Previous Post
Next Post





An electrostatic precipitator utilises electric forces to separate solid particles, or liquid droplets, from the gas in which they are suspended. The process gas, bearing particulates, is forced in the precipitator to pass over the discharge electrodes between a number of channels formed by collector electrode sheets suspended vertically. Electrostatic forces cause particulates to migrate towards these collector electrodes, whilst being borne through the length of the precipitator. Much of the particulate material adheres to the collector electrodes, leaving the carrier gas relatively clean. From time to time, both the discharge and the collector electrodes have to be mechanically rapped in order to dislodge accumulated particulate material, which then falls into a hopper and is conveyed away. Figure 1 shows a typical electrostatic precipitator with a cut away view showing the internals. The removal of dust or liquid droplets from the gas stream is dependent on the combination of five separate processes (refer to Figure 2). These are:

  1. The, imposition, by means of a corona discharge, of an electric charge on each dust particle, the polarity of which must be opposite to the polarity of the collector electrodes.
  2. The formation of an electric field which will react with the charged dust particles so that they are caused to move towards the collector electrodes.
  3. The deposition of the dust particles on the collector electrodes, their loss of charge, and subsequent agglomeration.
  4. The removal of the agglomerated dust particles from the collector electrodes Into the dust hopper.
  5. The extraction of the dust from the dust hopper.

A deficiency in any one of these processes will cause a loss of collection efficiency, and this will result in increased dust emissions. In order to optimise the performance of a precipitator it is necessary to be cognisant of all five actions and to relate to them the mechanical and electrical qualities of the precipitator as well as the nature of the gas and dust being treated.


FLS electrostatic type f
FLS electrostatic type f


figure two electrostatic precipitator
figure two electrostatic precipitator



2.1       Corona Formation

Particle charging occurs naturally but these charges are too small for effective precipitation. The most effective method of charging is by high voltage direct current corona. The corona is established between barbed or wire discharge electrodes maintained at a high negative voltage and plate or tube electrodes at ground potential. The negative corona may be visualised as consisting of two zones (see Figure 3). The active zone occurs around each discharge electrode where there is electrical breakdown of the gas and electrons are emitted which ionise the gas. The passive zone is the region between the active zones and the collector electrodes. This zone contains a dense cloud of rapidly moving gaseous ions of the same polarity as the discharge electrodes. These gaseous ions are repelled by the discharge electrodes and are attracted towards the positively charged collecting electrodes.

2.2                         Size and Electric Field Distribution

Establishment of the corona depends upon the electric field distribution in the inter- electrode space being non-uniform. This in turn is dependent on the dimensional difference of the electrodes. Similar size and shape electrodes would result in a uniform field existing and no corona would be formed. The field intensity varies from high at the discharge electrode to low at the collecting electrode (refer to

Figure 4). When too high a voltage is applied to the discharge electrodes there is a complete breakdown of the gas and a flash-over occurs. If this were allowed to persist a short circuit would occur between the electrodes and precipitation efficiency would be reduced greatly. For maximum efficiency the electrode voltage should be maintained at a level necessary to achieve a controlled rate of sparking of typically 50 sparks per minute.

2.3                       Charging and Collection

Within the passive zone a small proportion of the dense cloud of rapidly moving lonised gas molecules collide with and attach themselves to dust particles. These dust particles thus become negatively charged and are attracted towards the collecting plates. They adhere to the plates and then give up their charge.




A key factor in determining whether or not a dust may be separated from a gas stream is the electrical resistivity of the dust. This is defined as the resistance to current flow of a one centimetre cube of dust, when a potential is applied uniformly over opposite faces of the cube. Resistivity is measured in units of ohm.cm. The resistivity of a dust
is significantly affected by the gas conditions under which it is measured. When measuring dust resistivity it is necessary to simulate the actual or predicted conditions within the precipitator.
It would be necessary to inform the laboratory performing the resistivity tests of the actual operating temperature and gas humidity. For a preheater or precalciner kiln plant it would be necessary to give the gas conditions for both when the raw mill is operating and when it is stopped. Dusts with resistivities that fall between 106 and 1011 ohm.cm are unlikely to give rise to precipitation problems. Dusts with resistivities less than 106 ohrn.cm lose their charge so rapidly that they are repelled back into the gas stream by the collector electrodes. This re-entrainment results in increased dust emissions.

Many dusts in the cement industry have resistivities higher than 10 Figure 5). Such particles once charged, are reluctant to lose their charge to the collector and thus adhere strongly and resist dislodgement. The resulting layer of
negatively charged particles on the surface of the collecting electrodes repels the approaching charged particles and thus reduces the rate of precipitation. The dust layer on the collector also impedes the flow of current which results in a voltage drop across the layer. This in turn reduces the voltage drop between the discharge electrode and
the surf ace of the dust (the precipitating voltage), which again reduces the rate of precipitation. If the voltage drop across the dust layer is greater than the breakdown voltage, then a reverse ionisation condition will exist. This results from a positive corona discharge at the point of breakdown of the dust. Precipitation efficiency is reduced by the lowering of the flash-over voltage and by producing positive ions which decrease the rate of particle charging.

