Mechanical Maintenance of Cement Rotary Kiln
Mechanical Maintenance of Cement Rotary Kiln
Mechanical Maintenance of Cement Rotary Kiln
IN
MECHANICAL ENGINEERING
Submitted By
i
MALLA REDDY ENGINEERING COLLEGE
Maisammaguda, Dhulapally, Secunderabad. 500 100
(Affiliated to JNTUH - Hyderabad)
Ph: 040-64634234. Fax: 040-23792153
------------------------------------------------------------------------------------------------------------
Department of Mechanical Engineering
CERTIFICATE
This is to certify that the project work entitled STUDY ON MECHANICAL
MAINTENANCE OF CEMENT ROTARY KILN has been submitted in partial
fulfillment of the requirement for the award of degree of Bachelor of Technology in
Mechanical Engineering discipline of JNTUH, Hyderabad for the academic year
2013-14 is a record bonafide work carried out by
External Examiner
ii
ACKNOWLEDGEMENT
I feel ourselves honored and privileged to place our warm salutation to our
college MALLA REDDY ENGINEERING COLLEGE and the DEPARTMENT
OF MECHANICAL ENGINEERING which gave us the opportunity to have
expertise in engineering and profound technical knowledge.
I sincerely thank our internal guide Mr. Raveendra, professor for his
valuable and constant encouragement throughout the project period.
I wish the gratitude to our principal Dr. S. Sudhakara Reddy, principal for
providing us with the environment and mean to enrich our skills and motivating us in
our endeavor and helping us realize our full potential.
R.VIJAYAKUMAR
iii
CONTENTS
NO TITLE PAGE
Abstract 1
List of figures/ tables/ screens 2
1. Introduction 3
2. Shell inspection 5
5. Refractory linings 65
5.1. Introduction 65
5.2. Stresses on refractory linings 66
5.3. Mechanical stress 67
5.4. Slope 69
5.5. Thermal stress 70
5.6. Scheduled maintenance 71
6. Gears 72
6.1. Flange mounted gears 73
6.2. Spring mounted gears 74
6.3. Regular visual inspection 75
6.4. Daily/monthly 77
6.5. Vibration 80
6.6. Annually 81
6.7. Axial run out 82
6.8. Radial run out 83
7. Seals 86
7.1. Kiln inlet seal 86
7.2. Kiln outlet seal 87
7.3. Seal designs 87
7.4. Common causes for seal problems 88
7.5. Inverted seal 93
8. Conclusion 95
9. References 96
ABSTRACT
1
List of Figures/ tables/ screens
FIGURE PAGE
TITLE
NO NO
1 Cracks extending from filler bar welds 9
2 Mechanical loads 10
3 Circumferential cracks 11
4 Ripped shell 15
6 Shell distortion 17
7 Migrating tires 24
10 bolted design 27
13 Boss hole 78
15 Gear nomenclature 80
2
1. INTRODUCTION
The rotary kiln consists of a tube made from steel plate, and lined
with firebrick. The tube slopes slightly (1–4°) and slowly rotates on its axis at
between 30 and 250 revolutions per hour. Raw mix is fed in at the upper end, and the
rotation of the kiln causes it gradually to move downhill to the other end of the kiln.
At the other end fuel, in the form of gas, oil, or pulverized solid fuel, is blown in
through the "burner pipe", producing a large concentric flame in the lower part of the
kiln tube. As material moves under the flame, it reaches its peak temperature, before
dropping out of the kiln tube into the cooler. Air is drawn first through the cooler and
then through the kiln for combustion of the fuel. In the cooler the air is heated by the
cooling clinker, so that it may be 400 to 800 °C before it enters the kiln, thus causing
intense and rapid combustion of the fuel.
3
Mechanical maintenance is very important for maintaining the kiln perfectly to
achieve good productivity. This mechanical maintenance includes various topics i.e.,
heat transfer, thermal expansions, distortions etc.
Identifying and troubleshooting the problems are discussed with further details
inside the chapters.
4
2.1 REGULAR OBSERVATION REVEALS PROBLEMS
A. CRACKS
B. HEAT DAMAGE
C. DISTORTION
6
figure1: Cracks extending from filler bar welds
The unit walks-by starts with the shell, the unit’s largest component and then
covers all its support and drive components. Each chapter of this manual completes a
section of the walk-by. Your own walk-by should be conducted daily, with the main
focus on changes in condition from one day to the next.
This chapter focuses on shell inspection and the three main areas of shell
stress; cracks, permanent shell damage from thermal distortion, and temporary
shell problems caused by unit misalignment.
7
Figure2: Mechanical loads
2.3 CRACKS
The severest sign of shell stress is cracks in the steel plate. The shell stresses
that lead to cracking are caused by the following factors:
Thermal expansion
During the normal operation of any kiln shell the temperature fluctuates,
causing the shell to expand and contract. The tire support pads expand and contract at
a different rate than the shell, creating stresses in the welds used to attach the bars to
the shell. These stresses are created from both longitudinal and circumferential forces.
Friction
Tires are mounted loosely on the shell to allow for the different rates of
thermal expansion of the tire and the shell. As a result, the tire will have
circumferential movement relative to the shell. This is referred to as creep or slip.
There is generally a sliding component from this action that creates stress in the welds
attaching the tire pads to the shell.
8
Tire Thrust
As the unit is moving axially by
carrying roller adjustment or thrust roller
positioning, a force is transferred from
the tire to the shell. This force is applied
on the retaining rings or tire stops and
consequently causes stress in the welds
that attach the tire stops and the support
pads to the shell.
Ovality
Excessive “flexing” of the shell as it turns causes stress in the shell plate. The
welds that attach the tire support pads create a “stress riser” that can lead to fatigue
cracks in the shell plate.
9
than longitudinal cracks as they
may travel sufficiently around a
shell in a short period of time
causing catastrophic failure.
Circumferential cracks most
often occur on or near a
circumferential weld seam or at a
shell opening such as a manhole or
a patch plate. They are the result of
poor quality welding.
10
The deflection and slope curves are important because it is used by the kiln
designer to locate the support piers such that the slope of the kiln shell is parallel to its
design slope. That is the slope value should be zero at the center of each support (the
red dashed lines). Note also how the shell sags between piers. When a kiln is
modified by a change in length, change in loading or change in the number of support
piers, the disregard of shell sag can lead to aggressive tire to side stop block wear.
The shell bending stress curve is used to determine the shell plate thickness.
Note that the saw tooth tips along this curve are where the shell plate changes
thickness. The stress on the kiln shell is greatest at these points and so these
circumferential welds should be the focus of your attention when making your daily
inspections.
The shear stress in a typical kiln shell is very low and is rarely if ever a cause
for failure.
11
Dye penetrant testing is probably the simplest NDT procedure to perform and
can usually be performed by site technicians. Its use is limited to surface cracks and
is a valuable tool in the actual weld repair procedure.
Cracks are a sign of total failure and should never be ignored.
A. Shut the unit down and position it where the emergency repairs can be most
easily performed.
Weld “strongbacks” perpendicularly across the face of the weld tear. A
“strongback” is a piece of A-36 iron approximately 1½” thick by 6” - 8” high, and
long enough to bridge the weld tear by at least 12” to 14” on either side.
B. Bring the unit back up long enough to completely run out the product that was
in the unit at the time of the breakdown.
C. Remove approximately five feet of brick on either side of the weld tear.
Refractory may already be missing due to the sudden misalignment caused by
the weld tear.
D. Position the unit where the “strongbacks” are removed and install adjustment
hardware on the inside of the unit to reunite the separated parts of the kiln
shell.
E. When the seam is welded together on the outside, remove the adjustment
hardware on the inside. Back gouge and weld the seam. Grind the seam as
flush as possible.
13
F. It is necessary to install new “strongbacks” around the circumference of the
failed seam. Otherwise the weld failure will move to the part of the shell that
does not have the benefit of the additional repair hardware.
This method is a Band-Aid! Because shell stress and misalignment are present,
the unit will not perform properly. Catastrophic failure will occur elsewhere on the
unit if proper repairs are not made. Installing a new shell section at the area of weld
failure, coupled with complete alignment analysis, is the most likely plan of action.
As a result of the unit being suddenly stopped and allowed to cool in one
position, the kiln shell will sag between piers. This in itself is not the cause for alarm.
The sag will diminish as the unit is brought up to production temperature. Generally
not all of the sag will disappear but the unit will true it well enough to operate.
