Have you read this article?
Solving the Problem of Build-up on Preheater ID Fans
Greater demands for throughput and efficient use of heat in the kiln have placed greater demands on kiln induced-draft fans. These fans have been designed with ever-increasing volume and static pressure requirements, as well as higher process gas requirements. The result has been larger fan rotors operating at very high tip speeds.
One side effect has been build-up on the impeller. Typically, the build-up is an extremely hard, brick-like substance, which can break off during operation, causing serious imbalance. Build-up on preheater ID fans seems to be widespread, with documented cases in Pakistan, India, Saudi Arabia, Ecuador, Mexico, Texas, and many other areas in the United States.
A study at Robinson Industries, Zelienople, Pa., examined each of the generally postulated causes of build-up in kiln ID fans. The object was to determine what variables could be adjusted to control build-up. Findings from the study were applied in the design, building, testing, and installation of a build-up-resistant preheater fan at the Holnam Texas L.P. plant in Midlothian, Texas.
Holnam Texas case study Build-up forming on the blades of the preheater ID fan at the Holnam Texas plant was requiring periodic shutdowns of about 24 hours - 16 hours for the fan to cool down and another eight to sandblast the blades and remove the build-up.
The original fan at Holnam was obtained from Robinson in the 1980s. Running at its original design capacity, the fan did well, but the problems developed as Holnam increased production. By 1994, the fan was running 30% over its intended capacity, and Holnam was having difficulty tolerating shutdowns, which were increasing in frequency.
Hoping for a solution to the build-up problem, Holnam turned to Robinson Industries, which was able to design a fan that resists build-up and therefore makes frequent cleaning unnecessary. Robinson determined that build-up can be reduced by matching the shape of fan blades, as closely as possible, to particle streamlines, so the impact energy of dust particles is minimized. Blade angle must be inclined enough to prevent the "hard" buildup on the blade's front surface and yet radial enough to prevent the "soft" buildup on the blade's back surface. Engineers determined that the best blade for the Holnam cement plant would be the backward-curved fan.
Robinson also determined that the new fan at Holnam would have to be larger than the old one, not only to accommodate the plant's increased production but also to accommodate the blade's new design. To receive the same performance for a given application, a backward-curved fan needs to be larger than a radial blade fan.
Holnam's parent company, Holder bank, also conducted research on the build-up problem, which corroborated Robinson's.
Robinson faced three challenges typical in the design of kiln ID fans. The fan would have to: 1) handle a steady stream of dirty air; 2) tolerate high temperatures; and 3) tolerate fast changes in temperature, from 450 ºF to 840 ºF. Robinson faced an additional fourth challenge specific to the Holnam plant, namely that the fan would have to handle extreme stress because of its size.
The original ID fan at the Holnam facility was rated at 3,500 hp and designed for a speed of 1,180 rpm. The new motor, while rated for 4,500 hp, would be designed for 880 rpm.
The size of the fan and its weight (67 tons including the wheel, shaft, housing, inlet dampers, bearings, and bearing pedestals) presented fabrication challenges. The rotor was welded on a positioner so the tilt would be perfect for welding at every point. In addition, the fan's size necessitated splice welds in the shrouds and web plates. All shrouds and web plates were subjected to X-ray testing.
The fan's plate steel was ultrasonically tested for defects. Before the welds, destructive tests were conducted on sample plates. After construction, a dry magnetic particle inspection was made of all final wheel welds. All bolts were ultrasonically tested for defects. The bolts were tested again during tightening to ensure that they were being stretched to the proper length.
Since the fan's installation in 1997, Holnam has not had to stop production because of build-up or any other fan-related problem.
Research behind the solution prior to the building of the Holnam backward-curved preheater fan, Robinson had conducted an extensive study into the problem of build-up in cement factory fans. The study entailed two approaches: 1) information gathering and 2) a laboratory simulation of cement plant build-up conditions. In the first case, information from several cement plants was collected to determine if a pattern of conditions leading to build-up could be identified.
- Theories as to build-up causes can be broken down into the following four primary groups:
1. Thermal: As process gas temperatures have increased over the years (from 350 ºF only a few years ago to 700 ºF and higher today), so too have build-up problems. Some particles carried by the gas stream have a lower melting point than others and may become "sticky" above 500 ºF.
2. Chemical: There are several chemical theories. Here are two of the main ones.
a. Some believe the presence of sulphur encourages the formation of gypsum, which is a very hard and difficult-to-remove material, formed from calcium carbonate in a temperature range from 700 ºF to 1,800 ºF.
b. Another chemical theory suggests that chlorine reacts with other components in the gas stream to form lower-melting-point salts such as NaCl, FeCl subscript 3, and KCl.
3. Electrostatic: Since the fan rotor is not grounded (due to the oil film of sleeve-type bearings), positively charged dust particles may be attracted to a negatively charged fan rotor (or vice-versa).
4. Mechanical: Build-up could result from fan blade geometry. Some believe that centrifugal force holds material in place on the back side of airfoil blades, backward-curved, and backward-inclined centrifugal fan rotor blades. Others believe that build-up results when dust particles impact the front side of the fan blades, melting or embedding themselves into the surface of the steel blades.
Laboratory test work a 32 inch diameter rotor was set-up in a closed-loop system at the Robinson Industries laboratory. The test set-up allowed variations in dust loading, fan speed, temperature, particle velocities, fan blade shape (rotor design), coatings, and types of steel surfaces. By controlling most process parameters closely, it was possible to change one variable at a time to determine its effect on the rate of build-up on the rotor.