When operating conditions are around 300oC the resistivity of a dust is dependent on the composition of the dust itself. Below 2000C resistivity is more dependent on the adsorbed film on the particle surface. Figure 6 Illustrates the effect of adsorbed moisture on the resistivity of a dust. From Figure 5 it can be seen that peak resistivities occur between about 200OC and 3000C.

In general, it may be said, that for gas temperatures below about 3000C, the greater the difference between the temperature of a gas and its water dew-point temperature, the more difficult will be the process of electrostatic precipitation.

A relationship has been found between dust resistivity and the content of water soluble alkalies in the dust. The higher the soluble alkali content of the dust, generally the lower the dust resistivity. Alkalies occur in the raw materials fed into the kiln and from the fuels burnt.

In wet and long dry kilns, alkalies are condensed on the kiln feed at the back-end of the kiln. A portion of this material is picked up by the kiln exhaust gases and is carried to


the precipitator. Some cement works analyse the alkali content of dust collected in each field of the precipitator. Using this information they can then decide if the dust from particular fields can be returned to the kiln, diverted to the cement mill section, or other associated plants.

In preheater and precalciner kilns the alkalies condense on the kiln feed material in the preheater and are re-evaporated at the back-end of the kiln. This sets up an internal alkali cycle in the preheater which results in a relatively low alkali content in the dust passed to the precipitator. This along with the high temperature and low moisture content of the gas is the reason for high dust resistivity.


4.1 General Arrangement

Figure 1 shows a typical arrangement for an electrostatic precipitator. Precipitators come in a variety of configurations generally dependent on the duties they have been

designed for.                                        Typically a precipitator has 2 to 4 fields with a separate
transformer/rectifier set per field, however some precipitators have split fields with a separate t/r set for each half.

Precipitator casings are usually constructed from 5 to 6 mm mild steel plate, the plate thickness employed, however, depends upon the suction to be applied to the precipitator. For corrosive environments stainless steels (3CR12 or 18/8) are often used. Older precipitators on Lepol kilns had casings constructed from concrete, lined with acid resistant bricks to resist attack from the kiln gases.

Hoppers under the precipitator can either be longitudinal, transverse or pyramidal. The precipitator application determines the arrangement required. Dust collected In the hoppers is continuously extracted by screw or drag chain conveyors.

Various arrangements for the inlet transition to the precipitator are shown in Figure 7. The choice of inlet is usually determined by the space available in front of the precipitator, the configuration of the ducting from the upstream equipment and the dust concentration of the gases.

Gas distribution is carried out by profiled steel plates or angles and perforated plates or grids in the precipitator inlet. Their function is to distribute evenly the gas across the total cross section of the precipitator.

Collector electrodes with surfaces fully exposed to the gas stream would have poor particle retention. Modern collector electrode designs provide aerodynamic shielding for the collection zones (refer to Figure 8). This profiling of the plates contributes significantly to their mechanical stiffness. Maximum recommended collector plate length is 15m constructed usually from 1.2 to 1.5mm thick plate. For long collector plates location pieces are normally used at 1/3 and 2/3 the plate length to prevent


deflection. Collector plates are usually fabricated from mild steel with stainless steel being used for corrosive environments. The plates are fastened to frame elements and spacers to maintain the constancy of the spacing between them. They are all connected to an efficient electrical earth.

Modern discharge electrodes are usually of the rigid mast design (refer to Figure 9). The discharge electrodes are suspended, between the collector plates, from a framework which is supported upon ceramic electrical insulators. The precipitating voltage is applied to this frame. Beneath the collector plates the discharge electrodes are located by another insulated frame. Discharge electrodes may consist of wires (see Figure 10) which are maintained in tension by cast iron weights. The breakage of a single such wire can put a whole section of the precipitator out of use, so modern discharge electrodes are designed to be more robust. They are usually made from rolled sheet metal incorporating a number of sharp edges or spikes, which provide the seats of the generation of the corona discharge.

Wire electrodes are able to be replaced with the rigid mast design as and when suitable opportunities ocCur.Discharge electrodes are normally made from aluminium quenched mild steel;stainless steel should be used for corrosive gases.

All precipitators have an external layer of insulation material to reduce gas cooling occurring on the inside walls of the precipitator. This lagging is protected from weathering by a cladding of sheet alurninium. For cold and wet climates, such as in the U. K., an insulation thickness of 150mm is recommended, using insulation with a density
of 90 kg/m3. Warmer or dryer climate countries can use an insulation thickness of 100mm with an insulation density of 60 kg/m3.