14
software programs to show you a real-time temperature as well as historical data on
temperatures in various zones of the kiln. A scanner can work well as a preventive
tool by providing information on potential problem areas, and allowing steps to be
taken before refractory fails and shell damage occurs.
When one area shrinks it may cause another area to bulge. The shell, generally
considered to be a cylinder, no longer has a straight axis. It is now bent, or kinked, or
15
as it is often described in extreme cases, it has a “dog leg” or a “crank shaft”
condition.
Depending upon the location of the bend, the “dog leg” or “crank shaft” can
also cause tires to wobble. Wobbling tires not only exert cyclical loads into the
rollers and shaft but also reduce the contact area between the tire and the rollers. This
increases the contact pressure. If excessive, this can cause surface spalling of the tire
and the roller faces. The tire wobble also makes proper roller skewing difficult, if not
impossible.
16
In the area of the girth gear, this will affect both the radial and axial run out of
the gear, causing accelerated wear of the gear teeth. Depending on the location of
the “dog leg”, the seal may be adversely affected as well.
Correcting the Problem: Depending upon the nature of the distortion, it may
be possible to reduce the bend in the shell by using localized heating and cooling
procedures. Results are not exact or reliable. In most cases a replacement shell
section is required to affect a reliable repair.
2.9 MISALIGNMENT
When the support rollers are not set correctly to proportionally share the
rotating load, the kiln is misaligned. This condition causes serious overloading stress
conditions, similar to those caused by a bend in the shell. In contrast to cyclical load
imbalance caused by a bend in the shell, misalignment creates constant load
imbalance because the rollers are not holding the shell straight. The magnitude of the
overloading is in direct relationship to the magnitude of the roller displacement. The
effect of misalignment can be the same as a permanently deformed shell; cracks,
premature refractory failure, hot bearings, etc.
Obvious signs of misalignment are excessive flexing, refractory failures, plate
cracking and roller shaft failures. One of the most obvious signs of misalignment is a
rise in amp usage. If amp usage is suddenly up, something is wrong.
17
Misalignment cannot be detected visually nor can it be determined on a trial-
and-error basis as can roller skew. In order to determine misalignment requires
careful alignment measurement and specialized tools and procedures. These
procedures are covered in the section “Alignment Analysis”.
Alignment measurement is performed by an outside service organization since
very few companies have a large enough number of units operating to warrant the
expense of acquiring specialized equipment and retaining qualified staff to perform
this task.
2.9.1 WHEN SHELL IS MISALIGNED
Or has distortions causing a bend in the shell...
A. Excessive Flexing
B. Refractory Failures
C. Plate Cracking
D. Roller Shaft Failures
E. High Power Drive
2.9.2 ALIGNMENT MEASUREMENT SHOULD BE PERFORMED
Unlike cracks or heat damage, kiln misalignment cannot be detected visually.
Careful alignment measurements and analysis of the results are the only way to
determine if roller adjustments can reduce the stress on the kiln shell. Such
adjustments once made where a significant misalignment existed often results in a
18
dramatic drop in kiln motor amperage verifying that the energy used to turn the kiln
has dropped.
Understanding the mechanical function of all the parts and being able to
identify what areas need attention, will allow you to make the decisions about what
can and cannot be handled satisfactorily in-house.
Identifying the symptoms early and utilizing the proper analytical tools can
allow time to formulate and implement a plan of action. Failure to do so in some cases
can lead to catastrophic failure or unplanned outages.
19
It is impossible to restrain the effects of thermal expansion on a steel shell.
The shell components, the tires and the filler bars, also undergo thermal expansion.
Because the expansion rate of the components is usually unequal to that of the shell,
the problem of managing these differences becomes a primary focus for maintenance
personnel. The way the tire and filler bars are mounted on the shell plays a large part
in solving the problem of unequal thermal expansion.
Any time that inspection of a rotary unit is carried out, the tires and tire
elements should be carefully examined. Once a small problem develops with these
components, larger problems are sure to follow.
20
3.1 TIRE MOUNTING-2CLASSES
FIXED
FOR UNITS WITH SHELL TEMPERATURES LESS THAN 200° F/100° C
In the broadest sense tires can be put into two classes; fixed tires and loose,
migrating or floating tires. Fixed tires are usually found on unfired vessels and
equipment whose shell temperature is below 200°F or 100°C. Loose tires are
normally required to accommodate differential thermal expansion between the tire
and the shell, especially on those shells whose surface temperature is high.
FIXED -There are many ways a tire may be fixed on the shell. Some are
welded directly to the shell. Others may be wedged, pinned, keyed, splined, blocked
or otherwise mechanically fixed. Fixed tires that come loose must be identified and
repaired. If tires require frequent repair, an evaluation of the method of mounting
should be made.
LOOSE - Loose tires are inherently more difficult to maintain because they
“migrate”, meaning that by design they are free to rotate on the shell. This migration,
even when sufficient and controlled, and even when appropriate lubrication is present,
will wear the mating parts (filler bars, stop blocks, rollers, etc.). A migrating tire also
becomes a problem when it becomes too loose to properly support the shell.
21
3.2 PROBLEMS OF
LOOSE (MIGRATING)
TIRES
Wear occurs on OD of
filler bars.
22
3.3 FILLER BAR
DESIGNS
FULL FLOATING
SEMI-TRAPPED
WELDED
RIM-MOUNTED
ON SHELL (BODY
FIT)
SPLINED OR
TANGENTIAL
SUSPENSION
Refractory-lined
shells requiring floating tires
demand the most service
performance from the tire
mounting components. Over
the years the “full floating
filler bar” design has evolved
and provides the best
Figure8: full floating filler bars
compromise to meet the
conflicting service demands of the migrating tire mount.
23
The full floating design incorporates retaining rings which bear against the
side of the tire, providing a much larger bearing surface than former designs using
only stop blocks. Significantly reduced wear results. The retaining rings are keyed to
the shell so that the wear surface is the side against the tire and not the side which
contacts only the keepers.
24
3.5 SEMI-TRAPPED DESIGN
This design involves fewer parts than a
free floating design so it is a little less expensive
to install. It allows the filler bar to expand and
contract, however it still has the potential for the
weld to crack.
Bolts are used to attach the bars to the shell. Look for loose bolts and bolts that
are missing or have sheared off.
25
When making an inspection of the filler bars of any type, be particularly aware
of any cracks, loose welds, or rolled edges. Often the smallest filler bar problem can
lead to the largest shell problems.
The floating filler bar design eliminates weld cracks caused by thermal
expansion and contraction.
3.8 TERMINOLOGY
A. Welded Filler Bar: Also called a tire pad. Welded to the shell.
B. Floating Filler Bar: Not welded to the shell but held in place by keepers and
stop blocks.
C. Side Keeper: Welded to the shell.
D. Stop Block: Welded to the top of the filler bar. Provides a bearing
surface for the tire.
E. Shim: Welded to the bottom of filler bars to accommodate an
out of round shell and to get accurate gap between the filler bar and riding
ring. Size is adjusted to allow for welding. Example: FB = 1” x 12” x 28”,
Shim = 1/8”, 1/16” & 1/32” x 11.5” x 27.5”.
26
F. Shim Keeper: Welded to the shell. Keeps the shim from slipping out
from under the filler bar if it is not welded.
G. Retaining Ring: Located between the stop blocks and the edge of the
riding ring. Provides a larger bearing surface than a stop block.
The clearance between the tire splines and the shell pads is taken up by the
slow wedges. As this clearance varies with tire to shell temperature difference the
wedges migrate in and out. This results in a tight assembly circumferentially at all
27
temperatures and yet leaves room for the shell to expand/contract radially within the
tire.
Advantages:
A. Minimum and stable ovality of kiln shell, irrespective of tire clearance.
B. No thermal stresses, since there is no impeded thermal expansion.
C. Because the tire shoes have high elasticity there is uniform support of each
fixing unit, even when the kiln shell temperature varies.
D. Separate reception of bearing load and the axial load.
E. Provides torque transmission from the tire to the kiln cylinder.
The riding rings provide substantial strength to the shell by maintaining shell
roundness. Because the shell naturally flattens out at the 12 o’clock position like a
balloon full of water, the riding ring system must maintain shell shape by preventing
flex. A support system must also provide an accurate and ridged method of mounting
the tire to the shell.
28
By design there is a difference in the size of the shells outside diameter (OD) and
the tire’s inside diameter (ID), the tire having the greater diameter. Because of this
difference the tire naturally wants to creep, or migrate at a slower rotational speed
than the shell. The shell is actually rotating at one speed and the tire is lagging behind
at a slightly slower speed.
