Top stage cyclone collection efficiency determines the quantity (Mt/hr) dust finally escapes from pre-heater section.
With a recommended cyclone geometry, particles feed size as kiln feed and dust from up stream gases has an important role in particles separation from moving gases.
Generally, top stage collection is carried out in single or twin cyclone(s) with an average efficiency of 92 — 94 %. That means, 6 to 8 % of cyclone feed dust is moved out (= 6 to 8 % * 1.66 = 10 to 13 % of clinker) which is again collected and reintroduced in the system.
The chemical compositions of escaped dust generally have higher L.S.F.and A.M. and lower S.M. value than original mix and this is an indication of major share of lighter and smaller particles are from limestone and argillaceous materials like clay. The hardness of such materials as mentioned is comparatively less than other additive(s) used as siliceous stone and/or iron ore, and some time bauxite if argillaceous material is not used.
Therefore with higher L.S.F. or more calcite particles presence, the dust sample intend to have more lime and proportionally higher loss on ignition while being tested.
Here, let us illustrate the relation between CaO and LOI of escaped dust samples and test results are collected from a running plant — which also went under successive increase in kiln feed rate during this period.
A small drop in individual sample LSF is found but the noticeable differences is observed in sample LOI as reported. Lime (% CaO) as tested in all samples, represents the fact as dust sample contains more calcite particles and simultaneously high CO2 loss was expected.
A graphical representation of LOI as found and as expected against individual sample LSF value is demonstrated as -
A gradual decline in LOI as tested against expected LOI is visible and same is again related with steady increase in kiln feed rate. As it looks, increase in feed rate is bringing down the % LOI in samples. This cant be the reason unless up-stream gases contains calcined particles coming from lower cyclones. The calcined material would be almost LOI free and its mixing with kiln feed at Stage-I cyclone can reduce % LOI as tested.
There is always a chance of a small fraction of calcined dust from PC and bottom cyclone gets mix with, as no cyclones are 100% collection efficient. And in normal way of operation when solid — gas mass ratio remains within an acceptable limit, presence of 2 to 3 % calcined dust is very common.
An approximation of containing calcined dust is calculated from the difference as earlier shown between actual and expected LOI. The base of calculation as considering 92 to 94% calcination which represents sample LOI between 3 to 4 %.
And calcined dust share in go away dust from Stage 1 is computed for an equivalent LOI to achieve desired calcination as mentioned above.
An additional 3 to 4 % of calcined dust is found present when kiln vol utilization factor remains above 5.5 tpd/m3 —keeping preheater dimension unchanged.
Considering a feed rate of 100 tph, 3 to 4 mt/hr increase in fresh feed could have been possible if this extra calcined dust being avoided. Which is equivalent to 50 to 70 tpd increase in an overall clinker production. Other than the advantage of lesser quantity to handle and again same to be de-dusted.
The system solids-gas mass ratio is defined as the mass of solids over the mass of gas inside the preheater cyclone and accordingly cross sectional area of each stage cyclone and riser ducts are calculated. The size of cyclones relates to the maintaining of desired gas velocity criteria and efficiency. The velocity is to be sufficient for the powder to be picked up and lifted and at the same time gas flow velocities in the immersion tube to be minimum to attain optimum separation efficiency.
Quantity of combustion and decomposition gases increases proportionately with kiln feed and the increase reflects in gas velocities within the preheater system.
The effect of cyclone inlet velocity on collection efficiency is found best at 16 m/sec and acceptable between 14 to 18 m/sec. Therefore, the velocity <14 and >18 m/sec would be detrimental to the separation performance - allowing more particles to escape. Similarly higher gas velocity at cyclone entrance may have a tendency to rise pressure drop - compelling induced fan to work more. This is to be added here, higher particles concentration in gases helps to improve cyclone performance but at the same time it may not have any effect on pressure drop.
Here we understand, an existing pre-heater system can adopt maximum utilization rate within an acceptable limit as marked in dotted line- shown in above charts. Beyond it, the system may needs up-gradation to over come possible bottle necks -
Fans capacity.
Replacing or modifying existing top stage cyclones.
Modifying existing cyclone geometry, mainly inlet portion and inlet spiral.
Modifying dip tubes in existing cyclones.
Modifying riser ducts.
Replacing dispersion/splash boxes.
Modifying D.C.D. and T.A.D.
Modifying kiln inlet chamber.
Equally, lower feed rate have an adverse affect on cyclone efficiency as it may not fulfill the required velocity for efficient solid - gas separation.
We had a long discussion on gas velocity and how it reflects on collection efficiency of a cyclone. And possible affect is unwanted dust particles being escaped when gas velocity exceeds the limit as mentioned.
Here we see, escaped dust analysis not only give us the information regarding particles character but it also highlights the performance status of entire pre-heater section.
Periodically or when ever there is a change in kiln feed rate, P.H. dust loss analysis and computing the same as mentioned before (Cal-2 ) - may provide an useful information about ongoing pyro-processing.
The world’s most famous management consultant Peter Drucker once said — “you can’t manage what you can’t measure”.