mechanicsupport.com

unk-unk applied to Aircraft Inspections

2013-03-05T08:09:00.003-08:00

"There are known knowns; there are things we know we know. We also know there are known unknowns; that is to say, we know there are some things we do not know. But there are also unknown unknowns – the ones we don’t know we don’t know." Donald Rumsfield, Secetary of Defense

Inspecting an airplane using best industry practices, applied by consensious and trained mechanics, with help and guidance by our regulator authorities, one is still left with rumsfeldian unknowns; Something still might be wrong.

When something does go wrong, often the mechanic is blamed because "he should have caught it". Before this claim can be made one has to determine what the mechanic knew and did not know and why. It could be:

He might have known but ignored the condition (known-known).
He might have known that a condition could or did exist (corrosion for example) but its severity or signifigance was unknown. (known unknowns)
He might simply have not known about the condition and therefore it never occured to him that there could be a problem. Example, might be an undisclosed manufacturing defect. ;This is the unk-unk (unknown-unknown). "I never heard of that problem."

The proper response depends on which of these three catagories the problem belongs to. If the problem falls into the unk-unk then the problem is not a mechanic issue.

unk-unk is a slang engineering term thought to have originated at Lockheed. The problem of "we don't know what we don't know" is a trap anyone doing maintenance can fall into, not grasping the significance (or danger) of an action or inaction. The best defence against unk-unk problems is to follow well established procedures and standards even though "It's always more fun to go off rapidly on your own and invent your very own personal mistakes rather than look up and actually study somebody else's stuffy reference book."

"if it flew in it will fly out" -- Feynman's Criteria

2013-02-09T08:39:00.000-08:00

If one is asked to sign-off on safety, based on "not having failed yet," in the presence of deviations in performance, then I suggest you read Feynman's Appendix to the Rogers Commission Report on the Space Shuttle Challenger Accident . Feynman^1. outlines the classic tension between managers and technicians (engineers or mechanics).

This is what Feynman had to say in regards to the success of flights in the presence of deviations:

"The acceptance and success of these flights is taken as evidence of safety. But erosion and blow-by are not what the design expected. They are warnings that something is wrong. The equipment is not operating as expected, and therefore there is a danger that it can operate with even wider deviations in this unexpected and not thoroughly understood way. The fact that this danger did not lead to a catastrophe before is no guarantee that it will not the next time, unless it is completely understood. When playing Russian roulette the fact that the first shot got off safely is little comfort for the next."

Metal particles from oil filter

Metal particles in the oil filter is evidence that the equipment is not operating as expected. Given the finding of metal shavings in this aircraft engine oil filter, using Feynman's criteria, should this aircraft be removed from service until a more through understanding of the problem is identified?

1. Richard Phillips Feynman was an American theoretical physicist known for his work in the path integral formulation of quantum mechanics, the theory of quantum electrodynamics, and the physics of the superfluidity of supercooled liquid helium, as well as in particle physics (he proposed the parton model). For his contributions to the development of quantum electrodynamics. http://en.wikipedia.org/wiki/Richard_Feynman

"For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled." Richard Feynman's famous conclusion to his report on the shuttle Challenger accident:

Buying used Crankshafts and Propellers

2013-01-30T14:10:00.000-08:00

When I had the engine overhaul shop two items that I would never purchase used were crankshafts and propellers. The reason being is that without a service history I could never be sure that the item did not have a hidden and dangerous repair. Here is a link for an interesting accident report that discusses these two issues.

Part of the reason was some of the games people played. A crankshaft that was bent way past limits would be "pre-straightened" and then brought to my shop for inspection and "yellow-tag". This occured multiple times both when I had a propeller shop and a engine overhaul shop. If I needed a crankshaft better to purchase an entire engine with logs, take the crankshaft and part out the rest. Don't be nieve.