4.2 Electrode Alignment

The rate of charging of the dust particles and their movement towards the collector electrodes is a function of the voltage applied to the electrode system. The higher the voltage, the higher the collection efficiency should be. The magnitude of the voltage that can be applied to a given electrode system is largely dependent on the alignment of the electrodes and their surface condition.

Collector electrodes must be evenly spaced, truly vertical, without bows, dents or any other form of distortion. There must be no sharp projections, cracks or holes which could give rise to the formation of a localised corona discharge.

Every wire type of discharge electrode must be properly tensioned and should be positioned exactly on the centreline between adjacent collector plates, or in the case of tube collectors, exactly on the centre line of its tube. Rigid mast types of discharge electrode must be straight and securely located.

It is inevitable that the normal manufacturing tolerances resulting from the forming of plate and tube collectors will prevent precise alignment, but every effort must be made to achieve the highest possible standard, with particular attention being paid to the


positioning and alignment of the electrode suspension points, and any location devices or guide systems that may be provided.

Typically, precipitator manufacturers specify a collector electrode straightness better than L/1000 where L=the collector length. This provides a guide to the alignment tolerance that should be aimed at.





This alignment tolerance relates to plate or tube collectors with spring or weight tensioned thin wire discharge electrodes. Where rigid or semi-rigid discharge electrodes are used, manufacturing tolerances similar to those of the collector electrodes can be expected. This makes alignment more difficult, since the manufacturing tolerances can summate. For 10 metre high collector electrodes, the minimum distance between the discharge electrodes and the collector electrodes can, therefore, be half the collector spacing minus+/一20mm, or twice the tolerance that can be achieved with thin wire electrodes. In practice, it is quite common for bowed electrodes to be matched in order to minimise the alignment error, although this can be tedious if a large number is involved.

The prime objective is to achieve a uniform discharge electrode to collector electrode spacing throughout the precipitator, since this will permit the highest voltage to be applied. If a precipitator contains electrodes which are excessively bowed, causing the discharge electrode to collector electrode spacing to be exceptionally small, the bowed electrodes must either be replaced or, if the number is small, the discharge electrodes must be removed from the damaged areas. In a plate type precipitator, the removal of a small number of discharge electrodes will have an insignificant effect on the performance of the precipitator, whereas the presence of just one discharge electrode with a reduced spacing will be very detrimental. In a tube type precipitator, it is essential that every tube from which a discharge electrode has been removed, is blanked off to prevent the passage to atmosphere of untreated gas.

To maintain the alignment of the electrode system against the influence Of the gas flow, electrostatic forces, etc, the top and bottom discharge electrode support frames must be cross-braced to provide a rigid box structure and the bottom frame must be equipped with insulated tie bars or stabilisers attached to the precipitator casing and carefully adjusted to prevent movement.

Maximum collection efficiency requires the application of maximum voltage, which is usually accompanied by some measure of sparking. Indeed, random sparking is considered to be a sign of a healthy precipitator. It is important however, that this sparking be confined to the body of the electrode system and, adequate electrical clearances must therefore be maintained throughout the connection between the rectifier set and the discharge electrode system.

4.3 Electrode Rapping

Vibrator type rapping devices must be firmly mounted and rigidly connected to the electrode systems. Every effort must be made to achieve maximum energy transfer to the electrodes with no losses due, for example, to contact of connecting rods with the precipitator casing. Correct adjustment of the vibrators is essential and particularly applies to the armature gaps of the electro-magnetic types.

The hammers of drop and tumble hammer rappers should be free falling and the hammer anvils should be securely fixed to the electrodes. The lift of the drop hammers should be checked and any worn lifting mechanism parts replaced.

The timing device must operate correctly and provide the facility to adjust the interval between successive raps and the duration of each rap.

Adjustthent of the rapping gear (amplitude, duration of rapping and interval between successive raps) is largely a matter for experimentation. There are three main rapping systems:-

Electrostatic vibrator or out-of-balance electric motor type.

Tumble hammers mounted on a drive shaft spanning the width of the precipitator casing.

.         Motorised cam operated devices.

For the collectors, the objective is to find the optimum between rapping too frequently, thus promoting re-entrainment, and not rapping frequently enough, which will lead to an excessive build-up of dust on the collecting surfaces and a reduction in precipitator efficiency. It should be noted that a thin layer of dust is always present under normal operating conditions. For the discharge electrodes, the object is to maintain the electrodes in as clean a condition as possible, since any build up in excess of a film will tend to suppress the corona, and again, reduce the efficiency of the precipitator.

In a precipitator with a number of fields in series it will be necessary to grade the rapping of the collector electrodes, with the frequency of rapping being highest on the inlet field and decreasing towards the outlet. It is also a requirement that the rapping sequence be adjusted to prevent the collector electrodes in all of the fields being rapped simultaneously.

The type of rapping device and the rapping force required is largely a function of the nature of the dust being collected. Chloride bearing dusts tend to be sticky and thus require a higher rapping force for effective removal. Collector plate hammer design should be adequate to give 200 G at the point of impact and 80 G at the farthest point.