By making a mark with soapstone from the side face of the tire to the surface of a
filler bar, or along a stop block, it is possible to witness the mark slowly separate from
the two surfaces during each rotation. This separation is a direct measurement of the
fit between the shell OD and the riding ring inside diameter.
When we go to the top of the kiln and measure the actual gap we find that it is
larger than the difference in circumferences (Tire bore circumference less shell filler
bars circumference) divided by . The reason for this is ovality, meaning the shell
sags across the top. Another way to think of this is that the shell and tire are not
perfect eccentric circles. If they were then the gap would be equal to the creep/ .
29
Even this assumes that the creep is the result of true rolling action with no slip or
hang-up.
The amplitude “s” of the resulting plot is the actual gap. The period “ U” is
the prevailing creep. U / s ≠ but something more likes 2 to 2.5. This ovality
ratio varies from kiln to kiln and tire to tire.
This may seem like a very academic issue but it has great significance when it
comes to calculating the expected filler bar thickness when reducing the gap to correct
ovality is necessary.
Although this is an excellent diagnostic tool its use is often limited by the
presence of thrust rollers and high speed kilns.
30
E. There is a direct relationship between the number and types of supports under
a riding ring, and their ability to distribute the unit load to the shell. Worn
filler bars, or supports, allow excess gap at the shell’s 12 o’clock position, thus
allowing excessive flexing or flattening. This reduces the shell support
provided by the riding ring, accelerates and compounds support pad wear, and
leads to shell cracks. Cracks are a sign of complete failure and will eventually
adversely affect the total operation of the rotary unit.
Excessive wear also allows the shell to move through riding rings, which mismatches
the rings to the rollers and the gear and pinion. This then increases the loads at the
contact area of the rings and rollers, gear and pinions, and increases hertz pressure
which leads to surface spalling.
When there is wear on the tire it will become more and more difficult to make
adjustments to the tire, and there will be gear and pinion wear, or even failure.
31
Figure11: several types of damage on riding rings
There is always some metal distortion present in the pinch point between the
roller and tire. On a microscopic level this can be seen as an area flattening out,
similar to that of a rubber tire on a car where it contacts the road. When a tire starts to
wobble or if for some reason the pressure in a localized area of the pinch point
becomes excessive, the distortion of the steel will go beyond the elastic limit. The
flexing of the work-hardened layer against the underlying softer mass will create
cracks between them. Once subsurface cracking starts, it propagates until chunks of
metal break free.
Ultimately, spalling, and the reasons for it, can cause the tire to crack in half.
32
Worn or broken filler bars, and the resulting loose tire, will produce high
ovality readings. Ovality measurement is a standard practice for measuring how much
a shell flexes. This technique was developed specifically for keeping an eye on how
much flexing the shell exerts on the refractory lining. It is discussed in detail in a
separate section called “Ovality Analysis”.
Also of concern is the extent of wear at the sides of the tire where it comes in
contact with its retaining bars or stop blocks. With high loading, these blocks or bars
will also be worn. Visual observation of the space between the tire and these parts
gives a good idea of what is going on. For a new assembly this space is only about
1/16”. Average space after years of operation will show about ¼” to ½” space. Any
more than that and the tire may run off the face of their rollers and the gear may
misalign on the pinion.
Heavy side loading of the tire against these retainers will also cause the
retainers to break off. Sometimes these may be welded on and the welds will break. If
they are bolted on, the bolts may shear. No matter how they are mounted, cracks and
heavy wear are easy to spot and are a sure indication that something is wrong.
Typically, mis-skewed rollers or a bent shell are the principal reasons for this
kind of damage. There are other reasons for this to happen as well. Fixing cracks and
33
worn parts, or replacing missing pieces may not resolve the problem. It is important to
carefully analyze the condition, since redesigning the way the tire is retained may be
required.
If the problems you have observed are severe, you may have to replace the tire or
roller. A way to solve a problem that has been caught early is to grind the surfaces of
the tire smooth. The roller should be ground at the same time. (See “Resurfacing and
Grinding” section). A third solution is to replace the worn filler bars and stop blocks.
Doing this will not only replace worn out components but reshimming the filler bars
will reduce creep.
If the shell has collapsed under the tire and filler bars, a section may have to be
replaced. This alone is a very good reason to be aware at all time of the potential
problems inherent in the design of rotary equipment.
34
3.13 THE QUESTION OF LUBRICATION
The question of lubrication between the bore of a migrating tire and the shell
or shell bars that the tire fits on has always been controversial. Although lubricating
the tire bore may seem to be a natural requirement of good kiln operation, some
experts advise against it. One argument is that “greasy” lubricants may attract dust
and debris which then act as grinding compounds and accelerate wear. The second
argument says that lubrication will promote slippage and creep, again, hastening wear.
1. Any dust and debris will be consumed by the grinding action, leaving nothing to
gall the steel.
True rolling action can never be assured and is at best, a transient condition.
Lubrication, therefore, is not applied to induce slippage, but to prevent and local areas
from galling and hanging up to the point where metal failure occurs. A greater amount
of creep may be seen with the use of lubricant than without, which, if it acts to polish
the surfaces, is infinitely more desirable than creep who is limited. Inhibited creep
will eventually tear the metal.
35
Galling occurs when dry steel slides by dry steel and the surfaces attach
themselves on a microscopic level and destroy themselves. The steel balls up and
forms slugs or spitzers. These are created at the sides of the tire where they contact
the retaining blocks. Lubricating here is essential to prevent undercutting the tire and
consuming the stop blocks. Lubricating the entire area is completely appropriate.
36
Roller adjustment can be classified in two categories.
A. Alignment adjustment (Discussed in detail in Alignment Procedures).
B. Skew
Skew, more than any other mechanical adjustment, is the least understood, the most
misused, and causes more mechanical problems on average than all the other
adjustments that can be made on a kiln combined.
4.1 DEFINITION
Skew is the adjustment made to a roller by pivoting the roller on the midpoint
of the roller shaft. This means making equal adjustments in opposite directions on
the upper and lower bearing housings of a trunnion.
This pivoting adjustment only changes the parallel relationship to the longitudinal
axis of the rotating shell, but does not affect (to any significant degree) any change in
the position of the shell either in plan or elevation views. In other words the roller is
pivoted but the shell is not raised or moved laterally.
This simple, but important concept must be understood completely before correct
roller adjustments can be made. Thrust control by skewing may be the single most
important adjustment which influences the optimum mechanical operation of the unit.
37
A. WHAT IS SKEW?
B. WHY SKEW ROLLERS?
C. WHY IS PROPER THRUST IMPORTANT?
D. WHAT IF MY KILN HAS HYDRAULIC THRUST ROLLERS?
E. WHAT IF MY KILN IS A “FULL THRUST” KILN?
F. WHY SKEW ROLLERS?
4.2 PRINCIPLE
The principal mechanisms for confining the axial movement of the shell are
the thrust rollers. They prevent the shell from moving downhill, which is a unit’s
natural tendency due to gravity and weight. Most thrust rollers are not designed to
take the full load of the unit for even a very short period of time, unless the unit is
designed with hydraulic thrust rollers or is a full thrust unit. (These units are
designed to accept the full thrust load on the lower thrust roller.) Skewing the rollers
properly will counteract the gravitational force of the kiln, and move the kiln
uphill. Improper thrust can add to the gravitational force, overloading the thrust roller
and causing failure.
Proper thrust achieves a somewhat delicate balance of skew on each trunnion
so that the unit will lightly bump the lower thrust roller during operation. Too little
skew will cause the unit to ride harder downhill than necessary, and wear on the lower
thrust roller and the thrust face of the tire. Too much skew will induce an unnecessary
amount of wear on the trunnion and tire faces, causing conical wear patterns to
develop rapidly.
Skewing the rollers for proper thrust ensures that all rollers are thrusting in the
same direction, and in like amounts. Uneven skew, i.e. one roller thrusting the kiln
downhill and the other roller on the same pier thrusting the kiln uphill, or if the rollers
on one pier are thrusting the same direction, and an adjacent pier is thrusting in the
opposite direction, will accelerate the wear on the tire and trunnion faces, as well as
cause other problems; increased power consumption, higher bearing loading,
increased contact stress on the tires and trunnions, etc.
Good preventive maintenance begins with a complete understanding of SKEW.
38
To best understand the concept, we take a cylinder and a board. These
represent the fundamental geometric shapes whose interactions illustrate what
happens when skewing adjustments are made.
