İhsan CEYLAN
Process Automation Division Manager
SICK Türkiye
The importance of effective kiln inlet process control is well understood. Optimising combustion control maximizes production throughput whilst minimizing fuel consumption, easily the biggest capital outlay for clinker production. A large cement plant can spend up to €50M annually on fuel, so optimization is vital to trim the energy bill. But it is easier said than done. The continuous gas analysis measurement at the kiln inlet is notoriously difficult, especially in terms of achieving high availability of reliable process data to assist process control operators. The importance of the measurement has driven plant operators to provide intensive support to maintain the kiln inlet analysis system, to the point that the necessary hours of daily manpower input are seen as “normal”.
Elevated process temperatures up to 1400°C combined with extreme dust loadings (to 2000 grammes / m3) place severe demands on the sample probe & sample system. Sample probe tube & sample probe filter blockages are commonplace and the probe is also threatened with build-ups around the probe or mechanical damage from falling materials. Plant operators must often accept daily intensive maintenance as an unavoidable necessity to keep the system operational.
Modern sampling probes are fitted with cooling systems (water or oil-based) to reduce sample temperature to avoid immediate oxidation of CO on metal surfaces & to protect the probe from thermal stress, as well as back-purging the probe sample filter with instrument air to keep it clean. However, in the majority of cases, sample probe designs in combination with simple cleaning systems are not sufficiently robust. So intensive maintenance input (both regular scheduled & non-scheduled) is still needed to clean blocked probe apertures and probe filters.
SICK has worked with a partner company to develop a more robust probe solution. A powerful “shock blower” (max 10 bar) targets the probe aperture with short blasts of compressed air. The probe aperture itself is carefully positioned & designed to make sure that after each back-purge, it is dust-free. During operation, the probe periodically rotates through 90 degrees and moves forward & backward by 20 cm to prevent becoming obstructed by build-ups within the kiln.
But the difficulties do not stop there. In more recent years, cement plants have increasingly turned to more cost-effective combustion materials, such as pet coke & alternative fuels to drive down fuel costs. These fuels bring with them additional challenges by increasing the presence of sulphates & chlorides in the kiln, which not only affect clinker quality and lead to greater build- ups, but also create a corrosive sample gas for the kiln inlet gas analyser.
The basic design of the kiln inlet analysis system has not changed significantly in the last 30 years. The sample system is centred around a refrigerating cooler, which creates an artificial dew point of typically 3°C, such that the gas exiting the cooler contains only 8000 ppm water vapour and will remain in the vapour phase as long as the ambient temperature in the cabinet is above 3°C. Further filters and condensate monitors remove smaller dust particles & check for moisture coming out of the cooler before the sample gas flows through a multi-component analyser by means of a sample pump with needle valve flow control. The multi-component analyser is an non-dispersive infra- red (NDIR) photometer capable of measuring CO & NO & sometimes SO2. An integrated paramagnetic or thermomagnetic module additionally measures oxygen.
The typical measuring ranges for the target gases are shown below.
CO 0 – 1 vol.%
NO 0 – 5000 ppm
SO2 0 – 5000 ppm
O2 0 – 5 vol.%
The NDIR photometer measuring cell design is normally based on two discrete optical channels. A sealed reference cuvette, filled with a reference gas (nitrogen) where no pre-absorption of infra-red light will occur. And a measuring cuvette, through which the sample gas flows and pre-absorption of infra-red light occurs according to the length of the cuvette and the concentration of the target gas (Beer Lambert Law).
The NDIR photometer cell design demands that the measuring cuvette is kept absolutely free of any contamination, i.e. in the same “perfect” optical condition, identical to the sealed reference cuvette. The presence of any dust particles or liquid droplets would immediately create a false attenuation of the infra-red light not related to the presence of the target gases, and so invalidate the measurement. This design requirement is at the heart of the challenge for the kiln inlet application. The sample system must provide the gas analyser with a sample absolutely free of any dust particles or liquid droplets. This explains why the sample system design found on the conventional kiln inlet gas analysis system features a complex assortment of different filter elements, each of slightly differing design in terms of removing dust particles of various micron sizes and/or liquid droplets.