4.4 Dust Extraction and Transport

Dust collected in the precipitator dust hoppers must be continuously extracted and the hoppers must not be used as storage vessels. If dust is allowed to build up in a: hopper and comes into contact with the electrode system the rectifier set will be short circuited and precipitation will cease. More importantly, if the dust rises up the electrode system to any significant depth there is the risk that its subsequent removal will impose uneven sideways forces on the electrodes, causing distortion and alignment problems that will be difficult to correct.

As mentioned in Section 4.1,there are three main hopper arrangements一longitudinal, transverse and pyramidal. The longitudinal arrangement is often preferred because the number of extraction conveyors can be reduced. For applications where it is a requirement to separate dust collected by different fields (on wet process kilns) transverse, or pyramidal arrangements are the preferred options. These two arrangements also offer the ability to carry out maintenance on conveyors under individual fields during kiln operation. Hoppers must be d esigned for the types of dust they are likely to handle in the future. If you intend to use a fuel in the future with a relatively high chloride content (4% Cl), then this will have an effect of the flowability of the dust to be collected. Chloride dusts require hoppers with steeply sloped sides. It is normal for hoppers to have heaters to stop moisture condensing during the heating up and cooling down cycles of the kiln. Excessive condensation would lead to corrosion of the hoppers, as well as blockages.

Typically screw conveyors or drag chains are employed to transport the dust from the dust hoppers back either to the kiln or raw mill circuit or to a dust collection silo.



4.5       Gas Distribution

In an installation where two or more similar precipitators are operated in parallel, control dampers must be fitted and adjusted so that the gas flow is equally divided between them, although it should be noted that this does not imply an equal division of the dust load. Gas flow measuring devices are not usually employed, but where each precipitator is equipped with a separate ID fan, a check on the gas balance can often be obtained from the fan motor ammeter readings.

The. gas distribution within individual precipitators is usually determined by model tests made before construction of the precipitator, followed by on-site measurements and fine adjustments at the time of commissioning. Ordinarily, no further ad justrnents need to be made, although it is, of course, essential that the gas distribution devices be maintained in a good state of repair. If it is suspected that the gas distribution is poor, it will be necessary to make a gas velocity survey of the precipitator and this work would probably best be carried out by the manufacturer of the precipitator who can

then also make any adjustments that may be necessary. Typically, variations in gas velocity of土25 to 30% of the average gas velocity are acceptable.

Apart from the need to ensure that the gas flow through the precipitator is reasonably evenly distributed and that there are no areas of excessively high or excessively low gas velocity, there is also the need to control the gas stream so that all of the gas passes through the electrode system to ensure complete treatment. There must be no short circuiting of gas under, over, or around the electrode system. In the case of a tube type precipitator, it is only necessary to ensure that the tube plate forms a gas tight seal between the tubes and the precipitator casing. In the case of a plate type precipitator, a system of gas baffles is employed to guide the gas flow and these must be maintained in a sound condition. Side baffles are used to prevent the passage of gas through the space between the outer collector plates and the precipitator casing and top baffles, sometimes making use of the roof beams, are used to prevent the passage of gas over the top of the electrode system. Gas baffles at the bottom of the precipitator are usually unnecessary when pyramid or transverse trough hoppers are used. With longitudinal trough hoppers, however, baffles are essential. These must be positioned between the field systems and extend upwards to a level just above the bottom of the collector electrodes and downwards to within about 450mm of the dust extraction device. The edges of the baffles must be sealed to the hopper walls by welding or some other positive means.

The pressure drop across a precipitator largely depends on the design of the gas distribution internals and also the inlet and outlet ducting arrangement. Typically the pressure drop is 20 to 30 mmwg but extremes can be as high as 150 mmwg.

4.6       Inleaking Air

The efficiency of an electrostatic precipitator is inversely proportional to the gas flow rate therefore all sources of air inleaks should be reduced to a minimum. Inleaks to the precipitator casing are particularly damaging to performance, since they can distort the gas flow through the precipitator and lower the flash-over voltage, causing localised sparking. If they occur in the area of the dust hopper, especially at the dust extraction points, dust which has already been precipitated will be picked up and carried out of the precipitator, substantially increasing the emission level. In addition, localised cooling due to air leakage can cause rapid and severe corrosion of the casing. Precipitator cladding should be inspected during kiln shut downs. Damaged cladding could result in loss of the insulation material. Localised cooling would then occur resulting in casing corrosion and inleaking air.



The rating of the rectifier set must be adequate to meet the demand of the precipitator. If either the current or voltage rating is too low, the performance of the precipitator will be limited. As an approximate guide, the voltage rating, based on the mean value into a precipitator load, should be equivalent to 4 kilovolts per centimetre of the discharge electrode to collector electrode spacing. The current rating, on the same basis, should be equivalent to 0.5 milliamps per square metre of collector electrode area for a plate type precipitator and 1 milliamp per square metre for a tube type precipitator. In addition, the set must have the facility for the output to be varied from a minimum to a maximum value, both by manual adjustment and by an automatic control device.

