Modern regenerative side or end port furnaces as well as recuperative furnaces profit from a stable, controlled combustion process. Such a stable and precisely controlled process will result in improved furnace lifetime as well as optimum, efficient fuel consumption. Energy saving up to 2% can be achieved which will result in a fast return on investment. The glass melting process will also profit from a stable furnace combustion space environment.
Keywords:
-
Excess of air
-
Parasitic air
-
Oxygen control
-
Glass furnace efficiency improvement
-
Stoichiometric combustion
To understand the principle of excess of air control we first need to understand how the combustion in a regenerative glass furnace is normally controlled.
Fuel flow (2); oil, natural gas or a mixture of both, is normally well measured and normalized (Nm3, kg or ideally; caloric values) mainly because of the financial value the fuel represents and the importance to balance the fuel consumption against the furnace pull rate. Fuel-consumption/pull is one of the most important bench marking figures in the industry. However, without the availability of a precise caloric value or fuel composition values, it will still be difficult to control the combustion. In other words; the fuel-air ratio would need to represent a precise amount of oxygen molecules needed to oxidize all carbon and hydrogen into CO2 and H2O.
Calculating and controlling the combustion air and fuel ratio, having the precise fuel and combustion air flow parameters would be an easy job to do if there was not what we call “the parasitic air”; air entering or leaving the regenerator and furnace system without being measured by the combustion air venturi. Since the whole regenerator and furnace super structure is built out of refractory bricks, there are obviously minor/major opportunities for air leaking into/out of the system. Also burner purge air needs to be taken into account.
Conclusion: Even if we measure the combustion air at the inlet of the regenerator very precisely, this value doesn’t represent precisely the amount of air available for combustion taking place in the combustion space of the furnace.
The exhaust flue gas oxygen value
According to Nernts:
If oxygen is to be measured, using a zirconium ceramic electrolyte and atmospheric air, with pO2 = 0.209, as a reference, this equation simplifies to
pO2 = 0.209 exp(-46.421 E/T)
The equation may be re-arranged, to allow calculation of the emf from a known reference concentration and a measured oxygen concentration:
E = 0.0496 T log(pO2/0.209 )
Since “T” represents the probe temperature there is a type R, S or B thermocouple inside the probe near the ZrO2 sensor.
Invensys-Eurotherm T2550/2750 system has ZI - Zirconia input module
Input Types: Analogue voltage, Channel 1 - mV (TC), and Channel 2 - 2V (Zirconia probe)
Thermocouple Input Specification (Ch1 ONLY)
Input Range: -77mV to +100mV
Zirconia Input Specification (Ch2 ONLY)
Input Range: -10mV to +1800mV
There are several significant advantages of controlling the excess air. First of all, next to the 20.9% oxygen which we need for combustion, the air also contains 78% nitrogen. Nitrogen is of no use to the combustion process but it has to be heated up to combustion temperature. In fact the combustion process produces, next to CO2, H2O, a lot of very hot useless nitrogen. Like nitrogen, the excess of oxygen represents an efficiency loss and has a nasty side effect as it tends to react with nitrogen at a high combustion temperature resulting in unwanted NOx emissions.
Generally: Using the exhaust flue gas oxygen values coming from the ZrO2 probe to tune the combustion air input to a level at which the air available for combustion, (thus including the parasitic air) together with the applied fuel reaches stoichiometry will increase fuel efficiency and improve NOx emissions.
Control considerations
The glass furnace melting process is very complex and one of the most important control considerations is try to control the process as smoothly as possible. Every fast control action will result in the introduction of unwanted disturbances and side effects. The control of combustion is already disturbed by the inevitable reversal sequence. However, the regenerative principal gives us an efficiency improvement of 80% so there is no way of working around that disturbance.
To improve the efficiency of the combustion process the fuel to air ratio controller has the task of finely adjusting the combustion process as close as possible to a stoichiometric value without the production of CO. Therefore the fuel /air ratio controller needs to take the parasitic air into account which is in fact the difference between the calculated value of air available at combustion and the actual combustion air available at combustion, measured by the oxygen probe in the exhaust. Such a combustion controller normally uses the Lambda value for control which is the ratio of actual air fuel ratio to stoichiometry for a given mixture. Lambda of 1.0 is at stoichiometry. Since parasitic air tends to vary during the time to reversal and will not be the same for both sides of firing, the controller needs to act differently for each firing side and will also need to store the historic values, the once from the time it was controlling that specific side before.
Concluding; although excess of air control is complex, a modern controller or process control system is perfectly capable of managing such a sophisticated strategy. Perhaps the most important part is to educate the furnace operators in order to make them understand how such a control works, to monitor its behavior and judge possible malfunctions. If applied correctly it will provide stable combustion, improved fuel efficiency and reduced emissions of NOx.