There is another substantial potential challenge to the proper functioning of the conventional NDIR photometer, that of acid aerosol corrosion damage. Since the kiln inlet sample gas contains high concentrations of sulphur dioxide, it is possible for sulphuric acid aerosols to be created in the sampling system. Over time, these aerosols are liable to condense as tiny individual droplets inside the measuring cell, where they will attack the reflective surface of the measuring cell and significantly change its optical performance. This severe damage to the NDIR measuring cuvette causes the analyser to fail & requires cost-intensive replacement of the measuring cell & lengthy non-availability of the system.
To fully understand the issue of acid aerosols, it is useful to look at the history of sulphuric acid manufacture. Prior to the current “Contact Process” the method to manufacture sulphuric acid was the “Lead Chamber Process”, so-called because a lead reaction vessel was used. The heart of the sulphuric acid manufacturing process was the oxidation of SO2 molecules to SO3, catalysed by oxides of nitrogen (NOx) in the presence of air. The SO3 molecules then immediately react with water to produce sulphuric acid, H2SO4. Comparing the kiln inlet application, it is clear that all the participating gases (SO2 / NOx / O2 / H2O) are present in the sample gas. This explains why severe corrosion of NDIR photometers is a well-known problem associated with the kiln inlet application, since system were first delivered in the 1980s.
Various solutions have been investigated. Initially, acid filters were employed in the belief that they would act as a trap to the presence of any acid aerosols. But they were not able to exclude this reaction mechanism & the opinion was even stated that acid filters may offer a site for the acid aerosol reaction mechanism to take place! Super-coolers with a -30°C dew point have been tested, but acid aerosols were able to pass through the cooler, and even behind a -30°C cooler, there is still water vapour (approx. 400 ppm) present in the sample gas.
Some kiln inlet gas analysis suppliers have gone further in combatting this issue and offer dosing systems as standard. The dosing system is based on a liquid reagent being continuously fed via an inlet into the sample system.
The drawbacks to this solution are that it is clearly no longer possible to measure SO2 behind such a dosing system & the dosing system itself creates additional maintenance in terms of the additional moving parts (peristaltic pumps), the need to handle & replenish the liquid oxidizing reagent & especially in terms of handling the sulphuric acid liquid waste generated.
What is also important to acknowledge is that the concentration of SO2 in the process gas is a key parameter in terms of evaluating potential risk of acid aerosol formation and subsequent corrosion.
The increased use of alternative fuels, especially pet coke greatly increases the concentration of SO2 in the process gas & so increases the danger of acid corrosion in the sample system & NDIR gas analyser.
Until now, this problem has been combatted as described above.
However, there is now an alternative gas analyser technology. There are two principal differences in terms of the analyser design. Firstly, it is an infra-red filter photometer, but not an NDIR photometer.
The filter photometer measurement is based on measuring the absorption of infra red light at two different wavelengths for any given target.
A rotating wheel sequentially positions infra-red filters at a (reference) wavelength at which the target gas doesn’t absorb & a (measurement) wavelength at which the target gas does absorb. The difference in the light intensity is a function of
the concentration of the target gas. To measure several gas, such as CO, NO, SO2, a pair of measurement and reference filters is required for each component. Secondly, the significance of the measuring cell being thermostatted at 180°C or to a maximum 220°C, is that the cell temperature is comfortably above the acid dew- point of the sample gas. Therefore, even in the presence of acid gas aerosols, the temperature is such that the acid aerosols stay in the vapour phase & pass through the measuring cell without being deposited on the cell walls. The final safeguard is that the measuring cell inert oxide layer is resistant to any minor amount of sulphuric acid molecules which might be deposited & that the cuvette walls do not in any case participate to any significant degree to the overall optical measurement performance.
In the last years, this innovative combination of a more robust self-cleaning sample probe, with hot extractive analyser technology has been applied in a number of cement plants to the kiln inlet application, particularly those at which severe problems have been encountered with the conventional cold extractive solutions. As well as being inherently so robust, the analyser has built-in intelligence, which allows it to intiate automatic back- purge of the sample probe (for example if it detects a reduced flow-rate)
The results are very encouraging and appear to confirm that kiln inlet process control operators now have access to an alternative solution, which is capable of delivering the necessary availability with greatly reduced maintenance input.
It is anticipated that the application of this new technology will continue to broaden within the cement industry corresponding to the increase in diversification of applied alternative fuels.