With the electrode system correctly aligned and the rectifier set in proper working order, the system should be verified by carrying out a still air load test. This is done by energising the electrode system under still, ambient air conditions. Under these con- ditions the maximum rated current output of the rectifier set should be reached without the onset of sparking. This test provides a useful reference if the output of the rectifier set is increased in steps, with the voltage and current being noted at each step. The readings thus obtained may then be used to draw the electrical characteristic. Figure 11 shows the electrical characteristic of a still air load test on one field of a clinker cooler precipitator.

Should sparking occur during the still air load test, the flash-over point should be located and the electrode system closely examined to identify the cause. Remedial action can then be taken and the still air test repeated until satisfactory results are obtained.

When the precipitator is on stream, a gas load test may be carried out in a similar manner to the still air test and this may be used to check the operation of the auto-control system. An example of this is shown in Figure 12. The working point when under auto-control is indicated and it can be seen that, immediately beyond this point sparking occurs, indicating satisfactory operation of the auto-control device.

A direct relationship exists between the maximum electrode voltage and the onset of sparking and for maximum efficiency, the electrode voltage should be maintained at the level necessary to achieve a controlled rate of sparking, say about 50 sparks per minute. However, under certain gas conditions the electrical characteristic may be extremely flat and the power required to achieve this state would then be excessive. The current rating of the rectifier set then limits the power input, but since the voltage is virtually constant under these conditions (see FIgure 13) the efficiency of the precipitator Is unaffected. In such a case, a continuous automatic dust monitor could be used to control the precipitator current in order to prevent power waste.



The efficiency of separation of a precipitator is a parameter which is of considerable interest both to the designer and the user, since from a knowledge of the separation efficiency for similar plants the designer can determine the size of precipitator necessary to meet a required duty, and the user can decide whether his particular precipitator is attaining the best performance which can be expected, or to see whether the performance has deteriorated for some reason, such as a change in the properties of the dust or in the operating conditions.

The efficiency of an electrostatic precipitator is usually defined as:

Efficiency 二 Weight of dust collected by the precipitator /               Weight of dust entering the precipitator


The question of the efficiency of a precipitator may be approached from a theoretical or an experimental direction. The theoretical approach consists of the development of equations in which the efficiency is expressed in terms of the properties of the dust, the field strength, the velocity of the gas stream, properties of the gas and the dimensions of the precipitator. The experimental approach consists of measurement of the quantity of dust entering and leaving the precipitator and relating these measurements to the different variables associated with the condition.

Ideally, the efficiencies obtained by the two methods should agree within limits, but in practice the divergence between the results is considerable and the conclusion must be that the present state of the theory of the subject is unsatisfactory. However, it should be noted that the experimental determination of the efficiency includes effects arising from the re-entrainment by the gas stream of particles which have reached the collector surfaces; of the transport into the gas stream of aggregates which have become detached from the deposited dust layer and of experimental errors. The efficiency deduced theoretically however, is based upon the rate of, deposition of the dust on the collector surfaces and does not include such Incalculable effects.

Nevertheless, a link must exist between the theoretical and experimental determination of precipitator efficiency and this becomes evident on examination of the fundamental efficiency equation theoretically deduced by Deutsch in the early 1920’s, and universally known as the Deutsch Equation.

According to Deutsch, the efficiency of a precipitator is given by:

Efficiency of electrostatic precipitator equation
Efficiency of electrostatic precipitator equation

It can be seen that the Deutsch equation expresses the efficiency in terms of three basic quantities, namely, the rate of passage of the gas through the precipitator, the dimensions of the collecting surfaces and the effective migration velocity. The first two quantities may be arrived at without difficulty and it is apparent that their contribution, if any, to the difference that exists between the theoretical and experimental determination of the efficiency of a precipitator is insignificant and thus this difference must result from differences between the theoretical and experimental determination of the effective migration velocity.

The effective migration velocity (W), or drift velocity as it is sometimes called, is the velocity component of the dust particles perpendicular to the collector plates and is given by:

The effective migration velocity (W), or drift velocity
The effective migration velocity (W), or drift velocity

It is a direct measure of the rate of precipitation of the particles since, for example, if a precipitator is de-energised, W and hence the efficiency, due to electrostatic effects, will reduce to zero. On the other hand, if W is doubled, the precipitator size may be reduced to half its initial value. Thus W may be interpreted as a generalised performance parameter for the precipitation process.

While the equation clearly indicates that W is proportional to the product of the electric fields and the conductivity and size of the particles, and inversely proportional to the

gas viscosity, it is apparent that other factors such as re-entrainment due to rapping and other disturbing factors, which would be included in an experimental determination, are not taken into account. Thus effective migration velocities computed theoretically are invariably greater than those obtained by substitution of measured’ values in the Deutsch equation, typically by a factor of about 2. It follows, therefore, that the true efficiency of a precipitator is in all cases lower than would be expected from theoretical reasoning alone.