Design Intent

2013-01-12T08:54:00.000-08:00

Failure to meet Design Intent is a good description to use for parts that just don't function as intended. The FAA defines "design intent" as part material, geometry, and material surface condition that delivers the form, fit and function required by the part design to meet the service life of the part. Design intent is recognized as including more than those requirements noted on the part drawing or quality control document. If a part meets its "design intent" then the part meets its definition.^1.

Example: When the Beech Sundowner was introduced it had a nasty habit of the engine quitting on short final when the pilot closed the throttle. We, with the help of the carburetor manufacturer, re-jetted the carburetor to correct the problem. As this was a new aircraft we submitted a warranty to Lycoming who denied the warranty as the carburetor was not defective as it was in conformance with the engineering drawing and therefore nothing was wrong with it!

The carburetor was obviously defective since it did not meet the "design intent". The concept of "Design Intent" is a powerful concept that reminds everyone that the part must function properly else it is not safe. We offered to take the engineer for a ride to demonstrate the engine failure on short final but received no response!

Other examples:

Valve sticking
Spark plug fouling

Accidents where design intent became an issue:

United Flight 585
Lockheed C-130A, N130HP

Caution is advised for maintenance when working on aircraft or components that are being operated outside of the original design intent. Existing maintenance processes may not be adequate. see N130HP.

We have been discussing "Design Intent" from the perspective of failure to perform as expected or required. A subcategory is anomaly.

Anomaly

Anomaly: If you want to place the blame on the pilot. Example "...distracted by the flap anomaly"
Failure: If you want to place blame on the device: Example "a failure of the left outboard foreflap"

Anomaly when applied to surface conditions:
An abnormal surface condition with chemical or physical properties that do not meet design intent. A similar expression is "surface flaw", although flaw implies that the part is not in conformity with the engineering drawing, while "anomaly" as defined by the FAA includes "design intent" such that the part may be in conformance with the engineering drawing, in other words manufactured correctly, but have a surface condition that does not allow the part to function as intended. Anomaly implies a condition left over from manufacturing and not a material degradation, such as corrosion pitting.

Safety Concerns

Manufacturers sometimes approach a design intent failure by not admitting the failure in the existing part but by offering a new and improved replacement part. This tactic reduces safety and places passengers at risk as it implies that the existing product functions with design intent.

Not fully disclosing known risks in EXISTING parts is in itself a safety hazard. "New and Improved" works well in the cereal industry but should be approached with some caution in the aircraft industry.You may prefer the original cereal rather than the new and improved one. The customer may not want to replace a "perfectly good" component with the "new and improved" one if they feel the existing part functions to their satisfaction -- not knowing that it has a design intent flaw.

1. Failure to achieve design intent is an example of "unk-unk". This slang engineering term, thought to have originated at Lockheed, is defined as " something, such as a problem, that has not been and could not have been imagined or anticipated; an unknown unknown."

What we know and what we don't know was best expressed by United States Secetary of Defense, Donald Rumsfield "There are known knowns; there are things we know we know.We also know there are known unknowns; that is to say, we know there are some things we do not know. But there are also unknown unknowns – the ones we don’t know we don’t know."

Why Things Break - Getting to know Fracture Energy

2012-12-16T09:28:00.000-08:00

Metallic structures break not because they are weak (low tensile strength), but because they aren't tough enough. The "high-strength Grade 8" bolt fails not by forces and stress, but by way of energy--and often at a very low force! As mechanic's repairing structures, strength tells us little about structural integrity. The engineer designs the structure to be strong enough to support the loads- the mechanic makes the structure endure.

Even the smallest scratch or crack can locally intensify stress beyond the materials ultimate strength (Inglis equation for stress intensity). There is no advantage in using a bolt that is stronger than it needs to be to support the load if a scratch or corrosion pit can easily concentrate stress beyond even its ultimate strength -- in fact it is often disadvantageous. Since every structure has cracks and scratches, there is something else besides strength that holds the structure together. Griffith showed us what happens when the local stress from a stress riser (corrosion pit) exceeds the material's strength; and it has nothing to do with strength but something called fracture energy.

