This job aid was extracted from the Online University.
INTRODUCTION
When we’re trying to determine the cause of a failure, we
have to treat the product and the process just like any good
detective would. We have to have a plan. We have to be
systematic. And we have to be careful that we do not destroy
any delicate evidence.
In this job aid we’re going to look at a plan that will help you
get to the right answer quicker, and we’re going to look at a
couple of common situations.
• First we ask questions so that we can understand the
situation.
• Then we make observations—things that you can and
should do with the customer standing right there.
• Third, we examine the evidence very carefully.
JOB AID
FAILURE ANALYSIS
• And finally, we develop conclusions.
We do these four steps. And we do them in this order, every
time.
ASK QUESTIONS
We start with the questions. When a customer says it’s not
working, and maybe even gives you his opinion about what
happened, we need to get all the facts.
• How was the engine being used just before it failed?
• Were there any noises at the time of the failure?
• Did you attempt to restart the engine? If you did, what happened?
• And — when was it last serviced?
You probably ask most of these questions when you’re writing up a repair order.
But now you might want more detail.
This is where the detective work really begins. You start to
look for clues. These are all quick checks that you should do
with the customer right there—and you don’t need anything
more than a spark plug wrench.
First, we examine the outside of the engine.
• Is the cooling air inlet obstructed?
• Are the cooling fins clogged?
• Is there a hole in the crankcase?
• Are there signs of oil leaks?
This is all very quick and can be done in about the time it
takes to talk about it.
Next, check the oil.
• How much oil is there?
• What does it look like—fresh and clean, or old and dirty.
What you’re looking for may not be as obvious as you might
think. Certainly you need to see if the oil level on the stick is
within normal operating range. All engines consume some oil
and air cooled engines typically use more than water cooled
engines.
There’s always the possibility that the owner never checked
or added oil to the engine. You will have to drain the oil to
fully understand its condition.
If the oil is black and a lot thicker than fresh oil, you may be
dealing with an engine that’s never really had an oil change.
The owner added some, but never drained the old oil. Or you
may be dealing with an engine that has been badly
overheated.
If you see sparkling metal particles in the oil, save them.
They may be a clue to the failure. You’ll want to check to see
if they are attracted by a magnet.
Fresh, clean oil may not be entirely good news. It may have
been added after the failure. Too late to do any good. If it
was added after the failure, it will float on top of the old,
burned oil.
If you see fresh, clean oil on the dipstick, make a mental note
to watch the color of the oil carefully when you drain the
engine.
Next, take a close look at the air filter.
• Are the elements clogged or damaged?
• Do you see any dirt trails around the seals and down the
carburetor throat indicating that dirt was bypassing the
filter?
• Is the oil breather connected to the air cleaner?
A “yes” answer to any of these is an important clue. If dirt
gets into the engine through damaged filter elements, a
damaged or incorrectly installed seal, or a disconnected oil
breather, it’s like running sandpaper on the engine parts.
Abrasive damage will occur to the cylinder, piston, and
bearing surfaces.
Dirt can be a problem even if it doesn’t get into the engine. If
it clogs the air cleaner elements, the fuel mixture becomes
too rich and can foul the spark plug, wash the oil from the
cylinder walls, and dilute the oil. Sometimes it will dilute the
oil so much that the engine seizes from lubrication failure.
Next, check out the carburetor. You want to make sure that
it’s secure to the manifold. See if the governor linkage shows
signs of damage or tampering. Is the carburetor throat
clean?
If the carburetor is loose, it may offer another path for dirt to
enter the engine. At the very least it will allow additional
unfiltered air into the engine, making the mixture leaner.
Since a too-lean mixture raises the temperature of
combustion, it can cause overheating, detonation, and
subsequent failure.
If the governor linkage is out of spec, either because its been
damaged or because someone has tampered with it, you can
suspect overspeed damage. And there’s another problem to
check for: black oil deposits indicate excessive blowby
through the breather.
The last step in your quick check is to examine the spark
plug.
• Is it fouled with oil or soot?
• Are there signs of overheating or detonation?
• Is there mechanical damage?