The basis of most precipitator designs is the substitution in the Deutsch equation of an effective migration velocity which has been found by experience to apply to the particular electrode arrangement to be used and to the nature of the dust to be collected.

Thus the steps would be:

  1. Carry out efficiency tests on existing precipitators, on plants similar to the plant
    for which the new precipitator is to be designed, arid calculate the effective migration velocity by substitution of the measured results in the Deutsch equation, which may be rewritten in the form:

2- Calculate the collecting surface area that the new precipitator will have to
contain in order to achieve the required collection efficiency at the anticipated gas flow rate, by substituting the value of W derived experimentally in the equation:

3- Select the size and number of collectors required to give this area and arrange them in a casing in successive groups such that the cross-sectional area presented to the gas flow will produce the desired gas velocity, typically about 1 metre per second.

This method is said to be not entirely satisfactory, but in fact, owing to the uncertain state of precipitator theory, it is the most convenient available at present. Each precipitator manufacturer has its own methOds and variants of the above. They, of course, possess enormous sets of records of designs and performances to use In their predictions.

When a precipitator has been designed and erected, the only factors over which the user has any control are the gas flow rate and thus the effective migration velocity. Therefore, for an existing precipitator:

Efficiency ”            W/V or W/Q

Typically, the effective migration velocity will increase with an increase in gas velocity to a maximum value after which the force exerted on the dust particles by the gas stream will dominate and the effective migration velocity will decrease and approach zero (see Figure 14).

When W and V are both increasing, there is a balancing effect which tends to prop up the efficiency curve, but as the rate of increase inW falls off, and particularly when an increase in V results in a reduction in W, a pronounced drop in efficiency occurs.

Where a precipitator is required to operate at a higher gas flow rate than that for which It was designed, the W/V relationship is of prime importance and must be carefully monitored to ensure that the working point does not lie in, or beyond, the turnover region of the W/V curve, where small changes in V will result in disproportionately large changes in efficiency.

Ideally, the W/V working point should fall on the near-level portion of the curve preceding the turnover point, where normal fluctuations in the gas flow rate will not significantly affect the efficiency.

The manufacturers of electrostatic precipitators are not in complete agreement on the question of the values of W to be used in the design of precipitators for the various cement making processes and thus, in order to obtain comparable tenders in response to an enquiry, and to ensure that the precipitator is adequately sized to enable it to cope with the inevitable process variations etc, it is necessary for the value of W to be included in the specification. At the present time, the following values are considered to be appropriate, for precipitators with plate to plate spacing of 300mm.

Wet Process Kilns                     0.09 m/s

Semi-Dry Process Kilns             0.085m/s

Dry Process Kilns                      0.08 m/s under direct operation

Kiln Bypasses                           0.06 m/s


These migration velocities all apply to collector plate spacings of 300 mm. Larger or smaller spacings change the migration velocity pro-rata, e.g. wet process at 400mm, w== 0.12 rn/s. It should be noted that these values of W are anticipatory and if, when operating a precipitator at the design gas velocity and inlet dust concentration, the specified value of W is not attained the efficiency will be reduced accordingly, and theoretically, the precipitator would need to be increased in size to make up the deficiency.

It is therefore important that realistic gas flow rates and effective migration velocities are used when specifying new precipitators and that the design specifications of existing precipitators are referred to and compared with actual performance data so that steps may be taken, where necessary, to extend or replace them before the gas cleaning requirements exceed their capabilities.



Dust precipitation depends on a number of factors but generally better precipitation can be achieved if the resistivity of the particles lies within the range 106 to 1011 ohm.cm. As previously mentioned resistivity varies with temperature and gas humidity as shown in Figure 5. For preheater or precalciner kilns, gas normally leaves the preheater at
about 350”C and 5% moisture content. With the raw mill operating part or all of the gas is used for drying of the raw materials. It therefore may be necessary for part of the kiln gases to be conditioned during raw mill operation, and all the kiln gases during a raw mill stop.

At Hope Works gas conditioning is carried out by spraying water under high pressure (2000 kN/m2 into the riser duct between Stage II and Stage I, spraying downwards at

a position just above where the raw meal enters. This reduces the thermal efficiency of Stage I and thus increases the overall kiln fuel consumption. Water is also injected under high pressure into the downcomer fed by the two outlet ducts from stage I cyclones. Water injection, at this point, is currently limited by the capacity of the particular dust handling equipment down stream. It is planned in the future to add all the conditioning water at this point and stop the addition between stage I and II cyclones.