What is easier to break, a piece of glass or a piece of plywood? Drop the glass and it shatters, drop the plywood and it doesn't. Plywood is tough; glass is not. Glass cutters work on glass by making a scratch across the glass surface, why doesn't this work on plywood? Because it takes a lot more fracture energy to break plywood than it does to break glass.

Glass work of fracture 1-10 joule per square meter
Plywood work of fracture 10,000 joules per square meter

Mild Steel work of fracture 100,000 to 1,000,000 joules per square meter
High Tensile Steel work of fracture 10,000 joules per square meter

Glass and plywood need to be strong (and stiff) enough to function in a structure (with a large safety factor), but maintaining the structure to assure that it doesn't fail often has little to do with strength and everything to do with controlling energy.

For metallic structures:

as strength and hardness increases, energy required to break them decreases,
as stress increases, stress corrosion cracking potential increases,
as strength increases, hydrogen embrittlement potential increases,
as strength increases, the ability to detect cracks before complete failure decreases.

How does this knowledge help the mechanic? In three ways:¹

Choose parts that have both adequate strength to support the load and also that require a high energy to break them.
Carefully inspect for scratches cracks, corrosion pits structures that require low energy to break them (high strength steels and aluminum alloys).
As long as that high-strength metallic part remains in service it must be protected from hydrogen embrittlement. The higher the strength, the more protection. Protection is by preventing corrosion and preventing chemical (particular acid) exposure.

Examples of applying our energy knowledge:

A high-tensile strength bolt is used on a trailer hitch (exposed to road salt). Since stress corrosion cracking (SCC) increases with tensile load, we might want to use a lower tightening torque than what standard bolt torque charts show. Tighten the bolt to satisfy the needs of the joint, add in a safety factor, but no tighter.This lowers the tensile stress and reduces the susceptibility to SCC. You might be better served by using a lower strength but higher work of fracture bolt if the requirements of the joint allow for it.

Cleaning the inside of your Chinook Helicopter (or underside of your new Jeep) with a household chemical cleaner exposes high-strength aluminum and steel parts to hydrogen embrittlement. Better to use a cleaner that has been tested for hydrogen embrittlement and approved by a major aircraft manufacturer.² The army lost on of their Chinook Helicopters and had to decontaminate the rest of the fleet by cleaning the insides of their Chinook Helicopters with Simple Green cleaner.³

Scratches and corrosion is more serious on high-strength metallic parts. Principle structural components carrying heavy loads and made of light-weight aluminum or steel alloys are fragile structures. Protect from scratches and corrosion. Strength in the presence of corrosion is only temporary.

Another consideration often overlooked is that high tensile strength alloys are more difficult to manufacturer. The possibility of quality and process errors increase with strength. What is labeled as "high-strength" might not be. Hydrogen embrittlement becomes increasingly more difficult to control during the manufacturing process.
FAA Advisory-Circular-AC43-205
Extreme Simple-Green is the aircraft version which meets Boeing-specification-D6-17487P and is designed to be safe on aircraft aluminum.

Additional Reading

Metal Fatigue, Cracks, and Turbo Mallards
Crack Detection Using the Unaided Eye
AD2009-16-03 SAP Cylinder Cracking
Stainless Steel Stress Corrosion Cracking - Primer for Aircraft Mechanics

Simple Spring Maintenance for Mechanics

2012-12-08T08:23:00.000-08:00

A bolt in tension acts as a spring

A spring is an elastic body designed to store energy when deflected. A tightened bolt is a spring. A cylinder hold-down stud is also a spring. Springs are "machine elements."

Springs are springy (stiffness) for two reasons:
1. Geometry
2. Elastic modulus

The loading a spring can take before bending or breaking is affected by:
1. The material's yield (bending) point or elastic limit. Anneal the spring by over-heating it and you reduce the spring material's yield point so it bends at a lower load. It's just as springy but over a much reduced range of load. A spring damaged by exceeding the yield point is called "stress relaxation", "creep", "load loss".
2. Material's fatigue strength or ability to work over many cycles. Often valve springs are shot peened to increase fatigue strength.
3. Hydrogen embrittlement. Early spring fracture for no apparent reason.

