SUMMER INDUSTRIAL TRAINING

PROJECT

Scope of Predictive Maintenance and its

implementation in Diesel Locomotives
Submitted by: (G9P6)
Ayush Sharma (Delhi Technological University; BLKOS - 379)
Ghanvir Singh (Delhi Technological University; BLKOS - 378)
Banti Singh (Delhi Technological University; BLKOS - 380)
Rahul Kumar Meena (Delhi Technological University, BLKOS - 385)
Devender Pratap Singh (Punjab College of Engineering and Technology; BLKOS - 429)
Priyanka Palewal (Indra Gandhi Delhi Technical University for Women; BLKOS - 427)
Lakshay Vaid (Lovely Professional University, Jalandhar; BLKOS - 391)
Sandeep Chandra (Bipin Tripathi Kumaon Institute of Technology, Almora; BLKOS - 430)
Santosh Kumar (G.B Pant Institute of Technology ; BLKOS – 263)
 ACKNOWLEDGEMENT
We take this opportunity to express our sincere gratitude to all those people
who have helped us during our industrial training at the Diesel Locomotive
Shed, Tughlakabad. We would like to show our greatest appreciation to the
highly esteemed and devoted technical staff, supervisors of the Diesel
Locomotive Shed.
We are highly indebted to them for tremendous help and support during the
training.

We are highly grateful to Mr. Balram Meena (S.S.E – Running (mechanical)) for
their constant teaching and support during the training.

In particular, we are grateful to Mr. Om Kant Sharma, SSE/TRG (D.T.C.) of

Diesel Locomotive Shed, Tughlakabad, who scheduled our training in the
various departments of the shed. We would like to thank all those people who
directly or indirectly helped and guided us to complete our training and project
in the Diesel Training Centre and various sections.
 TABLE OF CONTENTS

1. Introduction to Diesel Shed, Tughlakabad

2. Types of maintenances
3. Predictive Maintenance vs Preventive Maintenance
4. Scope of predictive maintenance in diesel locomotives
5. Analysis of components : Possible implementation of PM
6. Possible changes in existing preventive maintenance
techniques
7. Conclusion
 Introduction – Diesel Shed,
Tughlakabad
Diesel Shed, Tughlakabad of Northern Railway is located in NEW DELHI. The
shed was established on 22nd April 1970. It was initially planned to home 75
locomotives.
The shed cater the needs of Northern railway. This shed mainly provides
locomotive to run the mail, goods and passenger services. No doubt the
reliability, safety through preventive and predictive maintenance is high
priority of the shed.
To meet out the quality standard shed has taken various steps and obtaining of
the ISO-9001-200O& ISO 14001 OHSAS CERTIFICATION is among of them. The
Diesel Shed is equipped with modern machines and plant required for
Maintenance of Diesel Locomotives and has an attached store depot. To
provide pollution free atmosphere, Diesel Shed has constructed Effluent
Treatment Plant. The morale of supervisors and staff of the shed is very high
and whole shed works like a well-knit team.

Inception 22nd April1970

Present Holding 147 Locomotives
19 WDM2
37 WDM3A
08 WDM3D
11 WDG3A
46 WDP1
26 WDP3A WDP3A
Accreditation ISO-9001-2000 & ISO 14001
Covered area of 10858 SQ. MTR
shed
Total Area of shed 1, 10,000 SQ. MTR
Staff strength Sanction – 1357
On roll - 1201
 Types of maintenance

The basic types of maintenance include:

 Preventive or scheduled maintenance, where equipment or facilities are

inspected, maintained and protected before break down or other problems
occur.
 Corrective maintenance where equipment is repaired or replaced after wear,
malfunction or break down.
 Predictive maintenance, which uses sensor data to monitor a system, then
continuously evaluates it against historical trends to predict failure before it
occurs.[10]

Preventive maintenance
Preventive maintenance is maintenance performed with the intent of avoiding
failures, safety violations, unnecessary production costs and losses, and to
conserve original materials of fabrication. The effectiveness of a preventive
maintenance schedule depends on the RCM analysis which it was based on, and
the ground rules used for cost efficacy.

The type of maintenance adopted at Diesel Shed, Tughlakabad is Preventive

Maintenance. Maintenance of various locomotives is done on the basis of pre-
decided schedules.

Corrective maintenance
Corrective maintenance is a type of maintenance used for equipment after
equipment break down or malfunction is often most expensive – not only can
worn equipment damage other parts and cause multiple damage, but
consequential repair and replacement costs and loss of revenues due to down
time during overhaul can be significant. Rebuilding and resurfacing of equipment
and infrastructure damaged by erosion and corrosion as part of corrective or
preventive maintenance programs involves conventional processes such as
welding and metal flame spraying.

Predictive maintenance
More recently, advances in sensing and computing technology have given rise
to predictive maintenance. This maintenance strategy uses sensors to monitor key
parameters within a machine or system, and uses this data in conjunction with
analyzed historical trends to continuously evaluate the system health and predict
a breakdown before it happens. This strategy allows maintenance to be
performed more efficiently, since more up-to-date data is obtained about how
close the product is to failure.
 Predictive vs Preventive
Maintenance

PERT Chart of M-24/M-72 schedule of Alco Locomotives

(PERT Chart schedules are classic examples of preventive maintenance in Indian
Railways, similarly a 3 yearly, 6 yearly and a 12 yearly schedule is also followed)

Basis of Maintenance :
Preventive maintenance is triggered by time, events, or meter readings , hence ,
another way to say preventive maintenance is planned scheduled maintenance
(followed by Indian Railways). However, this time-based maintenance approach
may not accurately reflect the usage of a piece of equipment, and could lead to
unnecessary maintenance fixes regardless of the actual state of the equipment or
parts. At the shed it was observed that, during their scheduled maintenance, few
parts of locomotives, which were in perfect working condition, were still
discarded, leading to wastage of potential resources, money and time.

On the other hand, predictive maintenance is based on the actual condition of the
equipment rather than time or age factors. It is used to predict machine failures
before they occur, and also gives company enough time to schedule a future
service appointment in advance. This allows field service technicians the
opportunity to service the machine and prevent the failure before it actually
occurs.

Maintenance Techniques :
With preventive maintenance, since fixing machinery is based on time and the
breakdown rates of similar parts, the field service tech will replace or repair parts
based on their expected failure dates. This date is not based on performance data
of the specific machine.

For predictive maintenance, various advanced techniques including infrared

thermal imaging, vibration analysis, and oil analysis can be used to predict
failures. For example, in diesel locomotives, thermal infrared imaging can scan,
visualize, and analyze the equipment’s temperature. We can literally see which
parts on the equipment are “running hot,” which is invaluable information to
both manufacturers and field service technicians.

Costs and Savings

As far as maintenance costs are concerned, in preventive maintenance sometimes
parts are replaced even when there is no need for replacement.

But predictive maintenance has several cost savings that range from minimizing
the time equipment is not working to cutting down on the price of spare parts
and supplies.
While the investment into a predictive maintenance program can be costly to
install, it results in reduced maintenance costs and downtime. Field
service software can be used along with predictive maintenance tools to make
sure the locomotives are running efficiently.
 Scope of predictive maintenance
in Diesel Locomotives
Recent advancements in sensors and communication technologies have given rise
to the possibility of DAQ (data acquisition) i.e continuous data collection from
various systems and subsystems in trains, enabling monitoring of mechanical and
electrical conditions, operational efficiency and multiple other performance
indicators.

These new capabilities enable planning of maintenance activities with the

maximum interval between repairs, while minimizing the number and cost of
unscheduled outages. If the problem is detected early, major repairs can usually
be prevented. Today’s technology makes it possible to affordably collect huge
amounts of data from hundreds of systems in a single train, analyze that data in
real time and detect problems before they actually happen, hence, making it one
of the most promising aspects of the rail industry’s digital transformation.
Using data collected on equipment during operation to identify maintenance
issues in real time means repairs can be properly planned, with the benefits that
trains don’t need to be unexpectedly taken out of service for emergency or
unnecessary routine maintenance.

When it comes to Indian Railways focus tends to be on keeping the locomotives in

service for as long as possible, to get value out of the considerable initial
investment.
This means that technologies enabling predictive maintenance, reducing
operating costs and extending a fleet’s lifetime, have the potential to deliver huge
financial rewards. However, it also means that older trains currently in service,
that were not built for modern connectivity, require investment to be able to do
so.

