By Eduardo Niño de Rivera
Special Edition: John Amendola Sr
Download pdf version.
Download pdf version.
INTRUDUCTION:
There is a wide
consensus described in technical literature [see references] regarding the best
way to retrieve oil samples from lubricating systems. It is often stated that oil
test results will only be valid if they are performed on samples that are
representative of the oil in the system. A nonrepresentative sample may lead to
erroneous conclusions regarding the condition of the oil itself and of the mechanical
components it lubricates. If, for instance, the sample is more contaminated
than the rest of the oil, we may conclude that the gears and bearings are in
worse condition than they really are. Likewise, a sample that is cleaner than
the rest of the oil, may lead us to believe that there is nothing wrong with
components that may be approaching an imminent failure condition. In addition,
the raw data of the analysis is meaningless without a proper interpretation
Whether oil samples
are taken by plant personnel or by outside vendors, a technician with adequate
training may extract the sample or supervise how it is done. An oil analysis is
not a direct observation of the condition of the gears or bearings, it requires
corroboration and proper interpretation to be a useful diagnostic tool. This
article offers practical advice on how to properly take representative samples
and on how to interpret test results.
THE FUNCTIONS OF
LUBRICATING OIL.
Lubricating oil has
three basic functions:
·
Reduce
sliding and rolling friction in bearings, gears and seals;
·
Remove
heat generated at sliding and rolling surfaces; and,
·
Protect
surfaces from rusting and corroding.
Lubricating oil
reduces friction by forming a sufficiently thick film between sliding surfaces
to prevent them for making direct contact with each other. This requires the
proper combination of load, sliding speed and oil viscosity. If the oil film is
not thick enough, there will be direct metal to metal contact, leading to
micropitting and/or scuffing, where metal particles are removed from the
contacting surfaces. Although these particles will eventually sink to the
bottom of the oil sump or they will be trapped by oil filters or magnetic plugs,
they will remain in suspension within the oil for some time, contributing to
abrasive ware or indentations that further deteriorate load bearing surfaces.
Lower friction in
properly lubricated rolling and sliding surfaces contributes to keep the
operating temperature at acceptable levels by removing heat as it flows away
from the surfaces where it is generated.
The 3 regimes of
lubrication, represented by the gray (sliding surfaces) and yellow (lubricant)
images, are defined as follows:
Regime I: Boundary
Lubrication. There is significant contact between the asperities of the sliding
surfaces
Regime II: Mixed-Film
Lubrication. There is contact only in the highest peaks of the asperities
Regime III:
Hydrodynamic Lubrication: The oil film is sufficiently thick to completely
separate the sliding surfaces.
It is suggested but
not proven, that surface distress due to scuffing occurs in the borderline between regime I & II, it is an
instantaneous phenomenon. Surface distress resulting in micropitting occurs in
the borderline between regime II & III. Micropitting is a fatigue phenomenon and
requires some load cycling.
When oil deteriorates
due to time in service, contamination, exposure to high temperatures or adverse
environmental conditions, it loses the ability to protect the metal surfaces
against corrosion and abrasion thereby damaging the surfaces it was meant to
protect.
MONITORING THE
CONDITION OF LUBRICATING OIL.
The first step in
designing a condition monitoring program for lubricating oil is to reach a
clear understanding with a certified lab regarding how frequently the periodic
oil tests should be performed; the reported formats and contents, including
diagnostics and other information; referenced acceptable level for each
variable in the report; and the number, location, identification, size, bottle
specifications and gathering methods for the samples.
The Acceptable values
for each variable should be agreed upon by the plant’s analysts and the lab
based on the requirements of each machine.
Periodic oil tests
results and their evaluation over time, should provide the following
information [1]:
·
The
current condition of the oil;
·
How
long the oil lasts begore it needs to be changed;
·
The
presence of metallic particles as an indicator of bearing and gear wear; and
·
The
trends in the frequency of oil changes and in quantity and composition of
metallic particles and other contaminants mixed within the oil.
Oil tests performed
outside the preestablished schedule may be required to assess the condition of
gears and bearings when anomalous symptoms appear, such as an increase in
temperature, noise or vibration.
A well-designed routine
test program is a cost-effective way to determine whether oil characteristics
and contamination levels are within acceptable limits [1]:
·
Oil
viscosity at 40°C (100°C is also useful as representative of the operating
condition);
·
Cleanliness;
·
Water
Content;
·
Acid
Number; and
·
Metal
and other substances mixed within the oil.
If these measurements
are within acceptable limits and there is no other indication of deterioration
or damage to any components, the gearbox may continue in operation.
A single reading
outside the acceptable values does not necessarily indicate that there is a
problem with the gearbox, the problem may well be in the oil sample itself, so
before making any transcendental or costly decisions, it is convenient to take
a step by step approach to assess the condition of the components and to
determine the underlying cause of deterioration and eventual failure:
·
Compare
the unacceptable reading with other samples taken at the same time;
·
Submit
a new sample to double check the results;
·
Review
data provided by the condition monitoring system (temperature, noise and
vibrations);
vibrations);
·
Perform
more detailed oil test;
·
Perform
a direct inspection of the gears and bearings; and
·
Perform
a detailed cause of failure analysis on gears and bearings.