Impulse coupling spring

The mechanic protect the spring by:

preventing excessive heat¹ and,
preventing anything that might reduce fatigue strength. Fatigue strength is reduced primary by corrosion causing pits that act as crack initiation points.

The mechanic can describe a spring failure as:

Loaded beyond its yield point (spring collapsed or creep) possibly due to excessive heat lowering the yield point.
Fatigue fracture (it broke).
Hydrogen Embrittlement (manufacturing deficiency -- poor quality control of the first order)

Hydrogen embrittlement fracture in shroud spring

Excessive Heat
1. 475 F. can be considered as the maximum temperature for aircraft engine valve springs. Consider that cylinder head maximum is typically 500 degrees F. and this temperature is taken between the cold (intake) and hot (exhaust) portion of the cylinder. At 500 F for CHT it is conceivable that the exhaust valve spring could be at or above this temperature. Above 475 a couple of changes start to occur in the valve spring.

Fatigue life is reduced as the shot peened compressive residual stress is relieved.
Temper strength is reduced above 475 degrees.

From experience over-heated valve springs are often dark and crispy from oxidized oil. They also have collapsed meaning that their free length is reduced.

Caution when using magnets around machinery

2012-11-24T07:36:00.000-08:00

A magnet is a handy tool for the aircraft mechanic. There are a few cautions when using it because a magnet will magnetize iron that it comes in contact with. Magnetized parts in an engine will collect "fuzz" or small particles of metal. These particles may interfere with the proper operation of the the component or increase wear. You wouldn't want bits of metal attracted to the camshaft lobes! Two classic items that are damaged by a magnetic field are hydraulic lifters and impulse couplings.

Hydraulic Lifter

Hydraulic lifters: for example, have very tight sliding clearance between the piston and the body. Particles attracted to the lifter can bind the piston thereby preventing proper operation.

A magnetized hydraulic lifter interferes with the operation of the check valve. Whenever we had a hydraulic lifter who's bleed-down was not working properly we would send it though the Magnaflux Demag Coil and often this would restore its function. We suspect that they became magnetized when Mechanics used a magnet to pull them out of the Lycoming engine. The hydraulic lifter became magnetized as soon as the mechanic touched it with his magnet.

Magneto impulse coupling

Impulse Couplings:

At low cranking speeds a weak spring and gravity interact to position the flyweight so that it engages the stop pin and retards the spark for engine starting. Magnetized flyweights can overpower the weak spring and gravity and prevent proper operation. The field fix might be to overhaul the magneto or replace the impulse coupling when a simple demag is all that is required.

Irrational Positive Bias in Aircraft Maintenance

2012-11-20T11:38:00.000-08:00

In Decision Making - Bias I discussed how our unconscious mind injects bias into our decision making. Decisions what we think are conscious, rational decisions are made mostly by our unconscious bias mind ^1.; But what direction is our unconscious mind leading us toward?

"The brain has evolved to be an optimist" ^1.

Optimism and Confidence are necessary for Success. If you don't have optimism and confidence you won't be able to do it. Whether it's training for a sporting event, going to school, or setting life's goals, if you don't have confidence you won't work towards the goal and you won't succeed. The unconscious positive bias is how we survive as a species. Successful people have lots of optimism and confidence.

People will seek the most optimist answer

"Their just stress relief cracks" Chalk Airlines accident
One study found that most general aviation pilots believe they are less likely than other pilots to experience an aircraft accident (Wichman & Ball, 1983).
95% of pilots estimate their chance of being in an accident at a rate that is less than reality (O'Hare , 1990).
Researchers Dale Wilson and Marte Fallshore found for the specific accident scenario of VFR flight into IFR conditions, pilots were overly optimistic regarding their chances of experiencing such an accident and were also overconfident in their ability to avoid or successfully fly out of IMC (Wilson & Fallshore, 2001)²

Irrational Positive Bias

Tali Sharot in her book A tour of the irrationally positive brain, calls optimism bias "irrationally positive". Although we believe we are making a rational decision, it's often irrationally positive.