For implementing predictive maintenance in Indian Diesel Locomotives, there are

2 major things need to be done :
1) Installation of black box data recorders and sensors to meet statutory
requirements. This goes along with being committed to the mandatory
refit/overhauling/preventive maintenance. It would give the opportunity to
consider what could be done to improve fleet performance and address the
particular problems.
2) Connect these black box recorders with a platform that gave a real-time
view of the state of the trains. All captured data would be stored and
regularly analyzed on running time by a Machine Learning / AI Algorithm.

This platform should, obviously, be designed with extensive input from

maintenance engineers and other train experts. This will help identify exactly
what data is desired for acquisition. The team would also help identify
additional sensors that could be combined to provide useful information.

Right data would mean the right insight. Well-designed data analytics would
identify the faulty component and the problem. Changes in running conditions
are a precursor to failure, so it is possible to spot deterioration patterns well
before failure occurs. The maintenance engineer would be, hence, informed
which component and what to look for, saving time and cost of unnecessary
disposal of other components.

This approach has few prerequisites : There must be many high speed sensors to
detect the state/running condition of the component.

Having planned the capture of the right data, it is then relatively straightforward
to design suitable algorithms to monitor the components. Components can be
individually calibrated to measure what is normal and a well-designed machine
learning / AI algorithm will consider differences between usual and unusual
patterns/conditions. Parameters can be set appropriately.
The principles of this approach hold for a wide range of challenges : Collect good
quality data which is indicative of normal operations, design algorithms which
recognizes deviations from the normal and match changes to profiles of
problems. This enables quick identification when something is going wrong and
what it is likely to be – before failure occurs.

In the upcoming section, possible implementation of PM (predictive maintenance)

has been considered i.e how failure occurs in the component, what data should
be captured for acquisition and how would the data be captured.
 Analysis of components: Possible
implementation of PM

POWER ASSEMBLY

The term power assembly refers to an Electro-Motive Diesel (EMD) engine sub-assembly designed to be
"easily" removed and replaced in order to restore engine performance lost to wear or engine failure.
Typical of heavy-duty internal combustion engines used in industrial applications, EMD engines are
designed to allow the cylinder liners, pistons, piston rings and connecting rods to be replaced at
overhaul without removing the entire engine assembly from its application location. This increases
engine value, reduces downtime and allows the engine to be returned to true new engine performance.
Other terms such as cylinder pack, liner pack, cylinder assembly and cylinder kit are used in the engine
industry to describe similar assemblies. In the large-engine industry, the term "power assembly" has also
become generic and is often used to refer to the assemblies used in non-EMD engines where "power
pack" may be the preferred term, although both terms are functionally the same.
An EMD power assembly consists of the following components:

1. Cylinder head assembly (including valves, springs, keepers etc.) less fuel system components
2. Cylinder liner
3. Piston and piston rings
4. Piston carrier
5. Connecting rod
In an EMD diesel engine, since two power assemblies share a common connecting-rod journal, and since
the power assemblies are directly opposite each other rather than staggered as in a typical V-type
engine, two different power assemblies are required in a single engine. The difference between the two
assemblies is in the connecting rods. One connecting rod "big end" has to fit inside that of its companion
rod and the two types are referred to as "blade rods" and "fork rods". The "fork rod" is logically the
"master" as only it has a "rod cap", in this specific case referred to as a "basket", whereas the "blade
rod" is logically the "slave" as its "big toe" is designed to fit completely within, and is guided by, and is
retained by the "fork", and both are retained by the single "basket".To inspect the engine, it can be
manually "barred over" with a lever, but manual engine rotation is slow and inefficient. In some
applications manually barring the engine over can be difficult or impossible. The preferred tool for
engine rotation is an electrically powered, hydraulically operated "turning jack". The turning jack uses a
hydraulic cylinder and ram assembly that automatically advances to engage a hole in the flywheel.
When the ram reaches its limit, it automatically retracts and advances again to engage another hole. The
engine is then progressively rotated through its cycle and can be rotated in either direction by installing
the jack on either side of the engine. Not only is a turning jack faster and more efficient, it is also safer
since there is no risk of a barring lever coming loose and causing injury or damage. Also, with a turning
jack, there is no need for the mechanic to be in physical contact with the engine at any point during the
inspection process.

CRANKSHAFT
The engine crankshaft is probably the singular costliest item in the diesel engine. It is the
medium of transforming reciprocating motion to rotary motion. Output of the engine is
collected from crankshaft. WDM2 locomotive is having electrical transmission system hence
to convert Mech power of engine into electrical power a Generator is fitted at the traction
output end of the engine. The Rotor of the Traction Generator is mounted directly
on the crankshaft through flange joint. The crankshaft may be assembled type or two pieces
bolted type or may be single piece forging. Counter weights are provided for dynamic
balancing of the crankshaft; they are either bolted up or welded. The standard Locomotives
of Indian Railways are provided with single-piece crankshaft with welded counter weights. In
case of CLW/MAK engines the counter weights are bolted. The portions of crank shaft which
form the axis of rotation are called main journals where it is housed in the engine block
through main bearings and the eccentric portions where piston is connected through
connecting rod are called crankpins. In case of WDM2 type engine there are 9 main journals
and 8 crankpins to accommodate 16 pistons in a V shaped engine. Through internal drill
holes of crank shaft lube oil is supplied for lubrication of main bearing and con rod bearing
portion and the same oil is further circulated for piston cooling through internal drill hole of
con rod.
The ALCO crankshafts are manufactured from chrome-molybdenum steel equivalent to
SAE 4140. The process of forging is such that continuous grain is maintained. In
manufacture of crankshaft, following sequence of operation is generally followed: -
Forging and forming operation
Rough machining
Drill of oil holes.
Ultrasonic & Mechanical testing
Welding of counter weights & their X-ray test.
Stress relieving & shot blasting
Final machining & for giving fillet radius at crank journal corners and making oil holes.
Nitriding
Grinding Lapping
Static & dynamic balancing
Final inspection
There are two processes of surface hardening with details given below:-
Method of hardening Hardness Depth of hardness
Induction hardening C-40 0.124"
Nitriding C-60 0.012 to 0.015"
Generally for low HP engines the first process is preferred, as depth of case is more
and the crank journals and man bearing journals can be ground down to next step
size.
Predictive Maintenance
Schedule: POH
Procedure: After cleaning thoroughly, Dye penetration / Magnaflux test is conducted
todetect surface crack. The following dimensional checks are to be carried out:
Crank pin: Positioning it vertically check dimension at two locations justbeside two oil
holes (at two right angular planes in each location) to check ovality and taperness.
Nominal Dia: 6", Limit upto 5.996"
Ovality: .002"(max) Taperness: .001"(max)
Main journal: Position the crankshaft, keeping No 1 crankpin in verticallocation, measure
the dimension as that of crank pin.
Nominal Dia: 8.5", Limit upto 8.496"
Ovality: 0. 002" (max) Taperness: 0.001" (max)

Fillet Radius: Checked through a special gadget. (0.0005" filler gauge shouldnot pass
between the gadget and the fillet)
Eccentricity checking: Eccentricity is checked between any three
consecutivemain journals (1,2,3) is given by the distance between the center
points of journal 2 and the mid point of the line joining the center points of
journals 1
3. The limit of eccentricity is .001". Eccentricity is checked by the following way:
• Place the crankshaft horizontally on a "V" block supported at No3 and No 7 Main
Journals, keeping No 1 crank pin in vertical position.
• Mark Dial of a clock at the free end flange in this position, to understand angular
location of the maximum deviated zone.
• Record the readings of maximum deviation of every main journal along with their
angular location.
Crank web deflection: Checking of crank web deflection is one of the majorworks while
assembling engine.
Main generator rotor is coupled at one end of the crankshaft, whose other end is supported
on a bearing housed at the magnet frame. As such, for any dislocation of magnet frame, if
axis of armature does not completely align with the axis of the crankshaft, the unbalanced
mass of armature will cause uneven loading on crank web at different angular positions
during rotation. This causes DEFLECTION ON CRANK WEB, which will be changing at various positions
of crankshaft during rotation. Such kind of continuous cyclic variation of load leads to main
bearing seizure and breakage of crankshaft.
The crank web deflection is measured by fitting a deflection gauge at the located punch
mark on the 8th crank web (web nearer to TG) and manually rotating the crankshaft in
both the directions
The permissible limit of deflection on each side is ±0. 0008", TIR ±.0.0016".
If the deflection is beyond the permissible limit, correction is made by adding or
subtracting shims at the mountings of magnet frame with engine block. The magnet
frame is mounted at two locations with the engine block and at two locations at the
base. Adjustable shims are provided at the mountings of the magnet frame with the
block. The shims of the magnet frame with the base are fixed and normally not disturbed
during crankshaft deflection.