The above is a reasonable compromise, but for completely satisfactory treatment, to achieve a gas temperature of 1500C into the precipitator, a gas conditioning tower is necessary. A typical conditioning tower construction is shown in Figure 15. There are several possible combinations of gas inlet and spray nozzle positions that may be employed. Towers that operate concurrently with gas entry and spray nozzles located at the top, have been proved to operate the best in practice.

Spill-back nozzles are commonly used in which the flow of water is controlled by regulating a valve controlling the water flow returning from the nozzle. This method ensures water pressure at the sprays is not reduced prior to the sprays, which would cause poor atomization at low flow rates. Typically a water pressure of 33 bar at the sprays is necessary for efficient operation. The pressure at the pump delivery needs to


be 33 bar plus the pressure drop in the pipeline to the sprays. For preheater and precalciner operations it is best practice to have two separate feed systems to the spray nozzles for direct and indirect kiln operation (raw mill off and raw mill running). The optimum water flow rate is determined for these two conditions and the appropriate number of spray nozzles is selected for each condition. This ensures that efficient operation of the spray nozzles is maintained for both kiln operating conditions.

Inleaking air across the conditioning tower circuit should be minimal. Oxygen content in gas leaving the tower should be less than 1% higher than the oxygen content in the gas at the preheater fan exit. If a higher inleak is measured the source should be identified and the ducts and tower should be inspected for signs of corrosion.

With any conditioning tower there is the risk of spray malfunction leading to sludge formation. Safe access should be provided above the hopper base for cleaning blockages. Lorry access should also be provided under the tower to facilitate cleaning after a blockage. Drag chains are preferred to screw conveyors for dust transportation due to their ability to handle surges. The extractor may be reversible, allowing the dumping of dust which contains excessive moisture, before it can block the dust transport system.

7.1Conditioning  Tower Sizing

The following notes are presented as a guide to the basic requirements regarding the dimensions of a tower and assume that the nozzles will be of the spill-back type. These are:-
About 0.6 gramme of water is required per normal cubic metre of gas in order
to reduce its temperature by 1‘C. For an outlet gas temperature of 150”C the minimum active tower volume
required can be calculated on the basis that the water will evaporate at the rate of 18.25 kg per hour per cubic metre when the inlet gas temperature is 350”C. The active height/diarneter ratio should not be less than 2.5:1 or greater than 4.5:1. Typically h/d = 3.0 to 3.3:1.
The residence time should be approximately 10 to 13 seconds relative to the average gas flow rate, i.e.:-

Water sprays should be located approximately 1 metre below the top of the cylindrical section of the tower. This must be taken into account when sizing the effective volume of the GCT

The water pump capacity should be over sized by 33%. The maximum water capacity should be determined for maximum kiln production at expected inlet temperature and for direct kiln operation.

When calculating the effective volume of the GCT only the cylindrical sectionshould be calculated and not the inlet and outlet areas of the tower. In particular, the volume used should be that between the level of the spray system and the centre line of the tower outlet port. The evaporation rates indicated above apply for conventional spray nozzles. The rate of evaporation can be significantly increased by the use of sonic spray nozzles. These nozzles use compressed air to create a standing sonic shock wave which shatters the water into a fine mist. Typical water droplet size from sonic sprays is 100 microns which compares to 150 microns for spillback nozzles. Sonic nozzles can achieve up to 30kg per hour per cubic metre of tower volume, with exit temperatures between 140 to 1450C. There is, however, a significant additional power cost involved in the use of sonic sprays due to the use of compressed air for atomization which is typically in the region of 3 kWh/t of clinker. Usually the height:diameter ratio of a GCT with sonic nozzles may be up to around 6:1.



With a continuous automatic dust monitor, the performance of a precipitator can be checked continuously on a measured basis, providing an immediate indication of the effect of any change that has been made. It is therefore both a useful process instrument and an invaluable aid to performance optimisation. If a precipitator is not equipped with a monitor, its performance on a day-to-day basis and the effect of any changes that are made either to the process or to the mechanical and electrical condition of the precipitator can only be judged in subjective terms.

A dust monitor may be used, in conjunction with a microprocessor system or a PLC,, to control the operation of an electrostatic precipitator.

The equipment exists for a dust monitor to record, into a computer system, all events when dust emissions exceed the statutory limit, for how long and by how much. In certain states (Lander) of Germany, this information is down loaded by the Inspector and used to calculate penalties.

The use of approved continuous automatic dust monitors on kilns and coolers is mandatory in most countries of the EU and most states of the USA.


Wet and semi-dry process kilns, because of the low temperatures and high moisture content inthe exit gases, are well suited to electrostatic precipitation. The exception would be very small wet process kilns with high back-end temperatures. Cladding and thermal insulation must be well maintained on these precipitators to reduce the rate of corrosion of the precipitator casing. To combat corrosion, precipitator casings were constructed from reinforced concrete lined with acid resistant tiles. Examples of these precipitators can be found on many semi-dry kiln plants.

For most other cement works applications some form of gas conditioning is required to lower the temperature of the gases to be cleaned and to increase their water content.

