burners with special flame geometries suffer in performance.

Higher NOX level due to less mixing of combustion products with hot gas and excess air.

Poor heat distribution and penetration at low temperature.

Fixed heat source can cause local hot spots.

Temperature control usually within 15C.

Very poor control at low ramp rates or at low temperature.

Moderate initial investment.


Frequency modulation

Near-infinite turndown via control of pulsing frequency.

Air / fuel ratio is stoichiometric.


These problems do not exist, as burners are not fired in intermediary modes.

Lower NOX level at high fire due to better mixing of combustion products with hot gas.

No such problems because full stirring energy is utilized.


Moving heat source eliminates local hot spots.

Temperature control within 3C achievable.

Good control at low ramp rates or at low temperature.


Higher investment due to instrumentation and valves.

We return to our 1200-kilocalorie furnace example to illustrate this point. Our furnace has 12 burners, each providing 100 kcal in full-fire state. With a turndown ratio of 1:5, each burner when ignited will provide at least 20 kcal of heat input. When all 12 burners are ignited, the minimum total energy is therefore 240 kcal. For process temperatures requiring less than this minimum heat input, there will be temperature overshoots unless some of these burners are manually switched off. However, if the process subsequently passes into a high-temperature zone (which is a common feature in the sophisticated heat treatment cycles in use today) the extinguished burners will need to be re-ignited to provide the increased heat demand.

What problems does this continual extinguishing and re-igniting cause? First, because the burner blocks are in a relatively "cold" state, there is a risk of ignition failure at lower temperatures, which can jeopardize the process. Frequent manual intervention will also be required to supervise the re-ignition, although automatic re-ignition systems can be found in use today. Another shortcoming is that continual re-ignition will reduce the life of the ignition transformer, while the periodic thermal shocks will cause mechanical damage to the burner blocks and tiles. Switching off burners to prevent overshoots can cause yet another problem of non-uniform temperature distribution. This occurs because heat is now provided into the furnace through fewer, more localized areas rather than all along the length of the zone. One solution to this problem is to rotate or alternate the firing and non-firing burners so that there is a moving thermal head. However, this will impose further stresses on the ignition transformers, burner blocks, and other components.

Pulse firing combustion systems overcome these shortcomings, providing precise control over a wide range of temperatures from 150C to 1200C. These systems have been in use in Europe and the United States for the last fifteen years, where it has been well established that temperature control of this nature can be achieved very effectively and reliably with pulse firing technology. Like a conventional combustion system, a pulse firing system takes its furnace demand signal from a PID controller. However, this is where all resemblance ends. Unlike a conventional system where flame length and thermal energy input can have any continuous (analog) value, a pulse firing system operates only in two discrete and mutually exclusive states, termed high and low. These states are characterized by a predetermined fuel flow rate, which is achieved through a specially designed set of pipelines. The high state corresponds to the full thermal capacity of the burner. The low state may be either a completely extinguished burner, or a very low heat input just enough to sustain the flame.

The heat demand signal from the temperature controller PID loop is translated into an equivalent duty cycle factor, which determines the frequency

FIGURE 2: Schematic showing pulse firing sequence for four burners and timing at which each burner in a zone is fired. The frequency at which the burner switches between its pre-selected high and low states is proportional to the heat demand. Different start times are assigned to the burners to ensure that they pulse in sequence, creating a moving thermal head in each zone. For example, for a four-burner zone with a total firing cycle time of 20 seconds, a 40% heat demand signal from the PID means that each burner will be in high (full-fire) mode for 8 seconds and in low (turndown) mode for the remaining 12 seconds. Additionally, each burner will go into full-fire mode 5 seconds after the previous burner. The moving thermal head thereby created in the furnace has the effect of uniformly distributing heat input. Hot spot formation is also prevented because any given burner in a zone is only fired actively for a given period of time before the firing sequence activates another burner in a different position in the same zone.

A pulse firing system has many other significant advantages over a conventional fuel-fired combustion control system. Some of these are:

IMPROVED HEAT CIRCULATION AND PENETRATION: Effective transfer of heat from furnace to charge is achieved primarily by rapid circulation of the furnace flue gases around the charge. This circulation is itself a result of the stirring energy imparted to the medium by the velocity of the gas stream coming out of the burners. The stirring energy is proportional to the cube of the velocity of the gas stream. By way of illustration, reducing the burners to 20% of their capacity would therefore drastically reduce the stirring energy to 1/125th of its original value. In a conventional system, burners are usually not fired at their maximum capacity, and consequently the stirring energy of the gas stream is not fully utilized. Excess air conditions can be created to improve penetration and circulation, but this consumes additional energy. A pulse firing system eliminates intermediary firing levels and substitutes them with a full-fire mode with unthrottled air flow. The higher combustion air velocity and kinetic energy are transferred to the flue gas, resulting in better circulation of hot air within the furnace as well as better heat distribution and penetration. These lead to temperature uniformity throughout the furnace over the entire control range.