We place the board on the cylinder in a fashion similar to how a kiln or dryer
tire (the board) sits on a roller (the cylinder).
39
Looking from overhead (the plan view) the board is placed on the cylinder at
right angles to the cylinder’s rolling axis. When the board is pushed so that it rolls on
the cylinder it does not move either to the left or to the right.
This is referred to as the neutral position or zero skew.
40
If the cylinder is pivoted at its midpoint, its rolling axis is no longer at right
angles to the board. Setting the board in motion (straight ahead) causes the board not
only to move ahead but also to move sideways. The direction of the lateral motion of
the board depends on the direction in which the board moves, forward or backward,
41
Curving the board so that it takes the same shape as a shell tire, and supporting it on
two cylinders instead of one is the last step to make the basic elements like a true tire
supported by two rollers. Curving the board does not alter the action – reaction of
skew but it does introduce other aspects which can be significant wear factors. First
there is a change of pier to pier alignment. But since skew is so small this effect is not
significant. Then there is an increase in local stress concentration on the rolling
surfaces as well as the introduction of skidding or sliding. These last two can cause
severe wear.
42
When the cylinder is skewed with the flat board sitting on it, the board does not
change position in elevation or in plan view. When the board is curved as shown, and
the rollers are skewed equally, two things happen.
1) The board changes its elevation slightly
Assuming both rollers are skewed equally (only to simplify the calculation) the
change in elevation is:
E A 2 - (B - skew)2 A 2 B2
A Radius of tire + radius of roller
B = sin(Angle) A
43
cooler or dryer etc. Although the skew is significant at 0.25mm, the change in
alignment elevation, DE is not.
2) The line of contact between the cylinders and the board changes. The line of
contact is not really a line. It is an area defined by
I) the length of contact between the roller and the tire in the axial direction.
II) and the width of contact which varies according to:
A) Roller diameter
B) Tire diameter
C) Hardness of the material
D) Roller slope matching the tire slope
E) Amount of skew
It is most desirable to have the area of contact as rectangular as possible. e.g... view
(a). When skewing is required, which is the case for many units by design, then
clearly the minimum amount of skew to just balance the down thrust of the shell,
should be sought. The skewing should be shared equally by all the rollers. For
illustration purposes diagram (b) shows excessive skewing, so much so that only half
the roller face is in contact. Since the load this roller carries has not changed, the
44
stresses in this reduced area must necessarily be higher. Visually the stress volume of
the yellow shape at “a” must equal that of “b”.
We can see therefore that excessive skewing decreases the contact area and
increases the unit load, and stress, in that area. The contact area behaves similarly to
a car tire in contact with the road. The contact area actually flattens out and the
material in the flat area deforms. When this deformation exceeds the elastic limits of
the material, it fails.
Skewing causes edge loading as seen in “b”. This can be catastrophic if the skewing
is excessive. The symptoms would include mushrooming, edge cracks in the rim and
ultimately large pieces coming out of the loaded edge of the roller. Since some
skewing is required in most cases, changing the roller slope by shimming is
beneficial. Only bearing housings that have self aligning bearing sleeves or spherical
roller bearings are easily adjusted in this way. Bearing housings with fixed sleeve
bearings can be shimmed using tapered shims but this is a more complex procedure.
When the roller slope is adjusted for skew the load carried by the roller is
distributed as shown in “c”. The peak stress is moved back to the center of the roller;
45
the stress reduces towards the edges and is symmetrically distributed. This is a much
better distribution pattern and makes the effort to do this worthwhile.
On the upturning side of the kiln shell the downhill bearing is shimmed and on the
down turning side the uphill bearing is shimmed. The shim thickness is about 0.6
times the amount of skew.
Now we put it all together. The curved board is the tire and the cylinders are
the rollers. The effects of skewing the cylinders have exactly the same effect on the
curved board as it does with the flat board. Skewing causes the ring to be rolled into
the direction the contact surface of the cylinder is moving. Since the rings are
prevented from moving laterally this causes the rollers to shift in the opposite
direction. As long as the roller is free to shift it continues to do so until the roller
shaft reaches and seats on the thrust bearing.
When neither the ring nor the roller can shift, the thrust load is relieved by
slippage. Therefore, with skewed rollers we no longer have pure rolling action.
Slippage is another effect that causes problems. It can tear the rolling surfaces. At
the same time sufficient thrust bearing capacity has to be provided by the support
roller bearing assembly. It will almost always be possible to have an overly skewed
46
support roller generate more thrust than the thrust bearing can handle. The oil film in
the bearings becomes too thin, metal to metal contact occurs, the surfaces heat up
which in turn reduces the oil viscosity further, and the bearing fails. Once the thrust
bearing fails the heat generated is usually enough to fail the support bearing as well.
When support rollers are fitted with spherical roller bearings the situation is even
more critical since the thrust load and the support load both act on the one bearing
simultaneously. These will tend to fail more frequently than journal bearings with
thrust rings or thrust buttons.
Since a skewed roller no longer runs against the tire with a pure rolling action,
but induces some slippage, lubrication of the outside diameter with dry graphite is
highly desirable, and helps preserve the surfaces. Oil lubrication on the rolling
surface should be avoided as it can promote spalling.
Once again we have good cause to avoid skewing when possible and to limit it
to a minimum when it is required.
47
4.3 THE HAND RULE
Until you get used to visualizing the actual motions of the roller and tire, this
“hand rule” may help sort things out.
The palms: Stand and face the tire as it moves in front of you. If the tire surface is
moving up – hold your hands out, palms up. If the tire is moving down – hold out
your hands, palms down.
Fingers: Curl the fingers into your palm. They point into the direction the top of the
roller is moving. When palms are up the fingers curl up towards you which is the way
the top of the roller is moving. When palms are down they curl down and away from
you, again the way the top of the roller is moving.
Index finger: Points to the direction which the bearing is to be moved.
Thumb: Points to the direction the shell will move as a result. For (a) pushing the
left roller in will cause the shell to move to the right, and so on.
Remember, the thumb points into the direction the shell will move. The roller reacts
by shifting itself in the opposite direction of the shell.
ALWAYS USE AN ADJUSTMENT LOG BOOK.
48
4.4 DETERMINE THE BEARING STYLE
Type I.) Fixed Plain Sleeve Bearing with Thrust Buttons on the End Caps.
CHECKING AND DOCUMENTING THRUST
Checking the thrust on a housing that has the thrust buttons in the end caps is
pretty simple. Using a 3 or 4 LB. hammer, and lightly striking the end cap on or near
the center, will produce one of two different tones. One is a hollow “bong”, or empty
sound, which indicates that this end cap has no load on it. The other sound is a very
solid, high “ping” like striking an anvil, indicating that the roller is loading up against
this end cap. This style of roller is considered a “pusher”. When thrusted, the shaft
will load up against one end cap and push the kiln in the opposite direction. For
example, if the uphill end cap sounds hollow, and the downhill end cap sounds solid,
the roller is positioned downhill and is pushing the unit uphill.
Remember to sound both end caps, even though the first one you strike may
produce one of the distinct sounds mentioned above. If the roller is midway in the
bearing this will cause both ends to sound hollow.
49
DETERMINING THRUST DIRECTION BY ROLLER POSITION (Type I
Housing)
Both uphill and downhill bearing housings are keyed into the bases such that
the space between the thrust buttons is ¼ - ½” or 6 - 12 mm larger than the length of
the shaft. This allows the roller to have that much axial float. When the roller is
skewed to drive the shell slightly uphill, its reaction is to slide downhill. The normal
and expected position for all the rollers is to be in contact with the downhill thrust
button.
50
Type II.) Sleeve Bearing, Self-Aligning with Thrust Collars on the Shaft
The thrust collars are located on the ends of the shaft or on the shoulder of the shaft
near the roller. Visual inspection through the inspection ports of the housing allows us
to locate the gap. This is the gap between the thrust collars and the thrust bearing.
With the thrust arrangement as shown above, the normal expectation is to have the
gap on the downhill end of the shaft. This indicates the roller is positioned downhill
and is pushing the shell uphill.
51
DETERMINING THRUST DIRECTION BY ROLLER POSITION (Type II
Housing)
The normal and expected position of all the rollers, if slightly skewed to push
the shell uphill, is to position itself downhill. With Type II style bearings we then
expect to see no gap on the uphill side and a ¼ - ½” or 6 - 12 mm gap on the downhill
side. Tapping the end covers on this style of bearing housing does not tell us anything.