CAM SHAFT
In diesel engine the cam shaft performs the vital role of opening and closing inlet and
exhaust valves and operates Fuel Injection Pump for timely injection of fuel inside the
cylinder. Usual practice is to provide 3 cams for each cylinder the two outer cams being
for exhaust and inlet valves and the central cam being for fuel injection.
ALCO engines have cams integral with camshaft. Each camshaft section takes care of two
cylinders. After profile milling of the cam lobes the cams are given for induction
hardening. Subsequent to this the cams are put on profile grinding machine. The
individual camshafts are joined together by bolting. The location of dowel hole is of
importance as it determines the relative angular position of one camshaft section with
respect to the adjacent one.
In order to avoid wrong assembly respective catalogue Nos. of camshaft sections are
punched onto the shaft. Care has to be taken to see that the correct section is fitted in
correct location. The rifle hole is made in the center of the shaft for lubrication of cam
bearings. Lubrication to cam lobes is provided by oil coming from valve lever mechanism
via the push rod.

Material composition: The ALCO camshafts are made from AISI specification1050
with following metallurgical composition.
Carbon 0.48 to 0.53%
Manganese 0.60 to 0.90%
Chromium 0.15 to 0.30% Max.
Phosphorus 0.025% Max.andSulphur 0.025% max.

Predictive Maintenance
Schedule: POH & Major
Inspection: Check cam profile through profile gauge. (If damaged or worn out the cam
segment is changed.)
  Setting of Cam shaft & Valve/ FIP timing: Proper assembling and setting of
camshaft is essential to ensure correct valve and FIP timing.
 Timing Mark & Pointer
o Timing Marks are provided on Timing Disc (Main Generator fan) fitted with
Main Generator armature, mounted on crankshaft. (Relation between
crank shaft, armature and armature fan are maintained through dowels,
provided)
o TDC, INJ Pointer is mounted on the block to read the relative position
ofcylinder corresponding to pointer and timing disc. (Pointer needs to be
calibrated during engine overhauling by finding TDC with the help of dial
indicator or trammel gauge.)
 Setting of cam shaft
 Assemble the cam segments as per correct sequence and order (Part Number
indicates the sequence and dowels fix their angular relation). Thus left and right
side camshafts are formed.
 Position the crankshaft to 1R TDC. Match both side cam gears with crankshaft
gear in such way that cut marks (line mark) on cam gear should perfectly match
with block edges.
 Now fit both camshafts matching with the key holes of cam gears.
 Secure the camshafts finally on cam gear by tightening it properly through
locknut.
 Valve Timing
o Rotate the crankshaft and bring the corresponding cylinder to compression stroke
(ensure compression stroke by feeling free rotation of push rods.)
• Check the gap between the valve stem and yoke. It should be 0.034". If
not adjust it through valve lever adjustment nut and yoke adjustment nut.
(d) FIP Timing
• Rotate the engine to bring the injection point against the
corresponding cylinder.
• Match the body cut mark of FIP at inspection window with the given
line mark on guide cup .If not, adjust it through timing allenscrew
provided at the bottom of the FP lifter.

CYLINDER HEAD
The cylinder head is held on to the cylinder liner by seven hold down studs or bolts
provided on the cylinder block. It is subjected to high shock stress and combustion
temperature at the lower face, which forms a part of combustion chamber. It is a
complicated casting where cooling passages are cored for holding water for cooling
the cylinder head. In addition to this, provision is made for providing passage of inlet
air and exhaust gas. Further, space has been provided for holding fuel injection
nozzles, through Nozzle Sleeve fitted in the cylinder head. Valve guides and valve seat
inserts also fitted in the cylinder head casted body.
In cylinder heads valve seat inserts with lock rings are used as replaceable wearing part.
The inserts are made of stellite or weltite. To provide interference fit, inserts are frozen
in ice and cylinder head is heated to bring about a temperature differential of 250 F and
the insert is pushed into recess in cylinder head. The valve seat inserts are ground to an
angle of 44.5 whereas the valve is ground to 45 to ensure line contact. (In the latest
engines the inlet valves are ground at 30° and seats are ground at 29.5°). Each cylinder
has 2 exhaust and 2 inlet valves of 2.85" in dia. The valves have stem of alloy steel and
valve head of austenitic stainless steel, butt-welded together into a composite unit. The
valve head material being austenitic steel has high level of stretch resistance and is
capable of hardening above Rockwell –34 to resist deformation due to continuous
pounding action.

The valve guides are interference fit to the cylinder head with an interference of 0.0008"
to 0.0018". After attention to the cylinder heads the same is hydraulically tested at 70 psi
and 190 F. The fitment of cylinder heads is done in ALCO engines with a torque value of
550 Ft.lbs. The cylinder head is a metal-to-metal joint on to cylinder.
The cylinder head castings are made from special alloy cast iron as per specification
given below: -
Material composition:
Total carbon 3.00 to 3.40%
Silicon 1.80 to 2.20%.
Sulphur 0.12% to 0.8%.
Phosphorous 0.15 Max..
Manganese 0.65% to 90%
Chromium 0.20% to 0.40%
Nickel 1% Min.
Molybdenum 0.35% to 0.45%
ALCO 251+ cylinder heads are the latest generation cylinder heads, used in uprated
engines, with the following feature:
-Fire deck thickness reduced for better heat transmission.
-Middle deck modified by increasing number of ribs (supports) to
increaseits mechanical strength. The flying buttress fashion of middle
deck improves the flow pattern of water eliminating water stagnation
at the corners inside cylinder head.
 -Water holding capacity increasedby increasing number of cores (14
instead of 11)

-Use of frost core plugs instead of threaded plugs, arrest tendency

ofleakage.
-Made lighter by 8 kgs (Al spacer is used to make good the gap
betweenrubber grommet and cylinder head.)
-Retaining rings of valve seat inserts eliminated.
Benefits:-
-Better heat dissipation
-Failure reduced by reducing crack and eliminating sagging effect of
firedeck area.

Predictive Maintenance and Inspection

Schedule: Yearly
• Cleaning: By dipping in a tank containing caustic solution or ORION-355
solution with water (1:5) supported by air agitation and heating.
• Crack Inspection: Check face cracks and insert cracks by dye penetration test.
• Hydraulic Test: Conduct hyd. test (at 70 psi, 200°F) for checking water
leakage with the special attention at nozzle sleeve, ferrule, core plugs and

combustion face.
• Dimensional checks:
(a) Face seat thickness: within 0.005" to 0.020"
(b) Interference:
I.Valve seat insert to housing: 0.0015" to 0.0035" (Stellite)
0.003" to 0.005" (Weltite)
II. Valve Guide: 0.0008" to 0.0018"
III. Yoke Guide: 0.0015"
(c) Projected Height:
I. Valve Guide: 2.25"
II. Yoke Guide: 3.210" to 3.272"
(d) Clearance between valve and guide: 0.004" to 0.007"
(e) Thickness of valve disc & Insert: 5/32" (new) 3/32" (min)
(f) Straightness of valve stem: Runout should not exceed 0.0005"
 (g) Free & Compressed height (at 118 lbs.) of springs: 3 13/16" & 4 13/16"
• Checks during overhauling:
(a) Ground the valve seat insert to 44.5°/ 29.5°, maintain run out of insert
within 0.002" with respect to valve guide while grinding.
(b) Grind the valves to 45°/ 30° and ensure continuous hair line contact
with valve guide by checking colour match.
(c) Ensure no crack has developed to inserts after grinding, checked by
dye penetration test.
(d) Make pairing of springs and check proper draw on valve locks and
proper condition of groove and locks while assembling of valves.
(e) Lap the face joint to ensure leak proof joint with liner.
(f) Blow by test:
I. On bench blow by test is conducted to ensure the sealing effect
of cylinder head.