Dry process heat exchanger kilns make use of the gases in the milling and drying circuit which automatically produces conditions which are conducive to electrostatic precipitation. When the kiln is operating as an independent unit however, water must be evaporated in the gases, preferably by means of a purpose designed gas conditioning tower, to both cool and humidify the gases. The requirement is that sufficient water be evaporated in the gases to achieve a final gas temperature of 150oC and that the evaporation is complete.

Precipitators are commonly used on kiln bypass streams. Typical bypass gas temperatures are 1050oC and this is reduced to about 350oC by bleeding in ambient air. The gas is then usually cooled further to 150OC in a conditioning tower, although some hot gas precipitators do exist. The particle size of bypass dust tends to be very small making it difficult to collect. The dusts also tend to be sticky in nature due to their high alkali or chloride content, making dust handling very difficult.

Gas conditioning is usually required for long dry process kilns, although some older plants operate with hot gas precipitators. These are operated at gas temperatures of around of 4000C, with the temperature being controlled by the addition of ambient air and sometimes a quantity of water sprayed into the back-end of the kiln.

Electrostatic precipitators have been used for many years to de-dust waste gas from clinker coolers. Their operation tends to be inconsistent due to the large variations in gas conditions brought about by kiln flushes, etc. The gas is generally at a high temperature, has a low moisture content and the dust has a high resistivity. Water sprays are often used to condition the gas to improve precipitator operation. Most modern cooler installations use fabric filters down stream of a heat exchanger to de– dust the cooler exhaust.

For open circuit cement milling, and older generation closed circuit milling, water Is usually injected into the mill for cooling and this is usually sufficient to condition the dust for effective precipitation. For effective operation the water dewpoint temperature of the air at the inlet to the precipitator must be in excess of 450C. Modern closed circuit cement mills have separate mill ventilation and fresh feed Is partially cooled by the coarse returns from the separator. This reduces the need for cooling with water injection and results in higher resistivity dust in the gas streams. The trend for recent cement mill installations is that they are dedusted by bag filters.

Precipitators have been employed on indirect fired coal milling circuits however this is not common and fabric filters are the preferred de-dusting equipment.


For reliable and optimum precipitator performance it is necessary to have a preventative maintenance programme. This programme must incorporate short term reliability to reduce the likelihood of breakdowns and unnecessary plant down time. It must also look at long term aspects which could result in gradual loss of collecting efficiency and lifetime of equipment with special emphasis on corrosion and wear.



Improved precipitator performance has been achieved On many plants by installing microprocessor based automatic controllers. These controllers offer features such as;

Back corona detection, to set optimum currents for operating conditions with

high resistivity dust.

Rapping with reduced current level and/or increased current level in a down stream field.

Quick power ramping after a flashover.

Energy management controlled by a signal from the dust monitor.

Pulse energisation has been used for many years to improve precipitator operation on high resistivity dusts. These devices enable current to be applied in pulses to the field systems and these can be controlled in amplitude and frequency. The equipment is expensive and unlikely to be applicable to cement works dusts.
An increase in precipitator size, costing the same, is likely to be much more successful.
A Variovolt system has been marketed over the past few years to improve existing precipitator performance. This is a transformer-rectifi~ set which applies secondary current at a frequency between 1000 to 2000 Hz to the field. The increased current, frequency results in a near straight line wave form which enables the applied kV to be
very close to the flashover kV. This enables a higher kV to be applied to the field and therefore an improvement in

collection efficiency.



To obtain the best performance from an electrostatic precipitator the two most important requirements are first that the electrode system be correctly aligned and have the ability to accept a voltage of the magnitude necessary for the precipitation process to take place efficiently and secondly that each rectifier set be adequately rated to produce that voltage. Questions of gas distribution, gas conditioning, effectiveness of rapping, etc, can all be deferred until these two basic requirements are met.

When it has’ been established that the electrode systems and the rectifier sets are without fault, the question of precipitation proper can be considered.

For those applications where the gases to be treated require conditioning by the introduction of water (hot and/or dry gases), the water spray system must be effectively maintained and operated so that the condition of the gases at the inlet to the precipitator complies with the design specification with regard to temperature and humidity.

All possible measures should be taken to reduce the amount of gas that a precipitator has to handle, and in particular, air inleak must be reduced to a minimum. Air inleak can also cause dust pick-up and gas flow distortion.

The frequency of operation of the rapping gear must be optimised to avoid re- entrainment or excessive accumulation of dust on the electrodes, which could adversely affect precipitator efficiency.

Dust collected in the hoppers should be continuously extracted to prevent any possibility of a build-up contacting the electrode system.

To provide immediate indication of the effect of any changes made to the process or the precipitator and to aid performance optimisation a continuous dust monitor is necessary. In many countries it is now a legal requirement to monitor compliance with environmental legislation.




Previous Post
Next Post

Leave a Reply