NEAR-INFINITE TURNDOWN: We have already seen that in a conventional oil-fired combustion system the highest turndown ratio is typically 1:5, so that a minimum of 20% of the rated heat energy still enters the furnace even when all burners are turned down. This may be greater than the process requirement and can lead to temperature overshoots, especially at low temperatures. In a pulse firing system, the pulse control algorithm automatically adjusts the ratio of high-fire duration to low-fire duration depending on heat demand so that turndown ratios as high as

TABLE 2: Sample data from STEP3 pulse firing test conducted at Wesmans R&D furnace testbed on 3rd August 1999

TIME PID OUTPUT TEMP SP LOAD
TC AVG
TEMP DEV AIR TC
15:59:00 8.8 161.9 163.0 1.1 186
15:59:10 8.6 162.0 163.5 1.5 184
15:59:20 8.4 162.1 163.5 1.4 183
15:59:30 8.3 162.3 163.5 1.2 182
15:59:40 8.2 162.4 163.5 1.1 181
15:59:50 8.1 162.5 163.3 0.8 179
16:00:00 8.3 162.6 162.6 0.0 177
16:00:10 8.3 162.7 162.9 0.2 175
16:00:20 8.3 162.8 163.0 0.2 174
16:00:30 8.3 163.0 163.0 0.0 172
16:00:40 8.3 163.1 163.3 0.2 171
16:00:50 8.3 163.2 163.5 0.3 170
16:01:00 8.3 163.3 163.5 0.2 168
16:01:10 8.3 163.4 163.5 0.1 168
16:01:20 8.3 163.5 163.5 0.0 166
16:01:30 8.3 163.7 163.5 -0.2 166
16:01:40 8.3 163.8 163.5 -0.3 165
16:01:50 8.3 163.9 163.4 -0.5 164
16:02:00 8.3 164.0 163.3 -0.7 163
16:02:10 8.3 164.1 163.2 -0.9 162
16:02:20 8.3 164.2 163.5 -0.7 162
16:02:30 8.3 164.4 163.5 -0.9 162
16:02:40 8.3 164.5 163.5 -1.0 162
16:02:50 8.7 164.6 163.4 -1.2 161
16:03:00 8.7 164.7 164.0 -0.7 160

 

FIG. 3: Control histogram of data obtained from R&D test of pulse firing system at Wesmans furnace testbed.

1:50 (or 2% of rated heat energy) are possible. If the low-pulse state indicates a completely extinguished burner instead of a turned down state, then even higher (and potentially near-infinite) turndowns are possible.

PRECISE LOW-TEMPERATURE CONTROL: Because of the extremely high turndowns achievable with a pulse firing system, furnace and load temperatures can be controlled even at as low as 150C to 200C, which is unattainable with conventional combustion systems. Additionally, as a result of the moving thermal head and enhanced flue gas circulation, temperature uniformity of up to +/-3C or better can be achieved in practice.

IMPROVED BURNER EFFICIENCY: Burners are designed to deliver their most efficient performance at full-fire, where almost complete combustion of fuel takes place. In a conventional system, when burners operate at intermediate ranges, incomplete combustion occurs. Pulse firing reduces this problem by eliminating intermediary firing conditions.

LOWER SPECIFIC ENERGY CONSUMPTION: Five separate mechanisms contribute to improve fuel consumption in a pulse firing system. First, burners operate at their highest efficiency throughout their control range. Second, because of improved heat transfer due to better circulation and penetration, less fuel is required to bring the load up to the desired temperature. Third, precise low-temperature control allows exact matching of heat input to the furnace demand, reducing temperature overshoots and their corresponding energy wastage. Fourth, pulse firing systems do not use slow-acting control devices such as modulating motors, and can therefore react dynamically to changes in heat demand, further reducing energy wastage. Fifth, conventional systems employ mechanical air/fuel ratio regulators, which raise the lower limit on the permissible minimum amount of fuel consumed in turndown mode. Pulse firing systems utilize solenoid valves, which have no such restrictions and consume less fuel at turndown.

With the objective of bringing this highly effective technology and its benefits to Indian industries, Wesman has developed its own in-house pulse firing solution, which we have christened STEP3 technology. In our system the dual-state high-low pulse firing concept has been refined and taken one step further to a triple-state low-mid-high firing system (hence the name STEP3) which offers even better control and uniformity. For low-temperature applications, the burners switch between low-fire and mid-fire. At higher temperatures they automatically switch between low-fire and high-fire. The duty cycle factor of the burners will be automatically adjusted by the STEP3 control system to achieve very high rates of rise whenever required.