II. Blow by test is also conducted to check the sealing efficiency of

the combustion chamber on a working engine, as per the
following procedure:

a) Run the engine to attain normal operating temperature (65°C)

b) Stop running after attaining normal operating temperature.
c) Bring the piston of the corresponding cylinder at TDC in
compression stroke.
d) Fit blow-by gadget (Consists of compressed air line with the
provision of a pressure gauge and stopcock) removing
decompression plug.
e) Charge the combustion chamber with compressed air.
f) Cut off air supply at 70 psi. through stop cock and record
the time when it comes down to zero.
g) 7 to 10 secs is OK., if less check the leakage.
h) To check leakage, charge continuously at 70 psi
-Leakage through TSC indicates head defective.
-Leakage through Sump indicates defect in Piston or Liner.
(g) Tale-tell hole checks: Tale tell hole in cylinder head tells about the
condition of cylinder heads in running condition as per the following:
 I. No leakage: OK
II. Fuel droplets: Upto 2 drops/min OK.
If more, Nozzle leak off rate is high.
III. Fuel Mist: Nozzle seat defective.
IV. Water leakage: Nozzle Sleeve cracked.
V. L/Oil leakage: Rubber ring on Nozzle perished.

LINERS

Liner forms the wall of the combustion chamber as well as it also guides the movement of
piston inside it. Liners are mainly of two types i.e. (a) Dry liner (b) Wet liner.
(a) Dry liners are those, which does not come in direct contact with coolant but fits in
as a sleeve inside an already complete cylinder. The temperature of the inside surface
of dry liner is higher than corresponding wet liner. Dry liners are in use in only very
small engines.
(b) Wet liners are those, which not only form the cylinder wall, but also form a part of
the water jacket. ALCO Locomotives are fitted with wet liners, which have slight
interference fit on upper and lower decks. In addition to this, synthetic rubber seals of
suitable qualities are to be used, one on the upper deck groove (Si Rubber) and two on
middle deck (Viton rubber).

The ALCO cylinder liners are made of high strength close-grained alloy cast iron heat
treated to relieve stresses. The liner metal composition of a typical ALCO engine is given
below: -
COMPOSITION
(Unalloyed cast iron grade 17)
Carbon 3.00 -0 3.50%
Silicon 1.70 to 2.30%
Sulphur 0.12% Max (mandatory)
Phosphorus 0.15% Max (mandatory)
Manganese 0.60 to 0.90%
Chromium 0.25 to 0.60 %
Molybdenum 0.35 to 0.70% (mandatory)

Predictive Maintenance & Inspection:

Yearly inspection, 3 yearly & POH renewal

The cylinder liners suffer from the following major defects:
(a) Wear in the bore (Nominal bore 9”.Max allowed 9.009”)
Max ovality: .003”(max) Max taperness: .002”(max)
(b) Loss of interference in the top & bottom decks.
In the bottom deck portion, in between Liner and block a sleeve is used, made of
spheroidal gray cast iron. Hence in case of losing interference or any other defect
the sleeve is renewed.
Interference between block to sleeve: .004” to .008”
Int. between sleeve to liner: .0005”to .0015”
As the liners form water jacket with the engine block, hence for proper sealing
one rubber ring of Si- rubber and two rubber rings of Viton rubber are used at the
top and bottom deck respectively. It is essential to change the rubber rings while
renewing or removing liners.

PISTON
The piston is the most important component in the diesel engine as it forms a part of
the combustion chamber and also takes direct part in transmission of power. It is,
therefore, necessary that the designers and users must know the essential details
about the piston. The combustion of fuel results in large amount of heat being
developed. Out of this about 18% of the heat is absorbed by piston only.
The functions of the piston are: -
(a) It compresses the air to required pressure & temperature.
(b) It receives the thrust of expanding gases and transmits the force through
connecting rod (for rotating crankshaft).
(c) It forms the crosshead through which side thrust due to angularity of connecting
rod is transmitted to the cylinder wall.
(d) With the help of piston rings it prevents leakage of gas from combustion chamber
to crank case.
Guiding factors for dimensions are as follows: -
(a) The top portion of the piston is in contact with direct heat of combustion. Inspite
of cooling arrangement, it takes up more expansion and as such the need for more
clearance at this location.
(b) Relief has to be provided at the piston pin located area to prevent seizure of
piston due to bulging of material at this location in course of working.
Ring Grove Insert
The top most ring bears the maximum burnt of high pressure hot gases. This result in
heavy wears in the upper ring groove. In order to over come this problem, Ni-resist
ring insert is fitted in the uppermost ring groove. Ni-resist rings apart from being
dove tailed in Aluminium casting/forging, are molecularly bounded to the Aluminium
body by AI-FIN-process.

Piston material:
In many ways cast iron is best-suited material for manufacture of piston. The reasons
are as follows: -
(a) Co-efficient of expansion matches with cylinder liner whereas Aluminium has got
twice the co-efficient of cast iron.
(b) Heat conductivity is 3 times better than Aluminium.
(c) Compressive strength is much more than Aluminium at high temperature.
(d) Wear is less than Aluminium.
But the two main disadvantages with cast iron piston are: -
(a) Weight of Aluminium is 0.097 1bs. per cubic inch whereas in place of cast iron it is
0.284 1bs. per cubic inch. Thus cast iron pistons are about 3 times heavier than
Aluminium piston in weight.
(b) Possibilities of cylinder liner being scored are more in case of cast iron piston.
The factor of weight has become more over riding in view of the high speed of the
modern diesel engines and hence Aluminium alloy pistons are favoured. ALCO 251
engines pistons are of Aluminium alloy with composition given below-
COMPOSITION
Copper 5.8 to 6.8%
Zinc 0.10% max.
Manganese 0.20 to 0.40
Titanium 0.02 to 0.10
Vanadium 0.05 to 0.15 %
Zirconium 0.10 to 0.25 %
Silicon 0.20% Max.
Iron 0.30% Max.
Magnesium 0.02% Max.
Other 0.15% TOTAL
Aluminium - remainder
These pistons are in two parts i.e. the piston body (or skirt) and the ring carrier having
interference fit. The joint between the ring carrier and piston is welded at the crown
by inert gas welding.
Mahle has developed single cast Al alloy piston, reducing the chances of dislodging of
ring carrier during working.
Steel cap pistons are used in Fuel efficient and upgraded engines.

PISTON RINGS
The main functions of piston rings are: -
a) Sealing of combustion chamber and thus prevents blow by of air and high
temperature combustion gasses from getting access to crank case.
b) Scraps down excess lube oil from walls of cylinder liner and thus prevents reaching
lube oil into combustion chamber.
Piston rings are made of malleable grey cast iron with open graphite structure and a
hard pearlitic matrix. The piston ring operates during a part of its life under
conditions of marginal lubrication hence material composition has important role in
this regard.
Piston rings are used in combination to perform the above functions. They are either
5 ring combination or 6 ring combinations. Now a days 5 ring combinations are in
use.
Compression Rings: 1. Square Face 1. Square Face
2.Taper Face 2. Taper Face
3. Taper Face 3. Taper Face
Oil scrapper Rings: 4.DoubleTaper face 4.DoubleTaper Face
5. Conformable 5.Double Taper face
6.Conformable
In the latest fuel efficient engines barrel faced piston rings are used in place of square
faced compression rings and both the oil scrapper rings are conformable rings.

CONNECTING ROD
Connecting rod is a member connecting piston and crankshaft and is a medium for
converting the reciprocating motion to rotary motion. In four stroke engines during
the compression and power stroke the connecting rod is subject to high compressive
load. In suction stroke it undergoes high tensile stresses. In case of two-stroke engine
the connecting rod is only subject to compressive load. Connecting rod length is
usually about 4 to 5 times of the crank radius. They are I beam sections of fine-
grained, fully killed alloy steel forging. Connecting rods are having a fine-drilled hole
from the big end to the small end for transporting oil for lubrication at small end
bearing and piston pin and for cooling of piston.
The connecting rod assembly consists of: -
(i) Connecting rod, (ii) Connecting rod cap (iii) Piston pin bushing (iv) Bearing Shell
upper (v) Bearing Shell lower (vi) Connecting rod bolts and nuts.
During assembly the bolts are to be tightened with specified torque value and elongation
upto .015” to .018”. Connecting rods are mostly made of carbon steel or alloy steel
forging. The metallurgical composition of connecting rod is given below in percentage.