52
Determining thrust on Type II style housings is a matter of removing the
inspection port and examining the position of the roller. When the ports are removed
you will see (below) where one thrust washer is tight by noticing that oil has been
wiped clean from its surface. This can only be seen on the roller on the down turning
side of the shell. The other should show a gap in which the oil runs freely over the
thrust washer. This type of roller is considered a “puller”. This means that the shaft
will move until it seats against the thrust collar.
53
Type III.) Spherical Roller Bearings (No separate thrust bearings)
This is the most difficult type of bearing to deal with for setting skew. The
previous style of bearings is specifically designed to utilize the “action-reaction”
phenomenon of skew by allowing room for a small amount of axial shift. That ¼ - ½”
or 6 - 12 mm float is essential for setting skew correctly. With spherical roller
bearings there is no accommodating float to show us skew direction. Spherical roller
bearings are mostly installed on smaller faster-turning units. Faster turning means a
proportionately higher thrust for any given skew. Unfortunately these bearings have a
low tolerance for thrust load. Consequently we see a much higher failure rate with
spherical roller bearings as compared to Type I and Type II bearings.
54
Type III.) Spherical Roller Bearings (No separate thrust bearings)
By fixing a dial indicator as shown, thrust load may be detected if the unit can
be reversed. Often there is roll-back when a unit is stopped. Any thrust load will tend
to tip the bearing housing slightly. Upon roll-back the thrust reverses direction. There
will be a small amount of axial movement on the bearing housing. The greater the
thrust load, the greater the amount of movement. Usually the fixing ring is mounted
on the down-hill side bearing. This then should be the housing to which the dial
indicator is mounted. If it is mounted on the other bearing, the “free” bearing, then the
outer race may move within the housing and the movement may not be detected by
the indicator. If reversal is not an option, then slapping a broom handle wrapped with
a greased terry cloth across the face of the roller will also do the trick. As the strip of
grease goes through the pinch point, the thrust is relieved and the bearing housing
jumps. This technique is obviously limited to a one time use.
Loosening the hold down bolts may be another possibility to release some
axial movement on the housing. Safety is always a consideration to be heeded.
55
look very polished. Its surface will be mirror like, almost chrome plated in
appearance and be very reflective. A poorly adjusted roller with excessive skew will
appear dull and gray by comparison. In the extreme it will be very rough and have
tiny flakes of material coming off its surface.
Then the Wipe Test
Take a cotton rag, wipe across the face of the roller. First wipe from the
discharge end towards the feed end. Then wipe the opposite way. On the roller with
a dull surface there will be a distinct roughness in one direction, the fibers of the cloth
almost seem to catch on the grain. The other direction will seem much smoother and
the cloth will not hang up on the grain. On a shiny roller this difference will be
imperceptible. By wiping all the rollers it will be easy to judge which is the roughest
and which are the smoothest.
Any detectable roughness should be in the direction from discharge to feed
end. That means the roller is pushing the shell up-hill. Roughness feed-end to
discharge-end means the roller is reversed skew, pushing the kiln down-hill.
56
roller is in contact this causes the kiln to be pushed uphill for part of the rotation. As
the wobble then moves away from contact with the thrust roller, the kiln moves gently
downhill for that part of the rotation. The cycle then repeats.
What is “correctly skewed”?
Each roller is pushing the kiln uphill and with a minimum amount of thrust.
This minimum is defined by the condition above. The combined thrust of all the
rollers does not quite match the downward push of the kiln. In this way the kiln will
slowly and gently move down but then be nudged back up by the wobbling thrust tire.
57
the cooling jackets. Liberal application of water externally is good as long as it does
not get into the housing at seals or inspection ports.
Usually the best method of cooling is to use radiator oil cooler. Using a small
pump, a barrel of water, some lightweight, flexible tubing such as copper, and the
proper fittings, a radiator can be fashioned to continuously cool the bearing oil. The
suction side of the pump is connected to the oil drain on the trunnion. The oil is routed
through the pump and into the coils of tubing submerged in a barrel of water. New
cold water is constantly being run into the barrel to keep the exchange of heat as high
as possible. The oil is then dispensed onto the top of the trunnion shaft through the
inspection port. If desired, an oil filter can be connected into this system to filter out
some of the particulate. Caution must be used to keep the filter as free-flowing as
possible.
If a bearing is known to be problem synthetic oil should be used before any
moves are attempted. Synthetic oil retains its viscosity to 450°F [230 °C]. If a
petroleum oil is being used be prepared for the possibility of having to change oil “on
the fly” to a synthetic to sustain a high rise in temperature. Some synthetic oils are
not compatible with petroleum oils. The changeover must be total without cross
contamination. Continue flushing with synthetic until the change is complete.
If none of these methods bring the temperature under control, bearing failure
is imminent. Prepare to slow or stop the kiln. A slowed kiln may allow the problem
bearing to “seat-in”.
58
Moves can be properly measured using dial indicators, one for each bearing
assembly. Often the magnetic bases for the dial indicators are inadequate to hold the
indicator reliably over the course of an adjustment campaign. Weld brackets to the
base and use clamps to hold the indicators for 100% reliability.
Adjustments using the “flats” of the adjustment screw is good enough for “ball
park” adjustment but must never be relied on for recording the actual moves made.
The bearing housing may take some time to seat in. Leave indicators in place for as
long as 24 hours after the last adjustment, before recording the final bearing position.
From our previous inspection we have catalogued roller positions, surface
conditions, thrust direction and what problem bearings (if any) exist. From this we
can derive the most offending roller to the least, and sequence our adjustment
campaign accordingly.
Suppose we were required to do more than set the rollers to their correct and
minimum thrust. Suppose it was required to move them for alignment and skew as
well. Say our first roller needs a 15 mm (0.6)”) move towards the center line of the
unit in order to correct for alignment. This would then be an alignment adjustment.
We will use the roller reaction to guide our work.
The first move would be a small one of about 0.5mm (0.020”) in one bearing.
The bearing first moved would be the one which would bring the roller closer to
neutral. Wait about 20 minutes until the roller has had a chance to shift. This is also
enough time to catch any temperature rise in the bearing/oil sump. Record all
temperatures again. Trouble can be identified by a temperature rise anywhere, not just
in the bearing being moved. Assuming that the shaft journals and bearings are in
normal condition and no temperature rise was encountered, these steps would be
repeated as necessary until the roller shifts position. The roller’s shifting position
indicates that the neutral point has been crossed. This is the most critical aspect of
the whole procedure: to get the roller to shift position without any significant
temperature rise in any of the bearings.
Once it is seen that the roller shifts easily without a temperature rise, then the
size of the moves can be increased to say 2mm (0.80”) per bearing. The sequence of
the moves should alternate from one bearing to the other with the shaft sliding
across with each move. Waiting 20 minutes between moves is also unnecessary as
long as the shaft shifts easily with each move. The work can continue smartly
59
providing there are no other mitigating circumstances like a bowed shell, etc. This
continues until the average of the moves for both bearings reaches the desired total,
15 mm for this example. The final moves should be very small ones to leave the
minimum amount of skew on the roller.
Even the largest rollers, and there are some as large as 10 feet (3050 mm) in
diameter, will respond quickly to a 0.10mm (0.004”) skew adjustment. Naturally all
the work must be monitored with dial indicators and must be done with the unit in
operation.
This procedure is used with units that have sleeved bearings. See “Two-Pier
Alignment” for the procedure using spherical roller bearings and pillow blocks. The
principle of roller reaction is always valid even though thrust direction is not seen by
axial roller shift. Secondary techniques need to be used.
60
This is not desirable.
Toed-in rollers can balance each other, one pushing the tire down as hard as
the other is pushing it up. If situation exists the shell may be “floating”. Floating
means that the shell is balanced between thrust rollers. But it is not the desired
situation. Left uncorrected there is unnecessary wear and tear on the whole support.
Left for long periods of time the tires and rollers will wear into a cone shape.
How do you quickly identify if a roller is skewed?
Look at the surface. A roller with little or no skew will polish up to a mirror
finish. If the surface is dull and gray and appears rough, then skew is present and
probably excessive.
61
This is also not desirable
Rollers should be parallel to each other, set in the same direction on all piers.
Often measuring between bearings or shafts, as “a” above, may reveal that they are
parallel and not toed-in as in our previous example. Unless these measurements are
tied into a common reference line, the above situation will not be identified.
Once again this situation could be present with the shell “floating”. Assuming
from that observation alone that all is well, will lead to excessive wear and tear of all
the support components. Careful inspection using the simple techniques already
described will show this immediately.