Examples of irrationally positive bias:

Indian Airlines Flight 440 crash found to be crew error in letting the aircraft descend below glide-path.

"... the control tower had discouraged the pilots from landing due to the density of the fog at Smolensk airport at the time." Tupolev-154 2010 Air Force. Ninety-six people died.

"It'll be OK, I'm in a hurry and gotta go"

"It flew in so it will fly out"

"It's been that way for 2 months"

"It doesn't look too bad to me"

"and the more they flew the more they demonstrated that the problem had no consequences."

When operating or maintaining aircraft, an irrationally positive bias needs an external control system to avoid failure.

How do we limit Irrational Optimism Bias from aircraft maintenance?

Standards
Culture

Standards: Pilots are familiar with "minimum descent altitude". Even if we "think" we can safely go below the minimum descent altitude. We are instructed not too and,
Culture: we know of stories of those who thought they could and failed. Dr. Tali Sharo offers a similar answer where we develop "plans and rules to protect ourselves from unrealistic optimism."^3.

'If you want to fix a problem, you can’t just fire the responsible person. You have to fix the organization, or else the next person to take the job will just experience the same pressures. Like Columbia after Challenger, the harmful behavior persists." Interview with Diane Vaughan about Space Shuttle Challenger Accident.

For the aircraft mechanic, standards are maintenance regulations, rules, manuals, and Acceptable Methods and Practices. If a part is worn beyond service limit then it's not airworthy and must be removed from service. Our mind might believe that the part can still function with that amount of wear (the Chalk mechanics thought that the airplane could fly with cracks in the wing); indeed, our boss or aircraft owner might insist that the part can continue to function safely; but that is not the issue -- the standard says we must not use it. The standard protects us from our irrational optimism bias.

When not working to standards, one must be "on-guard" and consciously aware of how positive bias might lead us to making the wrong decisions. New mechanics, or non-mechanics working on aircraft, may lack knowledge of the standards and culture and therefore have not developed a respect for the control mechanisms necessary to control irrational bias.

Successful People "bolt people think their fate is almost entirely in their own hands."
The first urban legend I heard as a pilot was the high accident rate of doctors in aviation. I don't know if this is true or not but it was a widely held belief at the time. Over achievers must have confidence to achieve, and their achievements give them even more confidence; their decisions have proven to be correct. But that confidence, when used to evaluate airworthiness, might lead them into danger. It's not just doctors. Successful people have confidence-- when mixed with ego it becomes a dangerous mix. Call it an unhealthy dose of "Irrational Positive Bias."⁴

NASA's managers had more confidence than the engineers on the failed launch of Challenger. They knew the standards of launch (temperature too cold). Yet they launched over the objections of their engineers. The decision to launch was not rational but driven by irrational positive bias.

Tribunals, Penalties and Sanctions

"Pilot Error"
"Failed to follow established procedures and regulations"
"Failed to follow the maintenance manual"

These are common phrases found in accident reports attributable to what caused the accident but aren't these the outcome or "effects" of another undiscovered (un-investigated) cause?

Possible the maintenance manual was not available? or could it be that the maintenance manual was available and understood, but the mechanic "felt" that it would be OK to do the repair in some other manner? Why weren't "established procedures and regulations" followed? Unless we know these answers then we don't know the true cause of the accident; Tribunals, penalties and sanctions do not serve as a deterrent and the whole exercise of accident investigation fails in its purpose of preventing future accidents.

Human Factors - FAA
More correctly renamed "Bureaucracy Factors" as the investigation is extended from the body of the person into the body of the organization in an inhuman attempt to re-integrate the human into the system -- further distorting the human's mental construct of reality. Instead of blaming someone we blame something. It just so happens that something usually has deeper financial pockets than someone. It is much more lucrative to fine the airline than to fine the mechanic.