Even when all the oil
test variables are within acceptable values, it is important to keep track of
their evolution over time. An upward trend in one or more variables may indicate
that there is a deteriorating condition which will go undetected if current
reports are not compared to past records. Spotting upward trends may allow us
to take corrective action while the gearbox is still in operable condition,
preventing catastrophic failures or unplanned machine downtime.
This cost-effective approach
provides a solid base for an accurate diagnosis and sensible decisions.
WHERE TO TAKE SAMPLES
The oil within the
system is not a homogeneous mix, so samples taken from different locations
provide different pieces of information that may be put together to provide a
more comprehensive view. If possible, AVOID TAKING SAMPLES FROM areas that are
clearly not representative of the oil that reaches the gears and bearings, such
as:
·
The
bottom of the sump, where debris settles in much greater concentration than in
the rest of the oil;
·
Downstream
from filters that remove contaminants and wear debris generated at load bearing
surfaces; or
·
Right
after the oil has been changed.
Ideally, samples from sumps
or reservoirs should be taken at approximately the middle of the oil depth, at
least two inches away from the casing walls and from the immersed rotating
components.
In circulating oil systems,
samples may also be taken from the return lines, this should always be before
the oil goes through filters. Samples taken from turbulent oil flow at piping
bends are more representative of the general oil condition than those taken
from the walls of straight tubing with laminar flow.
Be consistent. When
taking an oil sample, extract it from the same relative location. Also do so at
the same time; i.e 30 minutes after shutdown. If the system allows access while
running, take the sample after the unit has reached a steady state operating
condition. This can be verified by checking when the gear unit operating
temperature has stabilized.
HOW TO TAKE SAMPLES
Samples must be free
from outside contamination:
·
Use
new or clean, perfectly dry bottles, tubing and other tools that may be in
contact with the oil;
·
Clean
ports, valves and plugs before connecting or opening;
·
If
possible, avoid using the same tubes and hoses to retrieve different samples;
·
Flush
and clean any tools or equipment before reusing them to collect a new sample;
·
Flush
stagnant oil from pipes or tubing before collecting samples; and
·
Properly
identify each sample taken (equipment, sampling point, time and date).
Use bottles or containers
that meet the specifications agreed upon with the lab (material, shape and
size).
CONSISTENCY OVER TIME
Permanently installed sample
ports and valves (A) provide a very consistent means of retrieving samples from
the same spot every time. These ports and valves can be conveniently located in
return lines or in the gearbox casing to take representative samples as
discussed above. If there are no ports
conveniently located in the casing, a bent pitot tube may solve the problem or new
ports may need to be adapted on the gearbox.
|
A
|
|
B
|
|
C
|
|
D
|
|
E
|
In circulating oil
systems, samples may be taken from return lines, preferably at pipeline bends,
where the flow is turbulent (B). Avoid taking samples through side walls of
straight-line laminar flow (D) and do not take samples of oil that has gone
through a filter (E). Low pressure lines
will require vacuum pumps to extract oil samples.
Samples from sumps may
also be taken with vacuum pumps, introducing a rigid or plastic tube through vent
or filling plug openings (C). From outside the casing, it is difficult to
control de exact location at the end of a flexible tube, so particular care
must be taken to avoid taking samples from the bottom or near the walls of the
casing. For better control and consistency, the flexible tube may be attached
to a more rigid dip stick, this will make it easier to place the suction at
approximately to the same spot every time.
|
B
|
|
D
|
|
E
|
|
C
|
|
C
|
|
D
|
|
E
|
SUMMARY:
THREE KEYS TO USEFULL
OIL ANALYSIS
1.- SAMPLES. The data
returned by the oil tests is only as good as the samples submitted:
·
The
samples must be representative of the oil in the system;
·
Samples
must not be contaminated during the gathering process;
·
Samples
must be taken consistently over time;
2.- LAB REPORTS.
·
The
tests should be performed by a certified lab;
·
The
reports must return relevant data; and
·
The
acceptable limits must be properly set for each test variable according to the
requirements of each machine.
3.- INTERPRETATION.
·
The
gearbox may continue in normal operation if all measurements are reported
within acceptable limits and there is no other indication of an anomalous
condition;
·
A
single reading outside the limits should be confirmed before further action is
taken;
·
Oil
analysis is a useful indirect observation that points us in the right direction
for immediate action but confirmation from other monitored variables, such as
temperature, noise and vibration, or from direct observation of the components
should be considered before committing to transcendental or costly decisions;
and
·
Upward
trends in reported values over time are early indicators of deteriorating
mechanical components, costly repairs and unexpected down time may be prevented
by addressing the underlying issues while the gearbox is still in operable
condition.
CONCLUSION:
Oil analysis is
important not only because it lets us know if the oil is preforming properly as
a lubricant, but also because the oil captures liquid and solid matter that is
either penetrating from the outside or is being generated within the gearbox,
reflecting the degree of deterioration of seals, bearings, gears and even the
gearbox casing. Further, discrepancies and trends in the reported data may provide
advanced warning of a deteriorating condition in mechanical components. Oil
analysis is, therefore, a good indicator of health of the gearbox, however, it
is not a direct observation of the actual condition of its components; it is
useful as a non-invasive, cost-effective observation that can either sustain
confidence to continue operating when the measurements are reported within
acceptable limits and there is no other indicator of distress within the
gearbox, or it can provide early warnings of anomalous conditions, yet, the findings
must be confirmed by other condition monitoring methods or by direct
observation of the internal gearbox components before transcendental or costly
actions are taken.
References