62
Ideal Placement.
Unfortunately many designs require that support rollers be skewed. The thrust
mechanisms of these designs are inadequate to support the entire downward thrust of
the shell. This is especially true of large long rotary kilns. Since most of this type of
rotary trunnion-supported equipment is installed on a slope, there is a natural
component of force acting in the axial direction of the shell. If this force cannot be
completely managed by the thrust mechanism(s) it is the skewing of the support
rollers that must help out.
Skewing is a compromise. Skewing accelerates the wear and tear of the
support mechanisms but then allows smaller, less costly thrust mechanisms to operate
successfully. If skewing is insufficient the thrust mechanisms will fail prematurely.
If skewing is excessive additional wear and tear of the support components takes
place and the thrust mechanism can still fail. If rollers are skewed against each other,
wear and tear takes place but the advantage supposedly gained by skewing is lost.
The maximum performance life of rotary equipment that requires skewing, can only
be achieved by skewing correctly and keeping it to a minimum.
The amount of skew shown in the illustration may be sufficient for most installations
63
BENEFITS OF PROPER THRUST
A. REDUCED WEAR RATE.
B. REDUCED STRESS ON TIRE.
C. REDUCED ELECTRICITY CONSUMPTION.
OTHER CONDITIONS EFFECTING THRUST CONTROL
A. LOAD
B. OPERATING TEMPERATURE
C. AMBIENT CONDITIONS
D. SPEED
E. LUBRICATION
64
5.1 INTRODUCTION
A rotary kiln is lined with bricks designed to withstand the tremendous heat
that is generated inside the kiln shell, and the mechanical stresses that are present
during the countless rotations of the shell. Bricks can be made from a variety of
materials, and are chosen for their ability to protect the shell, for their durability, and
their ability to maintain the heat necessary for the process which goes on inside the
kiln.
The lining bricks are wedge-shaped or tapered, designed to fit the curvature of
the shell. If you look inside a kiln shell you will see thousands of bricks, laid up in
rows around the circumference of the shell.
There are several methods of installation, including the screw-jack method and
laying the bricks with mortar. The method chosen should suit the type and size of kiln
being used, and of course, should follow the recommendations of the refractory
specialist.
65
In this section we will look at what happens to brick linings over a period of
time and why linings fail. The bricks offer many clues that can help identify the
probable causes for refractory failure and may help you determine a plan of action.
This section will not discuss how to choose bricks. The choice as to what brick
grades are suitable for the individual zones in a rotary kiln depends on the process and
the location within the kiln. Each kiln will have unique requirements. This can best
be determined by the refractory specialist that visits your plant regularly and becomes
familiar with each kiln’s operating characteristics.
There is a lot to be learned by examining an old brick lining.
66
These requirements want contradictory physical properties in the refractory. Good
thermal characteristics come at the expense of mechanical resistance. Good
mechanical resistance makes poor insulators. Chemical resistance properties, like
glass for example, also make for either a poor insulator or are weak to resist bending
stress. The ideal brick is often thought of as one made of a chemical and heat
resistant rubber! An absurd idea but it does highlight the challenge of refractory
design. For this reason different refractory materials are formulated to fit different
circumstances. Selection of refractory is consequently extremely important to ensure
that materials are matched as best as possible to the requirements of the application.
67
SLOPE - Bricks are “pushed” from the feed end to the discharge end by gravity and
the pressure of the product being fed through the kiln. Bricks can be crushed against
retaining rings.
EXPANSION & CONTRACTION - Bricks expand up to 30% more than
steel. Usually the brick lining is fully expanded and compressed against the shell in
the operating condition, but during a rainstorm or kiln upset, the equilibrium between
the shell and lining is upset and the bricks may come under extreme pressure or the
lining may become loose. Either extreme can cause a collapse!
INCORRECT INSTALLATION –A shifting lining, broken bricks or a
collapsed lining are typical symptoms of a poor installation. Refractory installation in
a kiln is not the same as it is in stationary furnaces, which comprise 90% of refractory
installations. Ensure your refractory specialist is also a kiln specialist!
SPIRALING
5.4 SLOPE
Kiln piers are sloped from the feed end to the discharge end to allow the
process material to move from end to end. Even though the slope is not large, 2 - 4%,
it is enough to push the bricks toward the discharge end. Retaining rings, welded to
the steel shell, are sometimes used to prevent this from happening but the bricks can
press against the steel rings and be crushed.
The retaining ring is installed a short distance – equal to a few brick ring
widths – from the outlet of the kiln. If another ring is installed further up the kiln (not
in the discharge area) it should not be located in the vicinity of a tire.
THERMAL EXPANSION
69
5.5 THERMAL STRESS
Concave Melting Pits
Overheated brick sometimes called “duck-nesting”.
CONCAVE MELTING PITS
Standard-grade bricks that have no coating buildup and are overheated are weakened
and look melted. This usually is the result of the flame hitting the lining directly.
Adjustment of the burner eliminates this problem. The flame envelope is around
3000F and no refractory can withstand direct flame impingement.
THERMAL
Heat effects working on the lining
A. OVERHEATING OF BRICKS - wash out or “duck nesting”
B. OVERHEATING OF CLINKER - densification of the hot face
C. EXCESSIVE THERMAL LOAD - causes structural fatigue of bricks
D. THERMAL SHOCK - thermal tensions cause horizontal cracks
OVERHEATING OF BRICKS weakens the brick structure of the hot face. You will
see concave wear, so-called “duck-nesting”. This situation can be caused by poor
choice of refractory for the process and/or flame impingement on the lining.
OVERHEATING OF CLINKER changes the mechanical properties of the bricks
and causes densification at the hot face. Symptoms include lava-like coating solidly
connected with the bricks, and falling coating which will take off the densified brick
heads. This phenomenon is caused by the formation of an increased liquid clinker
phase infiltrating the hot face.
EXCESSIVE THERMAL LOAD generally occurs during overheating above 1700°
C without a liquid phase. This will cause structural fatigue of the bricks. The grain of
the brick matrix will look like brittle needles instead of a round grain.
THERMAL SHOCK is caused during sudden temperature changes. This generally
occurs during quick heat-up or sudden cool-down. Coating loss can also cause this
problem. He bricks will spall in layers.
70
The question is always, do we bring the kiln down for a longer period of time
and replace the whole burn zone and transition area, or do we just patch a small area
and get up and going again more quickly. The answer of course depends on prevailing
conditions.
PATCH & GO - Making small repairs throughout the year requires several cooling’s
and heating’s of the kiln. This, as we have seen, is hard on the kiln, causing bricks to
spall. It also adds lost production time and increased fuel costs. This method is
generally used only after trouble develops, such as a “hot spot” in the shell.
71
No discussion of troubleshooting kilns, dryers, and other rotary equipment
would be complete without the inclusion of the special problems of gears, pinions,
and drive trains. Understanding the mechanical function of these parts and the
differences in their design will help to focus on areas that may cause problems in the
future, or may already be causing problems.
The alignment and mesh of the gear and pinion, the lubrication that is used,
and the other components involved in the drive train, are all areas that should be
carefully inspected on a regular schedule.
Covered in this section are the two most common types of gear mounting
arrangements, their pros and cons, and a recommended inspection program.
Characteristics, advantages and disadvantages of different types of lubrication are
discussed as well.
72
6.1 FLANGE-MOUNTED GEARS are found on older kilns or on
equipment such as dryers and coolers that operate in a lower temperature range. They
are generally constructed in 2 to 6 sections, and the mounting flange is typically
welded directly to the shell. It is easy to reverse the gear to maximize service life if
the flange is still rotating true and in good condition. However, a common problem
encountered when attempting to
reverse or install a new flange-
mounted gear is that over time the
shell has incurred thermal damage
because of refractory failures.
Depending upon the location of this
damage, it can create a significant
amount of run out
In the shell where the gear is
mounted. If this is the case, the
mounting flange must be cut loose and remounted or a new flange must be installed.
In some cases additional centering of the gear can be accomplished by
redrilling the mounting holes in either the gear or the flange. This can also be limited
by the way the flange was originally machined.
73
A careful inspection should be made for these factors and consideration given
prior to planning any gear reversal projects.
6.2 SPRING-MOUNTED
GEARS are the predominant type of
girth gear arrangement for kilns. There are
two basic variations, the most common,
the tangential spring, and the horizontal
spring bar. The positive aspect of these
designs is that they allow the kiln shell to
expand without restriction, unlike the
flange-mounted gear.