Just like the eyes are the window of the soul, the spark plug
Now it’s time to get inside the engine. Before you even turn a
bolt, remind yourself to be careful so that you don’t destroy
any evidence.
Normally, when you’re tearing down an engine, you take the
parts straight to the parts washer. If you do that, valuable
information literally goes down the drain. Try to keep
everything just as it was when the engine failed.
Now examine the oil and engine very carefully.
First, we drain the oil into a clean container. Look at the oil as
it drains; it can tell you a lot.
If it starts our black and thick and then turns cleaner, it’s likely
that fresh oil was added after the engine failed.
Often it’s a good idea to strain the oil through some sort of
filter; you can use anything from cheesecloth to a clean shop
towel.
Look for things that shouldn’t be there;
• small metal flakes
• gasoline smell
• condensation in the oil
Any of these are clues to the probable cause of engine
failure.
Once you’ve drained and examined the oil, start
disassembling the engine. Examine each piece as you
remove it.
You’re looking for things like broken pieces in the crankcase,
cylinder, and rocker box when you open them.
Look at the oil on the parts for signs of coking or overheating.
You’ll usually need to clean the oil from the components for a
closer look. Do it carefully; use clean solvent and a soft
paintbrush to remove the oil—not a rag or stiff parts brush.
Remember, what you’re trying to do is remove the oil
and not the evidence.
There are only a few causes for failure of a Honda engine.
The most common are oil starvation and dirt ingestion or
both.
For instance, dirt ingestion typically leads to high oil
consumption which, in turn, can lead to oil starvation and
engine failure.
Dirt ingestion is usually very obvious—think of constantly
blasting an abrasive on all of the wear surfaces of the
engine. Instead of the polished, burnished finish, engine
components will have a satin finish.
This is what it looks like when the engine fails due to dirt
ingestion. Dirt makes an engine old very fast. Look for wear
on the cylinder walls, valve stems, and bearings.
You might also see dirt tracks down the carburetor and
through the intake. Keep in mind that, although the
carburetor is the usual entry point, the dirt may have come in
through the crankcase breather.
The first phase of oil starvation is aluminum rod material
transferring to the crank. Rod bearing material may seize to
the crank and spin in the rod.
Here is a crankshaft that failed due to oil starvation. You can
find clear signs of overheating.
The crankshaft journal should be a silver color with a sheen
of oil. This blue color means that the journal got hot enough
to take the temper out. The only way you get that kind of heat
is when there is no lubrication.
Here’s another clue. Aluminum from the connecting rod has
actually welded itself to the crank pin.
After enough metal transfer or removal, the rod will pound on
the crank journal, and the rod bolts will lose torque and begin
to loosen.
Eventually the rod fails, sometimes right through the block as
shown.
Valve faces will become dished from wearing on the
hardened seat. Ring end-gaps will increase, increasing oil
consumption and blowby. You can see the oil deposits in the
carburetor if the breather hose is still connected.
This is what happens when a good part goes bad. The idea is
to investigate thoroughly and determine the problem so you
can fix it properly.
It’s not always as easy as it seems, nor is the obvious answer
always the right one. Remember the process.
A customer comes in and says his engine failed due to a
loose rod bolt. The customer says that the engine has 200
hours on it.
Here is what an actual loose bolt failure looks like.
This is a GC160 engine with the rod bolts intentionally left
finger tight. The engine was run for an hour and then cooled
for 30 minutes.
The engine was started again and ran for 11 minutes before
self destructing.
You can see that there has been a parts explosion, but look
at the journals. They are bright and shiny with no
discoloration or metal transfer. Just what you’d expect from
an engine that had plenty of clean oil.
You don’t get rod-bolt failure on an engine that’s been
running fine for hundreds of hours.
Then why are the rod bolts loose?
Simple. Without oil, the rotating parts generate enough heat
to take the temper out of a crankshaft and melt aluminum
from the rod onto the crankshaft. That heat will expand the
aluminum rod more than the steel bolt.
The aluminum compresses before the bolt stretches and the
result is no more tension on the bolt threads.
Yes, you have a loose rod bolt, but it didn’t cause the
problem, it was caused by the problem.