Bill Freeman wrote:
Hi All,
A few comments on bearings that might be useful to the group. There are two
general classes of bearings, with a lot of variation in each class. The
first class
is plain or journal bearings, the second class is rolling element bearings,
like ball
or tapered roller bearing assemblies.
The journal bearing can be lubed with grease or oil and works adequately for
moderate loads and speeds when greased or oiled occasionally. In this mode,
there is contact or near contact between the shaft and the bearing shell and
there
is steady wear. This would be like a bronze bushing in an inexpensive
electric motor.
The really fantastic version of the journal bearing uses a pressure fed
supply of oil
and runs with literally zero wear when running. Let me say that again - a
pressure
fed, oil lubricated journal bearing can last essentially forever as long as
fed a steady
supply of clean, cool lube oil at an adequate pressure. The journal
actually pulls the
oil into the gap between the shell and the journal, and since the journal is
smaller than
the shell and inevitably offset, there is a wedge of oil formed which
increases the
pressure and lifts the journal off of the bearing. No contact, no wear --
one exception,
startup. Until the oil wedge forms the bearing operates like the bushing,
and has
significant wear. So journal bearings work best for higher speeds with a
pressurized
oil system, and operating at significant speeds and especially for long time
periods. Start
stop cycles are bad.
Rolling element bearings are great for low speed operation, and can operate
a very
long time with just fillup of clean grease, with seals to keep the grease in
and dirt out.
At higher speeds, the centrifugal loads on the rolling elements can get
significant, adding
to the load on the rolling elements and the outer race. This is not to say
that rolling
element bearings don't work at high speeds, just that they work less well as
the speeds
go up. The bearings will work with limited lube if kept very clean and
cooled, such as in
the crankcase of a two stroke motor where the lube is a mist of gasoline-air
mix with a little
bit of oil mixed in.The big disadvantage for rolling element bearings is
that the alway fail
in a very destructive mode. The contact area between the rolling element
and the races
is very small, eithera line or a point. Due to the nature of the resultant
stress field, the
maximum stress is NOT at the surface of the race or rolling element, but a
short distance
*below* the surface. This stress **will** eventually cause a subsurface
crack, which
will eventually grow and cause chunks of metal to pop (or spall) out. These
chunks
cause a lot of extra damage to both the bearing and anything else that they
can get to.
If you have a rolling element bearing in your "tranny" (PSRU) that shares
the oil system
with your engine, you need to give serious thought to detecting and
isolating bearing spall
material from the engine. A magnetic chip detector in the return flow path
would be a
great idea. Once a bearing starts spitting flakes, there is nothing to do
except replace it
and all the others that are running on the same load/time schedule. They
WILL be failing
soon after since this is a fatigue process which is dependent on load and
number of cycles.
By the way, a cycle is a stress reversal, usually a revolution. Detection
of the beginnings
of failure is VERY important. A non-inspectable sealed rolling element
bearing in a critical
location is an accident waiting to happen, IMHO.
Hope this helps form a useful foundation for selection of bearings for
different applications.
Bill Freeman
The AirCraft Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://home.earthlink.net/~rotaryeng/ http://www.linux.org