When installing a new gear or
reversing an existing gear, it is always
advisable to replace the spring plates and
the pins. The plates are subjected to
thousands of cycles over the years and as
such there is always concern for fatigue
Figure12: spring mounted gears
74
failure. Also, since the springs are welded to the shell, they are usually compromised
to some degree during the removal process. Trying to reuse the spring plates can
greatly inhibit the ability to achieve acceptable gear run-outs
Add life to your gears through regular inspection. Establish a check list for
systematic inspections daily, monthly and annually. The following areas are critical
elements of good gear maintenance
Mesh - Proper backlash and root clearance are critical to smooth gear operation. Too
much or too little clearance can cause problems that contribute to accelerated wear
and loss of tooth profile. Also important is the longitudinal relationship of the gear
relative to the pinion.
Contact - The contact pattern across the tooth face is a telltale sign of gear alignment.
If an axial run out condition exists, the contact pattern will vary with the rotation of
the kiln. Angular misalignment between the gear and pinion results in partial tooth
contact, concentrated loads and premature gear wear or failure.
Reducer condition - Inspect for vibration and elevated temperatures of the reducer.
These are usually signs of worn bearings or gears, or a lubricant problem.
75
Inspection ports - It is important to have
observation ports in the gear enclosure to
allow for adequate inspection of the gear
teeth. There should be ports on both sides
of the enclosure to allow inspection of the
depth of the mesh on both sides of the
gear, and a port in the area above the
pinion to allow inspection of the contact
pattern across the teeth, just after the gear
teeth are released from the pinion.
Most girth gears are a two-piece segmented style. This means that the splice
bolts will also need to be checked to make sure they remain tight.
76
Areas, boss pins and splice joints should be observed as they pass across the
pinion. In this area the gear is in tension before it comes into pinion mesh and in
compression after the mesh. This change of load from tension to compression will be
visible if there is any looseness of fit in the pins or bolts in the splices. Lubricant
should be present in both areas and movement can be detected as the lubricant is
squeezed or sucked into/out of the joint.
77
bearings or one of the bearings in the gear reducer. Analyzing the vibration frequency
by using vibration analysis equipment can help pinpoint the source.
Welds - On spring-mounted gears, check the welds attaching the spring plates to
the shell. On flange-
Mounted gears inspect the welds between the shell and flange. These are both
highly stressed areas subjected to cyclical loading and therefore susceptible to fatigue
cracking. Keep welds 1” (25mm) back of the tangent point.
Mounting and Splices - Visually inspect for loose or missing mounting bolts on a
flange-mounted gear. On a spring-mounted gear, the hole where the spring plate
attaches to the gear is called the “boss hole”. Inspect the boss holes and the pivot pins
for any sign of wear or excessive movement. Most girth gears are a two-piece
segmented style. This means that the splice bolts will also need to be checked to make
sure they remain tight.
The kiln will not generally be shut down to allow for physical testing or inspection
of the bolts and welds. Therefore you must learn to look for signs that indicate
potential problems. A simple way to look for loose bolts or cracked welds is to
watch them as they pass across the pinion. The gear is loaded in tension as it
approaches the pinion and switches to compression as it disengages the pinion.
Normally, everything is covered with lubricant and any movement such as a loose
bolt or cracked weld will result in “squishing” or “bubbling” of lubricant as this load
cycling occurs. If these signs are obvious, further physical examination is warranted
during the next outage.
Tooth engagement - Most gears
are manufactured with pitch lines
scribed on both sides. Under ideal
conditions the pitch lines should be
lined up or slightly separated during
hot running condition. If the pitch
lines are visible, watching them as the
kiln rotates can give you an indication
of radial run out and mesh conditions. Figure15: gear nomenclature
Unfortunately this is seldom the case. Lubrication can make the pitch lines difficult, if
78
not impossible, to see. Or, if the gear has at some point rubbed against the enclosure
the scribe lines are most likely worn away.
Another method of determining proper tooth engagement is to measure the
backlash, or the amount of clearance at the pitch line between the tooth space and the
corresponding tooth width. This measurement is only useful with new gears. Once the
gear teeth exhibit any signs of wear this measurement is no longer valid. In many
cases, a more reliable method of checking proper tooth engagement is by observing
the root clearance of the gear teeth throughout the rotation. It is important to note,
that while looking for root clearances, you must also look closely for a worn
tooth profile that could cause “false bottoming”. As a general rule of thumb, the
hot operating root clearance on most kilns should be approximately 5/16”. This is
based on the common 1 DP or 3” CP gearing. It is very important to know you’re
gearing so that actual clearances can be determined.
79
D. Pressure spot on total circumference ring gear. Manufacturing fault or new
pinion wearing into existing wear pattern of ring gear.
E. Pinion diameter enlarged on both sides due to wrong assembling or ring
clamping devices.
F. Ring gear joints and fixation of both ring gear sections must be checked.
G. Radial run-out of ring gear. The picture of contact is stronger over one half of
the circumference. Radial readjustment is necessary.
H. The ring gear has opened on both sides due to thermal expansion.
I. Ideal pattern.
6.5 VIBRATION
Check to see that the gears are running with proper root clearance or backlash.
Too little clearance will cause gear teeth to “jam” together as they mesh causing a
vibration problem as the top land of the tooth “pounds” into the bottom land. If you
observe a vibration that is intermittent, check first for interference on either end of
the kiln end of the gear enclosure. If there are no signs of any rubbing, it is most likely
that the shell has a bend that is causing the gear to run out. If this is the case, the
vibration will be in a direct relationship with the shell warp. As the bent area of the
shell passes across the pinion, the vibration can be felt as the gear teeth “’bottom out”.
If the vibration is constant throughout the entire rotation, the kiln is acceptably
straight but the contact surfaces of the riding rings, support rollers or filler bars have
worn enough to lower the kiln to a point of interference of the gear mesh.
Too much clearance creates excessive backlash, which under certain
conditions also causes a vibration problem. If there is a severe bend in the shell, even
though it is not an area that causes gear run out, or perhaps a large uneven buildup of
product coating, an unbalanced load is created in the kiln. As the heavy side of the
kiln passes over the top center and proceeds on the downward side of rotation, the
gear will try to “overrun” the pinion. If excessive backlash exists, intermittent
vibration will appear at a frequency equal to gear tooth engagement frequency.
Careful visual inspection of the gear for these conditions can help determine
the cause and then a solution to vibration problems. Usually a stroboscope must be
used for the inspection, and you will need to pay close attention to the occurrence and
duration of the vibration with respect to the gear mesh. A temporary clamp on current
80
probe can also be used on the drive motor to evaluate an intermittent vibration
problem. Typically if the motor current reads high during the vibration phase, the unit
is straining to lift the heavy zone of the kiln which is causing gear run out and a
resultant “bottoming” condition. Conversely, if the motor current is low during the
vibration phase, the motor is “loafing” as the bend in the shell advances to the down-
turning side and backlash allows the gear to “over run” the pinion.
6.6 ANNUALLY
A. Clean & measure amount of wear to teeth.
B. Fix leaking seals.
C. Check teeth for abnormal wear.
D. Change gear lubrication.
E. Clean bottom of gear guard.
During the annual inspection, since the kiln will be shut down, more thorough
inspections should be made of the gear wear patterns. Remove a portion of the gear
enclosure and with solvents, clean a 1-2 foot long area of the gear teeth. Measure the
tooth thickness on both the uphill and downhill edge and make note of the type of
wear pattern. Rotate the kiln and repeat this process in 4-6 other areas. Look for
ridges” or “steps” in the tooth profile. In some cases, if they are not too severe, these
areas can be ground down with a small hand grinder. This inspection should give you
a good indication of how the pinion has been meshing with the gear and if there is
excessive run out.
Gear lubrication should be changed out taking extra care to remove all spent
lubrication in the bottom of the enclosure. This is the area where most of the
contamination will settle.
Inspect the labyrinth seal of the enclosure for signs of wear. Fix the seal and
any holes in the enclosure as necessary. Keeping the enclosure in good condition will
help minimize dust and other contaminants from getting into the lubricant.
Lubrication contamination is one of the major causes of wear.
If the inspection of wear patterns indicates a run out problem, check both the
axial and radial run out. These measurements can only be taken while the kiln is shut
down. Also, if warranted, check the slope of the pinion. If any adjustments or
shimming of the pinion bearings are required, realign the balance of the drive
components and couplings.
