· Power consumption & product fineness
· Primary air heating requirements
· Mill fires
· Mill wear
Power Consumption and Product Fineness
Both of these performance aspects relate to the grindability of the coal, which is related to the amount of energy required to reduce the coal particle size. Low rank coals are not normally hard in the sense that they are relatively easy to deform, however much of the deformation is plastic rather than elastic, that is they are not brittle. Therefore the amount of deformation required to break the particles is higher than for many higher rank coals, therefore requiring a greater expenditure of energy.
The Hardgrove Grindability Index is essentially a measure of this energy requirement, though, as a small-scale simulation of the milling process, it has some shortcomings as a test. Figure 11 showed that there was a clear trend for HGI to reduce (more difficult to mill) as coal Carbon content dropped from 90 down to about 80. With a further reduction in Carbon content, there was less of a trend, with HGI values lying between approximately 30 (for an Australian coal) and 60. The HGI test is known to lack precision for low rank, high moisture, coals because the test result is highly sensitive to the moisture content for a
particular determination7.
Thus, the HGI has limited reliability in predicting the fineness and power consumption when pulverising a low rank coal at a given tonnage per hour. A second important factor not covered by the HGI is that low CV coals need to be milled at a higher tonnage rate for a given boiler duty, thus compounding the problem.
When a new coal is introduced to a power station, there are two possible courses of action for the operators regarding the running of the mills:
· They may retain all the previous mill settings and hope that the performance is
satisfactory,
· They may adjust the mill settings (including roll pressure, classifier setting) to suit the new coal.
ACIRL adopts both these approaches when testing coals in the pilot-scale vertical spindle mill:
· The mill is run at standard settings; power consumption and fineness are reported,
· The mill settings are adjusted to give a standard fineness of PF of 70% passing 75 μm.
Standard Mill Settings: Figures 19 and 208 show mill power consumption and fineness
respectively for standard mill settings. Figure 19 shows the typical trend for the power consumption to increase with lower HGI. However, it shows that most Indonesian coals perform better than expected, ie, the mill power consumption is lower than the trend for their HGI range. On the other hand, Figure 20 shows that the Indonesian coals produced coarser PF than the trend. This represents an enforced trade-off between power consumption and fineness when coals are milled in this way.
7 HGI of low rank coals may also depend on the history of wetting and drying before the test.
8 Mill performance data and coal quality data in this section came from projects
commissioned by coal companies and industry funding.
Figure 19: Mill Power Consumption (kW.h/t) at Standard Mill Settings for Indonesianand Australian Coals
Figure 20: PF Fineness at Standard Mill Settings for Indonesian and Australian Coals
Milling to a Standard Fineness: Figure 21 shows the trend of power consumption versus
HGI when the mill is adjusted for standard fineness. Under these conditions, Indonesian coals perform close to the overall trend, though the majority are still slightly better than the trend.
Figure 21: Mill Power Consumption (kW.h/t) to Produce Standard PF for Indonesian
and Australian Coals
The significance of these results is as follows:
· If an Australian coal is replaced by an Indonesian coal without mill adjustments, thepower consumption will probably be similar to that for Australian coal with HGI
about 10 points higher, so it is not likely to be excessive,
· The PF of the Indonesian coal is likely to be coarser than that of the Australian coal. This may not become obvious because not all power station operators sample and
analyse the PF,
· Coarser PF can lead to a deterioration in burnout efficiency. However, as will be
covered later, Indonesian coals inherently produce very favourable burnout efficiency, so this may not be a problem,
· If it is found necessary to adjust the mills to produce standard PF, then the mill power consumption is likely to be only slightly better than that of an Australian coal of the same HGI.
Calorific Value: The above comments are based on the same coal feed rate for all coals. Taking into account the need for a greater tonnage for lower CV coals, Figure 22 allows for this by plotting mill power consumption (kW) per unit coal energy input (GJ/h). The data spread is much wider than that of Figure 19, indicating that lower CV coals may be subject to limitations based on mill power consumption.
Figure 22: Mill Power Consumption per Unit Coal Energy at Standard Mill Settings for
Indonesian and Australian Coals
Primary Air Heating Requirements
Coal is dried in the mills as well as pulverised, requiring higher primary air temperatures for higher moisture coals. Based on the as-fired moisture content of the coal entering the mills, and assuming the PF exiting the mills contains a fraction (approximately half) of the air-dried moisture, the quantity of moisture removed can be estimated. The problem is compounded for low CV coals (which are often the high moisture coals) because a greater tonnage throughput of coal is required. Consequently a ranking of the drying requirements may be calculated as:
Drying Requirement = (Moisture Removed)/CVNAR (kg/GJ) (1)
Figure 23 (9) shows the Drying Requirement versus Carbon (%daf) for Arutmin, KPC, other Indonesian and Australian export coals. The Drying Requirement rises steeply as Carbon content decreases below about 78%.
Figure 23: Moisture Removed per Unit Coal Energy during Milling for Indonesian and
Australian Coals
In a given power plant there will be a top limit on the acceptable value of Drying
Requirement; the limit will depend on the mill and burner design. The specific limitations may arise from:
· The air-heater capacity may be insufficient to attain the required primary air
temperature,
· The operators may place a top-limit on the allowable mill inlet air temperature to
avoid the possibility of mill fires.
It is clear that, because of high as-fired moisture content, some Indonesian coals will not suit some plants that are not designed for them.
9) Based on data supplied by the Bumi Group (KPC and Arutmin coals) and the Barlow Jonker database (other coals).
Mill Fires
As indicated under the previous heading, high moisture coals may require higher mill inlet temperatures which increase the hazard of mill fires. Unfortunately, as well as having high moisture contents, low rank coals tend to be more prone to spontaneous combustion because of the reactivity of their organic matter. The two characteristics compound the mill-fire problem. For very low rank coals, special plant design features may include:
· Higher than normal primary air volume to reduce the temperature requirement
· The use of attrition mills and recycling some flue gas into the primary air10.
Mill Wear
It is a misconception that low HGI coals are inherently abrasive. Abrasive wear of mill components is normally caused by hard mineral matter in the coal, typically free silica (quartz) or iron pyrite. The Abrasion Index (or Yancey Geer Price Index) is a standard laboratory test to provide an indication of abrasiveness.
ACIRL’s pilot-scale vertical spindle mill has the capability of measuring mill wear, which tends to correlate, though not perfectly, with Abrasion Index. Indonesian coals typically have lower Abrasion Index than Australian coals and this is reflected in lower mill wear rates in the vertical spindle mill (Figure 24).
Figure 24: Mill Wear Rate for Indonesian and Australian Coals10) These measures are used for Australian brown coals (typical Carbon content 68% daf), which are not exported.