|
Oil Analysis Without Particle Counting?
Van Richard, Sr. Reliability Engineer, Georgia
Gulf
The first thing to do is evaluate whether the Lab manager might
be right, and the particle counting is not really needed. If the
lubrication program was advanced enough, perhaps particle counting
could be eliminated.
To be advanced enough, the program would need to be pretty far
along the path of best practices, i.e., new oil is filtered while
transferring; very good application and use of filtration; purification
skid is used regularly on large reservoirs; lubricant reservoirs
are well sealed and well maintained; and other analyses would
still be performed (viscosity, wear metals, contaminant levels,
water, TAN, etc.).
Dr. W. Edwards Deming, a driving force in the quality movement
and use of statistical analysis of data, also taught that the
best thing a company could do, as far as process control, would
be to revise their system or process so that no measurements needed
to be taken, because there was no longer any chance for the process
to turn out a bad result.
Simply, if the process is no longer capable of making a bad product,
don't waste time by continuing to measure for it. However, a power
plant, in addition to pumps and turbines, is almost certain to
have servo valves controlling fuel and/or steam valves. Hydraulic
servos have much tighter clearances than most any other lubricated
equipment.
Micron-sized particles, that would pose no threat to lubricated
bearings, can easily cause servos to wear, or worse, stick, and
can result in reduced capacity or a shut down for maintenance
on the servos. This generally makes particle count a vital parameter
for measurement. I would doubt that a program is advanced enough
that particle counting is not needed at all for a power generation
facility.
A likely compromise would be to reduce the number of equipment
that the test is performed on, or look for another lab, or possibly
a lubricant supplier, that would perform the test at a lower cost.
Some lubricant manufacturers may be persuaded to perform a specific
test on a few oil samples for a low price. If the particle counting
must absolutely be eliminated, samples could be looked at under
a microscope, a patch test can be done, or oil could be changed
more frequently.
See other responses
to this Readers Challenge.
|