Predictive Maintenance & Inspection:

Schedule: Yearly
Cleaning: Solution of ORION 516 in HSD Oil is used for Piston cleaning

Checks: Zyglo test for checking surface cracks.

Visual checks for checking damages in piston crown, Ring grooves, circlip
groove and Ni- resist insert
 FAILURE ANALYSIS OF COMPONENTS &
FAILURE INVESTIGATION
A part or assembly is said to have failed under one of the three conditions-
When it becomes completely inoperable-occurs when the component breaks into
two or more pieces.
When it is still inoperable but is no longer able to perform its intended function
satisfactorily- due to wearing and minor damages.

When serious deterioration has made it unreliable or unsafe for continuous use, thus
necessitating its complete removal from service for repair or replacement-due to
presence of cracks such as thermal cracks, fatigue crack, hydrogen flaking.

GENERAL PRACTICE OF FAILURE INVESTIGATION

The objective of failure investigation, and subsequent analysis is to determine the primary
cause of failure, and based on this determination, decide on corrective measures, which
should be initiated to prevent similar failures. The principal stages of investigation are-

Collection of background data and collection of sample-

All available information regarding the manufacturing, processing, and service history
should be collected. Particulars and condition of other affected components should also
be noted. Details about operating conditions must also be noted meticulously. Selection
of the sample should be done prior to starting the examination.

Visual examination of failed components-

After the receipt of the broken and affected components in a metallurgical lab, each
sample is registered against a particular sample number. The fracture face is cleaned
with K oil. And soft metallic brush. Location of the fracture must be done in relation to
some fixed corner or side depending upon the specimen. Examine the fracture face with
a magnifying glass to determine the type of fracture. Nature of stress raiser can also be
determined. Examination can be done for the presence of welding or reclamation marks,
wearing etc., and if possible, photographs may be taken.
To determine the nature of fracture and stress raiser-

On the basis of visual examination, fractures may be classified as:

Ductile fractures: it involves a reduction in area and neck formation at the location of
thefracture. Overloading is the main reason of this type of fracture. Generally found
in tough materials

Brittle fractures: the entire fracture face is crystalline without any origin. Sudden shock
orloading is the main cause of this type of fracture. There is no reduction in area of cross
section at the point of fracture. Generally occurs in fragile materials. However,
 sometimes a part made of tough material, can sometime fracture in a brittle manner if
that part contains a large enough flaw or if there is sufficient elastic or plastic constraint.

Fatigue fractures. They result from the application of repeated or cyclic stresses, each
ofthem may be substantially below the normal yield strength of the material. Fatigue
fracture face has two distinct zones. It is comparatively smooth and huge concentric
circles or marks originating from a single nucleus are present. They generally show
slight roughness as the crack grows. The remaining portion is crystalline in nature due
to the sudden fracture.

Non-destructive testing:- These tests include magnetic particle inspection,

ultrasonictesting, liquid penetrant inspection, and radiography. These tests are done
to find out surface and sub-surface defects. The magnetic particle inspection is done
on Ferro-magnetic components, while penetrant tests, ultrasonic tests, and
radiography tests can be done on all the components.

Mechanical testing:-Mechanical tests include hardness tests, UTS, elongation, bend

tests,izod-charpy tests etc. They help ascertain whether the component conforms to
the physical properties mentioned in the drawing. Nick break test is done on non-
ferrous materials to see segregation and oxidation.

Chemical testing:- Drillings of the component are taken to determine its

chemicalcomposition.

Macro examination:- Two types of macro examination are done:

Deep etch test to determine the grain flow and to decide whether the component is
forged, rolled, or cast. It gives indication of inclusion, segregation, rolling seam etc.

Sulphur print: this test gives the indication of Sulphur segregation, and is done by
pressing Silver Bromide paper dipped in 5% sulphuric acid on the polished surface of
transverse cut section
Micro-examination:- This determines the microstructure, inclusion, and mode of
heatreatment given to the component. This also tells about the presence of micro-
cracks, welding, structural changes due to working etc. a small piece of fractured
material is cut, including the region of fracture, and is polished. The final polishing of
longitudinal section is done on the polishing disc. Unetched micro-examination is
done on the polished surface after etching the micro-piece with suitable chemicals
under bench microscope, at magnification from 100 to 1000.
.

Instead of doing visual inspection we can use MI (machine learning) in this to full fill
our requirements .
1. Like for the ananlysis of crack and fractures we can use ultrasonic sensor which
use ultrasonic waves to inspect the material and stored the recived waves and
inspect the by creating a image using image processing analyse that using
previous collected data and reflect us as per our setup.
2. Similarly for handling excessive vibration in power assembly and compressor
We can use vibration sensor which record the vibration and compare that
recorded data with the previous recorded data also in initial there will be a
critical value so if present recorded value is more than that critical value then
our vibration snensor setup will generate a burger sound that will warn the loco
pilot regarding excessive vibration.
3. Similarly like this we can use temperature sensor to inpect the temperature diff
of main bearing and shell bearing both used on crankshaft. So in that case if
there will be a temp diff of 5 degree Celsius or more than that then this will
show on screen arranged in loco cab.
4. Now when the loco completed its journey then these recorded data will be
checked and servicing work will be done as per requirements then thus will be
called predictive maintenance.
LUBE OIL PUMP
The lubricating system in a diesel engine is of vital importance. The lubricating oil
provides a film of soft slippery oil in between two frictional surfaces to reduce
friction and wear. It also serves the following purposes.
 Cooling of bearing, pistons etc.
 Protection of metal surfaces from corrosion, rust, surface damages and wears.
 Keep the components clean and free from carbon, lacquer deposits and
prevent damage due to deposits.

The importance of lube oil system is comparable to the blood circulation system
in the human body. Safety of the engine, its components, and their life span will
largely depend upon the correct quality of oil in correct quantity and pressure to
various location of diesel engine.
The complete engine lubricating oil system is a combination of 04 oil systems.
These are:
(1) Scavenging oil system
(2) Main lubricating oil system
(3) Piston cooling oil system
(4) Soak Back or turbo lube system

1.Scavenging Oil System

The scavenging oil pump is a positive displacement, helical gear type pump. This
pump takes lube oil from 02 sources- from the engine oil sump and from the oil
strainer.
The pump feed lube oil to lube oil filter tank (also called Michiana oil filter). Oil
from the filter tank gose to lube oil cooler where it is cooled by the engine cooling
system. Oil then passes to lube oil strainer where it is filtered once again.
The oil filter (Michiana oil filter) contain 5 paper type filter elements. A bypass
valve provided across the filter tank and set at 40 PSI. If the filter is clogged and
pressure difference reaches to 40 PSI oil is bye passed to lube oil cooler. This
ensures adequate lube oil supply to the engine avoiding damages to the moving
parts.

2.Piston Cooling Oil System

There is a suction pipe (coming from the lube oil strainer) for the piston cooling oil
system and the main lube oil system. The piston cooling oil system pump receives
oil from a common suction pipe and delivers oil to the 2 piston cooling oil
manifolds extending the full length of the engine, one on each bank. A piston
cooling oil pipe at each cylinder directs a stream of oil to cool the underside of the
piston crown. This stream of oil also lubricate the ring belt. Some of this oil enters
oil grooves in the piston pin bearing for lubrication. Oil after cooling and
lubrication drains back in to the oil sump.

3.Main Lubricating Oil System

The main lubricating oil system supplies oil under pressure to most of the moving
parts of the engine. The main lube oil pump takes oil from the strainer housing
through a common suction. Oil from the pump goes to the main oil manifold,
which is located above the crankshaft, extends to the length of the engine.
Maximum oil pressure in the system is control by a relief valve in the passage
between the pump and the main oil manifold. The pressure relief valve is set to
125PSI, which relives excess oil back to the sump.

4.Soak Back Oil System: -

To ensure lubrication of the turbo charger prior to the engine start and the
removal of residual heat from the turbo after engine shutdown, a separate lube
oil pressure source is provided. This pressure system is controlled automatically
by the locomotive control system.
An electrically operated turbo soak back pump draws oil from the oil sump, feed
the oil through a soak back filter and finally to the turbo. A 70-PSI soak back filter
bypass valve is provided inside the soak back filter housing to bypass filter
whenever it cloggs to protect Turbo-charger.