Human Factors - The Optimist
"When optimists succeed they attribute it to their superior abilities. When optimists fail they attribute it to external reasons."

Normalization of Deviance

"And as they recurrently observed the problem with no consequence they got to the point that flying with the flaw was normal and acceptable. Of course, after the accident, they were shocked and horrified as they saw what they had done." Interview with Diane Vaughan about Space Shuttle Challenger Accident.

1. "Secrets of the Brain" PBS series

2."What is surprising is the effect of experience on ability biases. One would think that as experience increases, a person would gain a more realistic appraisal of their abilities. Instead, it appears that flight experience may lead to overestimates of one’s ability to both avoid and successfully fly out of IMC." OPTIMISTIC AND ABILITY BIASES IN PILOTS’ DECISIONS AND PERCEPTIONS OF RISK
REGARDING VFR FLIGHT INTO IMC

3. Tali Sharot "The Optimism Bias"

4. "Ignorance more frequently begets confidence than does knowledge." Darwin

Fluid Injection Injury

2012-11-13T17:26:00.002-08:00

Never again will you start "feeling" for leaks in a hose or tube carrying high-pressure after watching this!

Decision Making - Bias

2012-11-04T08:38:00.000-08:00

Essence of decision making whether one is a pilot, air traffic controller, or aircraft mechanic is:

Having the proper information at hand and,
being able to make the proper decision based on that information.

1. Having the proper information at hand

"A decision is thus no better than the information it is based on."

Could Chalk Airlines accident (N2969 killing all 20 people aboard) been prevented if better information were present? How many times have mechanics found the manufacturer's service department is "non- existent" or provides little information? Is the method of delivering critical information to the mechanic effective? "It's up to the mechanic" attitude toward service information doesn't lead to an effective solution nor enhance safety. Many times this is an organizational problem and responsibility.

Chalk Airlines mechanic's decisions, while with hindsight are fatally flawed, might not have been an incorrect decision based on the information present at the time. Throw in a dose of confirmation bias and the decision even becomes expected under the circumstances. This leads one to the conclusion that accidents such as this one are not a one-off event.

2. Being able to make the proper decision based on that information

Confirmation Bias recognition and management

We can have a lively discussion about what decision to make even when all relevant information is present. I don't understand why you don't agree with me and you don't understand why I don't agree with you. "People believe what they want to believe" This is Confirmation Bias. Interesting accidents caused by Confirmation Bias include Comair Flight 5191, Airbus A321-232 SK-473 and Airbus A320 UP-BWM. A more formal description of Confirmation Bias:

Confirmation bias is a phenomenon wherein decision makers have been shown to actively seek out and assign more weight to evidence that confirms their hypothesis, and ignore or underweight evidence that could disconfirm their hypothesis. As such, it can be thought of as a form of selection bias in collecting evidence.

We all have Confirmation Bias. It's not a defect but normal human behavior. Errors caused by Confirmation Bias are not solved through penalties. For the aircraft mechanic I believe there are two methods of reducing maintenance errors caused by Confirmation Bias:

You cannot inspect your own work. Often you cannot see your own mistakes, even glaring ones, because your mind has already decided that the work was done as you intended. An inspector does not have your bias (he has his own) and can better judge your work.^1.
Education and recognizing when Confirmation Bias might be present.

With 45 years in aviation I have learned that most mistakes (flying or maintenance) are not due to a lack of skill, but a lack of judgment. How often have I been told that "he is a good pilot". Fine, that's good, but as a human being can he make the right decision when the situation gets rough? For that we need to look at the soul of the person; the personality. Can he accept failure (land before his destination)? Or will his ego send him forward on a path of destruction? Its not about being good -- its about avoiding failure. Sometimes the more timid pilot is better at avoiding failure.