STEP3 uses a unique direct-spark ignition system that eliminates dependence on pilot gas burners even when using high-speed diesel (HSD). This reduces the cost of the LPG installation and accessories associated with pilot burners. The system can also run purely on LPG or on HSD/LPG dual-fuel systems. STEP3 can be offered as part of a new furnace project or as a revamping package for an existing furnace.

STEP3 is designed around a dedicated PLC program developed in-house at Wesman and uses intelligent routines to perform all sequence, process control, alarm and diagnostic functions. PID functions for each zone are also implemented within the PLC, which eliminates the need for external hardware controllers. A graphic control screen allows operators to interact directly with

FIGURE 4: Segment 1 (140C to150C @ 10C/hour) and segment 2 (150C to 200C @ 20C/hour), showing precise load temperature control at low temperatures as well as low rates of temperature rise.

FIGURE 5: Segment 3 (200C to 275C @ 80C/hour) showing precise load temperature control with the same system at a higher ramp rate.

FIGURE 6: Segment 4 (soaking at 275C) and segment 5 (275C to 245C @ -65C/hour) showing close adherence of load temperature to setpoint even at rapid rates of controlled cooling. the process. Dynamic graphics and animation help to visualize the process and firing sequence in real-time. Process inputs to the PLC are fed through programmed buttons on the interface, eliminating the need for hardware like LEDs, toggle switches, selector switches, keypads and pushbuttons on the control panel. Short-term trending of process parameters is displayed on the screen. Process and control functions are logically grouped on separate screen pages. Other features of STEP 3 include:

SEQUENTIAL AUTO-IGNITION AND FLAME MONITORING: After the furnace has automatically been purged, the PLC will attempt to ignite each burner one by one. Three successive ignition attempts are made for each burner. Burners which fail to ignite after three attempts are put into lockout mode, which requires manual intervention in order to re-ignite them. Continuous flame monitoring takes place during normal operation. If a burner should accidentally extinguish, the ignition sequence for that particular burner is automatically restarted.

BURNER AUTO-SHUTDOWN AND RE-IGNITION: The control algorithm utilizes an intelligent routine that continuously monitors the rate of rise of temperature, and extrapolates to calculate projected temperature overshoots or undershoots. Even with burners at maximum turndown, if the projected temperature overshoot exceeds and consistently remains above a programmed margin, then the PLC gradually shuts down burners one by one until the overshoot condition is removed. The shutdown condition is distributed throughout the burners in a cyclic way to avoid prolonged cold spots and uneven heat distribution. Similarly, with burners in this anticipative auto-shutdown mode, if a temperature undershoot is projected, the extinguished burners will gradually be automatically re-ignited. It is important to emphasize that the PLC takes these corrective actions in preventive anticipation before the temperature of the charge ever actually crosses the acceptable temperature deviation limits.

n SELECTIVE MANUAL EXCLUSION OF INDIVIDUAL BURNERS: Individual burners can be selectively deactivated and excluded from the system at the discretion of the operator, in order to adjust the total heat input required for different charges or cycles. The firing sequence for the remaining active burners automatically readjusts to ensure uniform heat distribution.

n CONTROL THROUGH AIR OR LOAD THERMOCOUPLES: Operators have the option of controlling the system either with air or with load thermocouples. STEP3 uses at least two air and load thermocouples per zone. Temperature control is achieved by taking an average of these readings. If one thermocouple fails in operation, control can still be achieved using the other healthy thermocouple.

INTERNAL SETPOINT GENERATOR: Up to twenty segments can typically be entered into an internal setpoint generator, which is part of the standard PLC program code. The setpoint parameter inputs are made through the interactive graphical display, eliminating the need for an external hardware programmer. Features include run/hold mode, forward/reverse mode, high-speed/low-speed mode, and segment advance/retract.

USER-SELECTABLE FLEXIBLE FIRING ZONES: Because STEP3 relies on programmed software code instead of hardwired logic, field inputs can be interconnected flexibly within the PLC. This allows several pre-programmed configurations of burner zones to be stored and activated by the user depending on the mass distribution and other properties of the charge. The end result is better temperature uniformity across the surface of the charge, even for charges with irregular or complex geometries, such as forged rolls.

FAULT DIAGNOSTICS AND ON-LINE HELP SCREENS: Alarm pages are programmed on the display screen to indicate alarm conditions and status as well as textual messages which inform the user regarding the probable causes of the alarm as well as recommended remedial actions. These alarms cover not only process parameters such as flame failure, ignition failure and temperature abnormalities, but also the operational status of components like temperature sensors, blowers and pressure switches.

DATA LOGGING TO PRINTER: In addition to graphical display of process parameter trending on the control screen, process data may also be sent as lines of digital text to a dot-matrix printer. This provides a long-term hard copy of the process history for the users reference. Alarm details as well as date and time of their occurrence are also registered on the printer.

Terms & Conditions Disclaimer Privacy Policy