FAILURES IN LUBE OIL PUMP

There are four factors, which effect the lube oil system pressure directly that is
lube oil pump discharge capacity, diesel engine temperature, pressure setting
value of Relief & Regulating valve and quality of lube oil. Some other factors like
choking of filters / strainer, low oil level in c/case, contaminated lube oil, low
idling speed and excessive wear/ clearance in bearings also effect the system
pressure. During running of diesel engine it is observed that lube oil contaminated with
water and oil level in c/case is increasing, which indicates water leakage inside the c/case. The
sources are leakage of cylinder liner bottom gasket & sleeve, cracked cylinder liner,
cracked cylinder head etc. Sometimes it is observed that lube oil contaminated with fuel oil,
which indicates nozzles dribbling or fuel leak off gallery cracked. It is also observed
that some engines consume high rate of lube oil, which indicates clearance between valve
and valve guide is more, engine piston rings worn out or turbo oil seal damaged.

PREVENTION
Maintenance reduction of screw pumps handling Lubricating Oil and Hydraulic Oil
can be achieved by eliminating oil leakage from mechanical shaft seals. Seal leaks
are the major cause of maintenance on oil pumps. Seal conversion kits have been
developed and proven that will eliminate seal leaks, allowing the existing seals to
operate for their design life. Seal less pumps have also been developed and
proven that completely eliminate the seal, and as an added benefit eliminate the
grease packed coupling between pump and motor.

SCOPE OF PREDICTIVE MAINTENANCE

There are few methods or sensor to predict the failure of lube oil pump such as:
 Ultrasonic leak detectors can be used for leaking valves and system leaks.
 Stereoscopic photography can be used for corrosion, fatigue cracking and
hull fouling .
 Infrared thermography can be used for leaky steam traps and hot or cold
firing cylinders.
WATER COOLING SYSTEM

After combustion of fuel in the engine, about 25-30 % of heat produced inside
the cylinder is absorbed by the components surrounding the combustion chamber
like piston, cylinder, cylinder head etc. Unless the heat is taken away from them
and dispersed elsewhere, the components are likely to fail under thermal
stresses. All internal combustion engines are provided with a cooling system
designed to cool the excessively hot components, distribute the heat to the other
surrounding components to maintain uniform temperature throughout the
engine, and finally dissipate the excess heat to atmosphere to keep the engine
temperature within suitable limits. Different cooling systems, like air cooling,
water cooling are adopted, depending on the engine design, working conditions
and service etc.. The advantage of having a water cooling system is that it
maintains a uniform level of temperature throughout the engine and by
controlling the water temperature, the engine temperature can be controlled
effectively.

FAILURE
1. A Bad Seal
The seal between the water pump housing and the engine block keeps coolant
from leaking out of the engine. It also keeps excess air from getting into the
cooling system. This seal can fail if it is improperly installed, if the vehicle isn’t
operated for weeks at a time, or if the coolant becomes corrosive.
2. A Loose Drive Pulley
A loose or wobbling drive pulley is another source of trouble. If the water pump’s
drive pulley is vibrating, the pulley bearings will wear and eventually fail. This
reduces pump efficiency before eventually causing failure.
3. A Bad or Broken Belt
If the belt isn’t sufficiently snug, it should be tightened to specification, and this
tightness should be checked regularly. A failed belt means a non-operational
water pump, which is as catastrophic as a broken water pump.
4. Corrosion
Corrosion only becomes detectable when the water pump requires repair. The
pump will begin to leak, the pump’s impellers will be corroded (thus reducing
efficiency), or the seal between the block and the pump will be damaged and
leaking.

PREVENTION
Most of the time, water pump failure is a result of insufficient maintenance. To
prevent water pump failures, consumers should be educated about the
importance of periodically draining and flushing the coolant in their cooling
system.

SCOPE OF PREDICTIVE MAINTENANCE

There are few methods or sensor to predict the failure of water pump such as:
 Ultrasonic leak detectors can be used for leaking valves and system leaks.
 Stereoscopic photography can be used for corrosion, fatigue cracking and
hull fouling .
 Infrared thermography can be used for leaky steam traps and hot or cold
firing cylinders.
 Shock pulse can be used for checking bearing condition.
 Material(non-destructive) testing can also be used for corrosion,fatigue
cracking and wall thickness reduction
FUEL OIL SYSTEM

The fuel oil system is designed to supply fuel to the engine in correct quantity and at the right
time according to the engine requirements. The fuel oil system draws fuel from fuel tank, filter
the fuel, pressurise the fuel, and inject the fuel into the engine in correct quantity in atomised
condition.
Fuel oil system consist of
 Fuel feed system
 Fuel injection system

LINE DIAGRAM OF THE FUEL OIL SYSTEM

Fuel feed system: -
Fuel is drawn from the fuel oil tank through a suction strainer by the fuel pump. The strainer
separates foreign particles from the fuel oil, and protects the fuel pump. The pump is designed
to supply adequate quantity of fuel to the engine at various speeds and load conditions.Fuel
then goes to primary fuel filter. This primary filter is provided with a 30-PSI bye pass valve with
sight glass, which should be normally empty. Whenever the primary filter is Choked/clogged
and the pressure difference reaches 30 PSI this bye-pass value open allowing the fuel oil
directly to the system, which can be noticed by the flow of bye-pass fuel in the sight glass.
Under such cases the primary filter element is changed.
The fuel then passes to 02 engine mounted secondary filters, which are of spin-on type.
Secondary fuel filters are also provided with a bye-pas value, which is set at 60 PSI. Whenever
the filters are choked/clogged and the pressure difference across the secondary filters reaches
60 PSI, this bye-pass valve opens and diverts the fuel oil back to fuel tank, avoiding damage to
fuel injectors due to unfiltered fuel oil. A bye-pass sight glass is also provided to indicate the
condition of the fuel secondary filters and the sight glass should be normally empty.
From the secondary filters the fuel oil is supplied to all unit injectors through fuel supply
manifolds located inside the top deck on the both banks. The governor controls the quantity of
fuel to be injected through the injectors to the engine.
Fuel injection system :-
Fuel supplied by the fuel feed system is always available at all the unit fuel injectors. The fuel oil
available at each injector are to be pressurized to very high pressure, timed and to be injected
in the cylinder in atomized form. The timing of each unit injector is decided by the camshaft
and the fuel is pressurized by the in-built fuel injection pump which is operated by individual
cam lobe of the cam shaft.
The quantity of fuel to be injected will be regulated and controlled by engine mounted wood
word governor according to the notch and load conditions. The governor operates fuel control
shaft, linkage mechanism and fuel racks. The individual fuel injector nozzle does the
atomization of the fuel to be injected in the cylinder.

FAILURE

 Pressure Problems

On modern automobiles, the average fuel pump pressure is over 60 PSI (pounds per square
inch). On classic cars with mechanical style fuel pumps, however, the pressure is much lower—
between four and six PSI. If you suspect that your fuel pump is not producing enough pressure,
there are two tests you can perform. The first is a simple pressure output test. Many
inexpensive old-school vacuum testers can read mechanical fuel pump pressure just as well as
vacuum pressure.

 Fuel System Leaks

Common place for a fuel leak to develop is in the rubber hose and metal tube that lead from
the tank to the fuel pump. Since the metal tube is exposed to the elements, it's not uncommon
for it to become rusted through to the point that the fuel is leaking out. Likewise, the rubber
hose can also become worn out over time, leading to fuel leaks. If you notice that this
component is damaged, replace it with a reinforced rubber fuel hose.

 Engine Oil Leaks

Where the fuel pump mounts to the timing case cover a gasket provides a tight seal. Although
this gasket is usually reliable, engine vibration can cause the bolts to come loose. When this
happens, oil can seep out around the fuel pump. If you notice a leak in this area, replace the
seal as soon as possible, as detergents in the engine oil can cause further damage.