Part II Irrational Positive Bias in Aircraft Maintenance

Further Reading:
Metal Fatigue, Cracks, and Turbo Mallards
Cracks in Aircraft Structures

Notes:
1. Who made the mistake? The mechanic who didn't do the job properly? The inspector who failed to notice the mistake? or the employer who failed to provide an inspector?

Confirmation Bias in Reading

Sometimes we read what we expect to read and not what is written. Did you notice the extra THE above?

When to use Thread Compound

2012-10-21T09:52:00.002-07:00

Thread Compound includes items such as:
Lubricants: oil, grease, anti seize,
Locking compounds: thread locker, loc tite
Thread sealants: pipe sealant or teflon tape or other thread sealing tape.

Applying antiseize to bolt

There are four reasons one would apply a thread compound to threads:

To prevent liquid or air leakage past the threads. Example, Tapered Pipe threads (NPT)
To prevent loosening of the joint. Example a thread locking compound.
To control friction (K) for better torque control and to meet the "conditions of torque" in the tightening specification you are using.
To prevent seizing, galling, and corrosion. Example, anti seize.

Your choice of whether to apply or not is governed by which of the above items you need to control.

There are four reasons one wouldn't use a thread compound on threads

The Maintenance manual says not to use it
In areas of high vibration where a thread lubricant or antiseize might reduce friction and allow the nut to loosen.
Thread compound might contaminate system fluid.
System temperature that is outside of the threads compounds temperature range. Could be too hot or too cold.

Further Reading:
How Antiseize changes bolt torque

1 Quart less than normal for flight school

2012-10-02T11:13:00.000-07:00

John,

One of the bits of "common sense" I've heard for years is that you should run an engine with 1 quart less oil than nominally "Full", because "it will just throw that extra quart out." I've never taken verbal or confrontational exception to such a statement, but it seems counterintuitive to me that a manufacturer would specify a full oil level (say 6 quarts) knowing that the engine would purge itself of that extra quart.

So, what's your take on this bit of advice from the old-timers? I don't want to install 6 quarts of oil only to lubricate the ramp or the countryside with the superfluous quart. And I want to provide our club members with the best information available as we begin to operate this new, and expensive, engine.

I believe it is good common sense to teach students (and instructors) to make sure that the engine has the manufacturer's recommended amount of oil in the engine. There may be a time when that last quart of oil is the only quart of oil.

Teach the student to reference authoritative sources for information and be sceptical of rat chatter from the hanger. For engine operation this is the engine manufacturer's operating manual; for the airframe this is the airframe operating manual. There are numerous other resources that the manufacturer provides. Not that there is not good information out there but there is also awful information that makes it difficult to decide which is good and which is bad.

One other error I see is that a specific recommendation might be good advise for some but not all. A good example of this is engine leaning. There is not one way to lean that is correct for all engines or all pilots in all circumstances. Another example is your question about 1 quart down; 1 quart down on a 12 quart sump is a lot different than 1 quart down on a 6 quart sump. Given that an engine can develop a problem and burn easily 2-3 quarts an hour without the pilot being aware; this could lead to disaster. Beware of blanket statements -- there are always exceptions.

Students and their instructors should start out with good habits. As they become more experienced they can then explore some of the other methods.

Is recharging a magneto rotor magnet necessary?

2012-09-26T13:57:00.000-07:00

Not during routine maintenance such as a 500 hour inspection unless the rotor or coil are removed from the magneto.

I see this as two separate issues:

Assure that magneto meets tolerance and specification compliance,
Maximize magneto performance

1. The allowable tolerance might be robust enough to allow the magnet to remain within specification in the anticipated operational and service environment. Minimum coming-in speed^1. can assure that the magneto as a whole is within tolerance and indirectly that the rotor magnetism is within tolerance. In this context it might be unnecessary to re-charge rotors.