The first priority was to see if the field type of build-up could actually be duplicated in the laboratory. After 168 hours of continuous operation, the rotor was stopped and the fan casing opened for inspection. A significant build-up (to 11/42 in. thick in some areas) was observed.
The build-up occurred on the leading edge of the blades (pressure side) and at the blade-to-center plate weld intersection. The build-up was heaviest at areas of high impact between the dust particles and the rotor surfaces, and along streamlines containing high concentrations of dust particles. In other words, the location as well as the appearance of the build-up was similar to that in field sites.
- Each of the possible variables was examined through experiments or field observations to determine how best to reduce or eliminate build-up:
1. Temperature: The operating temperature was reduced in 50 ºF increments. It was noted that the build-up was significantly reduced (but not totally eliminated) at temperatures below 500 ºF.
2. Blade shape: Tests were run using radial blade rotors, backward-curved blades, and airfoil-bladed designs. Results showed that build-up could be reduced with airfoil and backward-curved blades.
3. Fan speed: As the fan speed was reduced to 1,013 rpm, build-up was reduced. Lower fan speeds (resulting in lower impact velocities as dust strikes rotor surfaces) appear to reduce build-up.
4. Surface roughness: The fan rotor had four blades fitted with ground finish 316 SS blade liners. While stainless steel provides corrosion protection and an ultra-smooth surface, build-up was virtually unaffected.
5. Coatings: Several coatings were tried including two ceramic coatings, tungsten carbide, alonized, polished 316L SS and 410 SS and nickel boron. None reduced the amount of build-up on the fan wheel.
6. Sandblasting: With build-up already on the fan rotor, coarse sand was added to the recirculation loop, but to no effect. Either the build-up was too hard or the sand was not striking the fan rotor surface in the same areas as the smaller particle feed dust. Texas Lehigh reports that this method of cleaning was only marginally successful in the field, with build-up being evened out (high spots were removed) but not eliminated. Excessive use of sandblasting during operation can lead to rotor erosion and failure.
7. Sonic horns: Reports from various plants indicate that sonic horns are not an effective means of preventing build-up on preheater ID fans in cement factories.
8. Electrostatic grounding: A grounding brush was attached to the rotor shaft of a preheater ID fan at Box-Crow Cement (now Holnam Texas L.P.). The voltage to ground was measured at 0.00 V. This compares to 31 mullivolts static charge when an ungrounded rotor was in operation with feed dust recirculation at 700 ºF. There was no noticeable difference in the amount of build-up between these grounded and ungrounded setups.
- Field observations Reports from 20 plant sites have added some interesting insights on the causes of and possible solutions to preheater ID fan build-up:
è Build-up has occurred on all types of fan blades.
è Build-up has been reduced in some measure by lessening the impact velocity of dust particles striking the fan rotor. A larger-diameter rotor operating at a reduced speed will acquire less build-up. Reducing the operating temperature (possible by air dilution upstream from the fan or water spray in the down corner from the preheater section) greatly reduces build-up on kiln ID fans. While temperatures below 500 ºF (260 ºC) are necessary to achieve the reduced build-up, temperatures at or below 392 ºF (200 ºC) increase particle size, causing fan rotor erosion.
è In some cases, a rotor can be sized with an oversized shaft to make it less sensitive to imbalance caused by uneven build-up. This arrangement allows for longer operating cycles between shutdowns for cleaning.
è One or two plants have tried water sprays directly onto the 700 ºF rotating fan wheel. Rapid cooling (quenching) was reportedly effective in removing build-up, but thermal fatigue on the fan rotor was devastating. In at least one case, a rotor failed catastrophically.
Conclusions
è Build-up on preheater ID fans is definitely temperature and impact velocity related, which would support theories about lower-melting-point alkaline forming a sticky compound.
è Sulphur may be ruled out as a cause of build-up on preheater fans since no sulphur was found in the fuel in the lab test, though build-up still occurred.
è Formation of FeCl subscript 3 resulting from a chlorine reaction with the rotor's steel may also be ruled out. Build-up occurred even with stainless steel and other protective (non-iron) coatings on the rotor.
èThe recommended temperature of gas entering the kiln ID fan is 482 ºF (250 ºC). This temperature may be achieved with water sprays in the down comer from the preheater tower. The presence of moisture in itself may minimize build-up.
è The fan rotor should be designed for the smoothest possible flow lines. The backward-curved and airfoil rotor designs are best for this purpose.
è Fans should be selected to achieve minimum gas/dust particle velocity at the inlet of the fan rotor. Minimum velocity means minimum impact on the fan rotor. Consider using:
a) Double inlet fans instead of single inlet fans.
b) Larger-diameter and lower rpm fans.
c) A peripheral speed at the rotor inlet opening limited to 15,000 fpm and an inlet velocity not to exceed 7,500 fpm.
è Oversized shafting should be used to reduce sensitivity to imbalance. The design critical speed (including effects of bearing oil-film stiffness) should be at least 1.25 times the operating speed of the fan including the weight of 1-in.-thick (2.54 cm) build-up on all leading surfaces (pressure side) of the fan rotor blades. Careful attention is required to insure rigidity of the foundation up to the fan-bearing pedestals.