SCOPE OF PREDICTIVE MAINTENANCE

There are few methods or sensor to predict the failure of water pump such as:
 Ultrasonic leak detectors can be used for leaking valves and system leaks.
 Stereoscopic photography can be used for corrosion, fatigue cracking and hull fouling .
 Infrared thermography can be used for leaky steam traps and hot or cold firing
cylinders.
AIR COMPRESSOR

An air compressor is a device that converts power (using an electric motor, diesel or gasoline engine, etc.)
into potential energy stored in pressurized air (i.e., compressed air). By one of several methods, an air
compressor forces more and more air into a storage tank, increasing the pressure. The compressed air, then,
is held in the tank until called into use. The energy contained in the compressed air can be used for a variety
of applications, utilizing the kinetic energy of the air as it is released and the tank depressurizes.

Reasons for Air Compressor Failure

Air Leaks
As pipes, hoses, drains, seals and other components age, they begin to wear, crack and leak.
Sometimes leaks are audible and sometimes not. It’s important to detect leaks with ultrasonic
equipment.

Exceeding Pressure Limits

Running compressors at the required pressure is key to maintaining them. Running them at
higher pressures increases the workload on the system and can lead to failure. Exceeding
pressure limits also wastes energy and increases energy costs for performing tasks. This was
one of the main reasons of failures observed at TKD Diesel Shed.

Inappropriate Use
An easy way to eliminate failure is by identifying which processes are necessary for compressed
air and which are not. If a task can be handled by low-pressure air instead, as in the case of
cooling, it is better to use it instead of compressed air. This will reduce your energy
consumption and reduce the workload on your compressed air system.

Clogged Air Filters

Air filters get dirty fast, so it’s important to change them as frequently as possible to get the
most out of the compressor. New air filters maximize system performance and help the
compressor run as efficiently as possible. This keeps the compressor from working harder than
it should and keeps energy cost down as well.

COMMON FAILURES AND THEIR PREVENTIVE MAINTENANCE

Problem Cause Solution

Compressor will not 1. No electrical power. Turn on power.

operate. Push the reset button.

2. Low oil level. Check oil level. Replace your oil if necessary

3. Pressure switch not See pressure switch adjustment.

making contact.

4. Pressure in the tank See pressure switch adjustment.

is below the cut-in Replace pressure switch to one that has a
pressure. lower cut-in PSI

Excessive noise in 1. Loose pulley, Tighten any loose ends.

operation. flywheel, belt, belt
guard, cooler, clamps
or accessories.

2. Lack of oil in Check for possible damage to bearings

crankcase. Replenish the oil level.

3. Piston hitting the Remove the compressor cylinder head and

valve plate. inspect for foreign matter on top of the
piston. Add a new gasket and reassemble
Problem Cause Solution

the head.

4. Compressor floor Tighten the bolts on the air compressor. It

mounting loose. may also be a good idea to replace
your vibration pads

5. Defective crankcase. Repair or replace.

6. Excessive crank end Adjust and shim properly.

play.

Knock - same cycle as 1. Main bearings. Replace bearings.

R.P.M.

2. Connecting rod Replace rod.

bearings.

3. Loose flywheel. Tighten.

Knock occurs while 1. Connecting rod Replace rod.

compressor is bearings.
loading.

2. Wrist pins, wrist pin Replace piston assembly.

bearings.

3. Loose connecting Tighten.

rod nut.

Excessive oil 1. Restricted air intake. Clean or replace air filter.

consumption.

2. Oil leaks. Tighten bolts or replace gasket.

3. Worn piston rings. Replace rings.

4. Wrong oil viscosity. Drain oil, refill with oil of proper viscosity.

5. Compressor tilted Level compressor. Vibration pads may help

too much. with this

6. Scored cylinder. Replace cylinder.

Problem Cause Solution

Oil in discharge air. 1. Compressor air Clean or replace your air filters.
intake restricted.

2. Worn piston rings. Replace rings.

3. Excessive oil in Drain down to full mark on sight gauge.

compressor.

4. Wrong oil viscosity. Check viscosity.

5. Piston rings installed Replace crankshaft.

up-side down.

Compressor vibrates. 1. Mounting bolts Tighten.

loose.

2. Compressor not Level compressor so that all feet touch the

properly mounted. floor before tightening down.

3. Pulley and flywheel Realign.

misaligned.

4. Belts loose. Tighten belts.

5. Bent crankshaft. Replace crankshaft.

Air blowing out of 1. Broken first stage Replace valve plate assembly.
inlet. inlet valve.

Insufficient pressure 1. Leaks or restriction. Check for leaks or restriction in hose or

at point of use. piping. Repair.

2. Restricted air intake. Clean or replace air filter element.

3. Slipping belts. Tighten belts.

4. Service hose too Replace with larger hose.

small.

5. Excessive air Limit air usage to compressor capacity by

requirement. using fewer or smaller tools.
Problem Cause Solution

Receiver does not 1. Faulty check valve. Bleed tank! Disassemble check valve
hold pressure when assembly, clean or replace faulty parts.
compressor is
unloaded.

Excessive discharge 1. Dirty cooling Clean cooling surfaces of cylinder,

air temperature. surfaces. intercooler and discharge tube.

2. Poor ventilation. Improve ventilation or relocate

compressor.

3. Blown head gasket. Replace head gasket.

4. Restricted air intake. Clean or replace air filter element.

5. Worn valves. Replace valve plate assembly.

Possible implementation of PM
With today's growing desire for greater productivity, manufacturing processes and equipment
are being subjected to increased demands for additional capacity, testing accepted limitations.
As a result, the diagnostic capabilities of predictive maintenance have also increased, opening
new areas of responsiveness and diagnostic performance. The following represent some of the
most reliable and preferred technologies available for comprehensive analysis.
Lubrication/Oil Analysis
Surveys have shown that over half of all industrial breakdowns are the result of improper
lubrication, including under-lubrication, over-lubrication or use of the wrong lubrication. Oil
analysis, one of the oldest methods of predictive maintenance, is used to define three basic
conditions. These include:

1. Condition of the oil

2. Condition of the lubrication system

3. The condition of the machine itself

Performed and trended over time, oil analysis can pinpoint improperly performed
maintenance, operational practices, and incidents of contamination such as the addition of an
improper lubricant.
Machine Condition Monitoring

Sound analysis, one of the oldest and widely employed systems of detecting equipment failure,
is best performed on a day-to-day basis by someone familiar with the equipment. Sound
analysis is simply detecting something abnormal; if the sound becomes audible, it generally
indicates a component is beginning to malfunction and should be further inspected.
Vibration analysis is often used to determine the operating condition of rotating equipment by
identifying potential problems before they cause serious failures and unscheduled downtime.
This type of analysis can be used to detect deteriorating or defective bearings, mechanical
looseness, and worn or broken gears. It can also pinpoint misalignment or imbalance before it
can cause bearing or shaft deterioration.
Ultrasonic analysis enables technicians to identify and locate such potential problems as
bearing deterioration and leaks of compressed air or hydraulic fluid. This type of analysis can
also identify poor maintenance practices and machines that are operating beyond their original
design limitations.
Thermo graphic analysis provides visual images that represent variations in IR radiance of
surface objects. This type of analysis has been proven as an effective predictive maintenance
tool because mechanical or electrical breakdowns are often preceded by changes in operating
temperatures.
Infrared Analysis has the ability to detect an object's IR emissions, and translate into a visual
image. It does not have the capability to analyze and quantify specific temperature values.

Use of vibration sensors

Critical machine parts like separators, cooling towers, blowers, generators, pumps, bearings and motors
usually give hints they are damaged and about to fail. When a fan becomes unbalanced or bearings start
failing, they start vibrating more. Although this change wouldn't be noticeable to the human eye or ear,
it's easily detected by a vibration sensor .
Parts ideal for vibration monitoring:

 Separators
 Rolling stand bearings
 Motors driving hydraulic pumps

Vibration sensors serve as an early indicator of product damage or failure. By sending an alert as soon as they
detect abnormal activity, they help you solve problems before they turn into serious issues.
This screenshot shows a real time- view of a vibration sensor through the Sensaphone
Sentinel web interface.