81
6.7 AXIAL RUN OUT
82
taken on both indicators. The readings directly opposite each other are then
mathematically averaged to find the amount of “wobble” in the gear.
The reason for using two indicators is to negate any axial float of the kiln and any
“wobble” of the thrust rollers.
Illustrated on the chart above are typical curves for the amount of acceptable
axial run out of girth gears for kilns and similar roller-supported machines.
The allowable axial run out is a function of the gear diameter. For the most
part, gear manufacturers and OEMs of rotary equipment recommend 0.001” of axial
run out per foot of pitch diameter. This rule applies to new installations, and to gears
running at 5 rpm and over.
Another level of tolerances is shown for new installations with gears running
at less than 5 rpm. This is generally a 25% additional run out allowance.
The “acceptable tolerances” should be applied to installations where the unit
has been heated and expansion has taken place, or wear is present. In this case, it is
unlikely that original tolerances can be attained.
This is just a guide. Values specific for any particular gear should be
obtained from the gear manufacturer.
83
stem is held back.
The gear is stopped
at each of the twelve
teeth and an
indicator reading is
recorded.
A rigid setup
is required for the
dial indicator and
extreme caution
should be taken so
as not to disturb the
indicator placement during the procedure.
Take the maximum reading and subtract the minimum reading. This is the
radial run out. Compare this to the acceptable value for your gear. Values exceeding
the acceptable values will reduce the service life of the gear set.
Illustrated in the chart above are typical curves for the allowable radial run-out
of girth gears for kilns and similar roller-supported equipment. The allowable radial
run-out is, as with axial run-out allowances, a function of the gear diameter.
84
Gear manufacturers and OEMs of rotary equipment typically recommend 0.0015 of
radial run-out per foot of pitch diameter. This applies to new installations and to
gears running at 5 rpm and over.
The level of tolerances shown for new installations with gears running at less
than 5 rpm is calculated by adding an additional 50% to the run-out allowance.
The “acceptable tolerances” should be applied to installations that have been in use
for a period of time. These are unlikely to attain the original tolerances.
This chart is just a guide. Values specific for any particular gear should
be obtained from the gear manufacturer.
85
A seal is essential at each end to prevent ambient air from entering the process.
Air leakage can have varying effects on different process. A leaky seal may reduce
efficiency; create unsafe or unstable conditions.
Each seal is designed for the kiln diameter, maximum run out, temperature, and
pressure that it needs to isolate. There are many kinds of seals and each offer
advantages in their application.
86
In the rotary kiln, raw material is fed through the inlet chamber from the inlet end gets
discharged as the clinker into cooler the discharge hood. Both ends of kiln have fixed
hoods, where rotating kiln ends protrude in to them. such intrusions in to fixed hoods
leaves small gaps all around through which ingress of the false air take place. These
gaps need to be closed by providing seals.
Any discussion of drum end seal maintenance can be lengthy because of the
wide variety of seal designs used over the years by various rotary equipment
manufacturers. The designs are as simple as a piece of rubber belting or ceramic fiber
blanket, or as elaborate as utilizing counterweights, pulleys and air cylinders in an
attempt to create a more positive seal.
The primary purpose of any seal is to keep “tramp” air out of the unit to allow
better control of the process. A secondary but important function is to help control
87
dust and hot gases from escaping from the kiln or dryer if there is some sort of
“upset” condition.
To be successful, the seal for a rotary kiln or dryer must accommodate both
axial and radial movement. The major downfall of most seals is too many moving
parts susceptible to “sticking” or failure from dust contamination or heat distortion.
The three main causes for seal failure are abrasion from product contamination,
excessive shell run out in the seal area, and expansion and contraction of the unit
during operation.
For the purpose of this discussion maintenance issues of four common types
of seals will be reviewed. The four types are the graphite block seal, face plate type
seals, the outward facing leaf seal and the inverted leaf seal.
The graphite block seal has been around for many years with several
88
variations. The basic components of the seal are two rows of graphite blocks with one
end of each block contoured to fit the radius of the shell. The blocks are positioned
radially in slide holders that are spring loaded to keep them in contact with the shell.
The graphite blocks are staggered in the two rows with every other block overlapping.
Many of these installations have now been converted to some style of leaf seal.
The typical problems encountered with the graphite block seal are heat
distortion of the block holders, and dust accumulation causing the blocks to “bind” so
they can’t move freely in their respective slides. The result is that the springs can’t
keep the blocks in contact with the shell. Another problem is spring failure. Various
spring designs have been tried with little success.
The face plate type of seal has many variations, but the operating principle of
most of them is the same and therefore the maintenance issues are similar. We will
use the arrangement shown in this illustration as an example for general discussion.
The main sealing components are a floating seal plate mounted on the face of
the hood, and a fixed seal plate that rotates with the shell. The floating plate has some
type of flexible section attached to it which allows the seal to move with the shell.
The fixed and floating plates have a wear plate of brass or phenolic material
sandwiched between them and held in place with springs, counterweights or air
cylinders.
89
Contamination can cause this type of seal to stick. Typically the floating half
gets stuck and the sealing rings will disengage with any axial movement of the shell.
Product may also build up to the point that it deforms the flexible section of the
pressure assembly and causes failure.
COMMON PROBLEMS causing the face plate seal to leak and become
ineffective:
A. Failure of the device applying pressure to the seal plates. The device should be
inspected regularly.
B. The wear plate can become too thin or “scratched” and “gouged” from
contamination; the seal will leak.
C. Excessive run-out in the shell. These types of seals are particularly susceptible
t
o
f
a
i
l
u
r
e
i
f the shell run-out causes the fixed seal plate to “wobble”. The flexible
assembly will allow longitudinal movement but cannot compensate for
“wobble” of the rotating fixed ring.
90
Another common type of seal is the overlapping leaf seal. The two basic
91
Most direct-fired units will
operate with a slight negative pressure
in the hood on which the seal is
mounted. An outward-facing seal acts
to seat itself tighter against the shell
with that prevailing negative pressure.
Hence, it creates a positive seal against
infiltration of ambient air which is the
seal’s intended purpose. However,
most combustion is not altogether
“smooth” and flame turbulence, even
on the best burners, will cause bumps
and puffs of positive pressure. The
outward-facing seal has an inherent
disadvantage because the positive
puffs blow the seal open allowing hot
air and dust to blow out.
The leaves should be made of a material that can withstand the temperature
extremes to which they will be subjected, and still maintain spring pressure without
taking a permanent set. The leaf spring should also exhibit some abrasion resistance to
provide an adequate service life.
The major drawback of the outward-facing leaf design is their susceptibility to
damage from dust accumulation.
92
7.5 INVERTED SEAL
The spring action of the individual spring plates is strong enough to hold
against the negative operating pressure in the firing hood. The first advantage of the
inverted seal is that a positive puff will act to tighten up the seal, preventing blow-
back. With no blow-back there is no hot air flow, no dust can be conveyed through the
seal, and dust accumulation is significantly reduced.
Secondly the seal leaves run on the riding band instead of directly touching the
kiln shell. This provides a sacrificial member if wear occurs. The riding band is
mounted in such a way to negate shell run-out at the time of installation.
Since the shell run-out will has been all but eliminated the flexing of the seal
leaves is minimized thereby extending their service life.
The third advantage of the inverted seal is that at the bottom the seal plates are
above a hopper which carries away product spillage. Unlike the non-inverted seal,
which “catches” product spillage, the inverted seal does not create a shelf on which
product can accumulate. Phillips recommends spraying a graphite powder onto the
riding band under the leaves to
lubricate them whenever possible.
This will help extend their service life.
93
The inverted leaf seal does
have a space requirement but the
real plus of the inverted seal is its
simplicity of design. There are no
moving parts to “stick”. The design
incorporates a method to keep the
dust buildup away from the sealing
components and it is very forgiving
of shell run-out.
94
CONCLUSION
In this study project I came to know that while operating the rotary kiln, so
many critical conditions I seen while doing this project. To maintain the kiln properly
in such conditions mechanical plays a vital role in shutdown.
The frequent stoppages of kiln tends to increase the maintenance cost and
decrease the life of the kiln so perfect analysis is to be recorded in the operation and
perspective remedies are performed in shutdown.
By effective utilization of the man power and raw materials in repairing the
machines will give the excellent results in production rate.
95
REFERENCES
1. http://en.wikipedia.org/wiki/Cement_kiln
2.http://www.flsmidth.com/enUS/About+FLSmidth/Our+History/Our+Product+Bran
ds/Phillips+Kiln+Services
3. http://www.thekilndoctor.com/
96