VALVE SPRING

The valve spring is one of the most important yet over looked parts of an engine, and is a key
part of the valve train. The valve spring keeps the valves closed tightly against their seats until
the cam opens the valve, which releases pressure. The cam will then continue through its
motion releasing the valve and allowing the valve spring to do its job, and retract the valve back
into the head.
The spring was designed for 1500 rpm diesel engine and now used in 2200 rpm diesel
engine. The failure occurs with 2200 rpm engine, with similar pattern ie. at 1.5 to 2
turns. There may be one or more reasons for exhaust valve spring failure. Some of
these reasons are due to improper metallurgical properties of spring, improper design
etc. The effects of this failure results in deterioration of engine performance (brake
power & efficiency), failure of other components eg. cylinder head, valve guide, wear
in between valve and cylinder head.

FAILURES IN VALVE SPRING

1. Failure due to fatigue

In engineering terminology fatigue is a progressive structural damage of materials under cyclic
loads. Important categories of fatigue include: Mechanical fatigue due to fluctuating stresses Creep
fatigue due to cyclic loads at high temperatures; Thermal fatigue due to cyclic changes in material’s
temperature; Thermo-mechanical fatigue due to a combination of mechanical and thermal fatigue;
corrosion fatigue due to cyclic loads applied on corroded materials, Fretting fatigue due to cyclic
stresses together with the oscillation motion and frictional sliding between surfaces, etc.. Fatigue
failure occurs at stresses that are well below the yield point of the material
I.C. Engine valves are subjected to repeated cyclic loading due to valve train dynamics. Repeated
loading results in materials failing well below the yield strength. When the material is subjected to
fatigue, one or more tiny cracks usually start developing in the material, and these grow until
complete failure occurs.
2. Failure of valve due to wear
Wear Failure occurs generally at the seating face of valves and stem at which portion that slide on
valve stem guide. Wear mainly occurs due to two major factors, he first one is the impact force
between seating face of valve and seat insert, whereas the second one is due to sliding of the valve
on the seat insert during the action of combustion pressure. The rubbing action and impacting
action wear out the valve face. Typical mechanisms of wear include adhesive, abrasive, fretting,
erosion, cavitation, and contact rolling fatigue.

Wear is divided into three stages: running - in wear stage, stable wear stage, and sharp wear stage.
In running - in wear stage, the temperature is high. Under the action of high impact varying load,
the strength and hardness of surface material of gas - valve is decreased because of high
temperature, the plastic deformation initiates so the seat face of valves are worn out and the wear
mass loss sharply speed - up with the passage of time. With the increment of expulsion quantity,
surface roughness drops and practical contact area of gas - valve increases. Rmax data shows the
wear at seat face of valve which is the difference between maximum and minimum values of
measurement length.

Predictive Maintenance of valve spring failure

SVM (Smart Valve Monitoring) system.

The system has been designed to deliver predictive maintenance information from widely
distributed areas for use by operators to help reduce shutdowns and improve overall plant
efficiency.
This is made possible by the ability of the SVM to be integrated into an existing Ethernet and
fibre optic infrastructure, enabling over 100 monitored valve actuators on wellheads at
numerous locations to be networked over distances of 10 km or more.
The project sees SVM field control units attached to the ESD circuits for shut off valves, installed
in sets of two and four on more than 50 wellheads. Analysis is performed on the SVM server
computer in the centralized control room.
Partial stroke testing (PST) is a function used in a safety instrumented system (SIS) to enable
operators to identify possible failure modes on a shutdown or emergency shutdown (ESD) valve
without the need to completely close the valve which would disrupt the process. Partial stroke
testing is an accepted hydrocarbon industry standard technique that is quantified in detail by
regulatory bodies such as the IEC and ISA.

The partial valve stroke prevents unexpected failure on demand of the safety function and
demonstrates that certain potential problems that would otherwise go undetected, such as
spring fractures in the spring chamber of the pneumatic actuator, are not present.
Consequently, the interval for testing for these otherwise undetected errors can be extended.

The SVM is said to incorporate several features that are not available from other systems, as
well as providing detailed diagnostic data that allows the operator to plan for strategic
preventative maintenance. The key to its reliable performance is its separation from the valve’s
control system which enables operators to design the control system to suit the routine and
safety requirements without having to compromise for the testing programme.

The SVM system is powered by the control signal to the actuator’s solenoid valve. The
monitoring function is then provided by a pressure transmitter located between the solenoid
valve and actuator, which records the instrument pressure changes while the valve is moving.
Any change in the valve performance is detected and identified by a change in the pressure
wave exiting the actuator. The self-contained design of the SVM enables it to be used with the
most complex control mechanisms and makes it impossible for the SVM to prevent the valve
from closing on demand.
Design optimization of Valve Spring
Optical microscopy

The optical microscopy of the spring sample revealed a normal structure (containing ferrite
and pearlite phases) typical of high strength spring steel.

Hardness test
The hardness test is done on Rockwell hardness tester, the hardness is found to be 41.5 HRC
which is according to the drawing specification ie between 38 HRC- 48 HRC

Die-penetration test

The Die-penetration test is done to see the minor cracks or flaws on the surface. In this test
spring surface is first cleaned chemically, after 5-7 min fluorescent color is sprayed on the
surface, then for 10-15 min the surface is kept for drying and then after developer is applied.
Now clean the spring with tissue paper and the micro cracks or flows are seen to be glowing
under ultraviolet light. The surface is observed generally smooth.
Analysis using ANSYS

The Geometry was modeled in modeling software Solidworks 2017 . Then the model
was imported into ANSYS workbench 18.1 . The meshing was done with tetrahedral
elements. To increase the accuracy and bring the analysis to practical condition the
mapped meshing was done. Image shows ANSYS result in which the equivalent stress
distribution are shown. The maximum stress is concentrated at the point where the
failure occurs. The maximum shear stress is seen at the point where the failure occurs.
 Possible Changes in existing
Preventive Maintenance
techniques

While modern technology is advancing day by day - now multiple sensors can
stream data continuously, data processing has become faster and sophisticated,
there are new data capture devices more , certainly the implementation of
predictive maintenance would reduce operating costs, extend a loco’s lifetime
and potentially deliver huge financial rewards. However, switching to PM is a long
way to go.

So while the plans to switch to PM are laid out, here are some suggestions to
improve the existing maintenance techniques : -

Back at the shed, it was observed that failures occurred mainly due to 2 reasons :

1. Malfunctions in the engine and material used in the engine

2. Poor maintenance & negligence

Malfunctions in the engine and material used in the engine are unavoidable. Any
component has a certain life up to which it performs well, and after which it is
very likely to fail. These failures arise due to metallurgical phenomenon like creep,
fatigue etc. or due to overloading, failure of other engine components etc.

The loss of time and money due to these failures can be prevented only by
predictive maintenance, which would help in predicting such failures long time
before they even occur.

Poor maintenance & negligence however, if controlled, we can control the

failures occurring to some certain extent.
At the shed it was seen that :

 There is a lack of working staff : When 12 people are supposed to be

identifying errors/defects on one locomotives, there were only 5-6 people
working on one. And even then, one person has to look after 3-4
locomotives simultaneously. This reduces the effectiveness of maintenance
by a large extent.

Increasing the number of working staff will increase the effectiveness of

maintenance (proper installation of components, identifying issues etc.) .

 There is a lack of constraints on the working conditions of components : it

was seen that various components on the locomotive were operated above
their safe working limit. Engines and compressors were operated above
their safe RPMS (in compressor’s case, above 8 m^3/min) which caused
overheating of the piston and other components. This overheating caused
seizure, which caused the connecting rod, bearing and sometimes even
crankshaft to fracture.

A proper set of working conditions must be imposed strictly and further

supervision must be employed to avoid negligence.
 Conclusion
In this project we analyzed various failures and devised some methods to predict
component failures based on working conditions. The future looks promising, that
such a type of prediction indeed is possible. However, the effort to reach this goal
is high, it also means that older trains currently in service, that were not built for
modern connectivity, require investment to be able to do so.
The benefits are there, but they need proper planning and management. Various
studies would have to be done as to which data should be acquired. Installing the
required apparatus would have high initial costs, and possibly changes to the
engine design. Then after installation of sensors / DAQ apparatus is done, a
machine learning / AI Algorithm would be needed to be developed in order to
process the data and predict failures based on past failures and unusual running
conditions. Also various changes in schedule would have to be implemented, a lot
of testing would be needed.

Therefore, there is an important cost-benefit case which must be made ahead of

any major investment into predictive maintenance. But it doesn’t mean
predictive maintenance is out of reach. Once the initial setup is done, algorithm is
developed, all the investment would be worth it.