Many magnetos lack a documented service history so it is possible that the magneto rotor at some time could have been exposed to unanticipated and/or unusual situations that are outside of the manufacturer's definition of the service environment. One example might be fire or someone else attempting to re-charge the rotor with improper equipment or technique. In this context, replacing, re-charging, and other non-typical inspections are necessary to restore the rotor.

2. Proper recharging and installation can assure the magneto produces as much current as it is capable of. This can be desirable. Peak performance, rather than a range of performance within a tolerance zone might be a safety issue. Peak performance might allow the magneto fire the plug over a wider range of E-gaps than a magneto that is still somewhere within the tolerance zone but not at the top end. It might also help starting performance by producing more current across the spark plug gap.

There could be a conflict between these two issues as peak performance might fall outside the tolerance zone for some reason. i.e. manufacturer wishes to "de-rate" the magneto. This would imply a lower bound to the coming-in speed tolerance.

Removing the Coil or rotor:

Exposing the rotor to a open magnetic circuit for the first time can drop the magnetism from E to F.

Magnetic Hysteresis Curve

1. Minimum Coming-In Speed is the minimum speed at which the magneto will spark a 5 mm gap of standard design.

Why O-200 stick exhaust valves

2012-08-17T15:47:00.000-07:00

O-200 Exhaust port

I wrote this back in 2003

"Carbon build-up on the guide pushes the valve stem into the guide. If the guide is the original aluminum bronze style the guide quickly wears out. If the guide is a harder ni-resist style then the guide doesn't wear and the valve sticks..."

I also wrote this:

Warning! At the first signs of valve sticking ground the airplane. Follow the engine manufacturer's instructions for valve sticking.

Valve sticking can cause immediate and catastrophic engine failure. Sticky valves destroy your camshaft, push rods, lifter faces, etc. Solvent treatments are unsuitable in that while they take time to work your engine is being destroyed!

Broken rocker boss caused by stuck valve

A stuck valve as a much more of a safety issue than a worn valve guide. So now the industry has stuck valves on O-200 engines -- something that was very rare before the guide material was changed. Are you, the flying public, OK with this? It doesn't have to be this way. Better to fix the underlying problem so you don't have stuck valves.

Now granted the O-200 (including the C-85, O-300/GO-300) cylinder is a "little troublemaker" and requires more loving and care than cylinders on any other engine model. I used to get the comment: "its only a C-85". Yes, and if it isn't perfect it will bit you in the pocketbook. But it can be made to work without sticking valves. Guide wear to can be reduced to an acceptable rate even with the softer guide -- Heck that was the norm in the 1970's / 1980's.

Endurance Strength -- Strength Reduction due to several surface conditions

2012-08-11T08:22:00.001-07:00

The chart below shows the reduction in strength due to several surface conditions when the part is exposed to cyclic loading.

What is "endurance strength". When we think about strength it's how much force is required to fail the part -- either break or bend. Our part must be strong enough to resist the static load. However, many parts have repeated (cyclic) loads applied to them.; propellers and connecting rods are two examples).

Given enough of these cyclic loads the part can fail at a very different stress level than a one-time applied load. This point of failure when stressed repeadly is the endurance strength. We should be concerned about endurance strength for parts subject to large number of cyclic loads. The chart below shows the great reduction in endurance strength under various surface conditions.

Loss of Strength due to corrosion pitting:

The chart below is from Timkin Practical Data for Metallurgist.and shows endurance (fatigue) strength for various surface conditions. If we use the polished Lycoming connecting rod in the picture above and assume that the strength is 120 psi for a polished surface. With corrosion pitting the endurance strength drops from 70 psi to below 20 psi.

Lycoming Connecting Rod Corrosion

Read the note at the bottom!

Is it possible to use differences in CHT change when leaning to find engine baffling problems?

2012-08-02T16:19:00.001-07:00

Lycoming Cylinder Barrel Blue oxide films -- 550 F.

You lean the engine and the CHT changes as expected. But why does the CHT on some cylinders change more than others?

One reason will be that each cylinder is operating on a different point on the mixture ratio graph. Unless each cylinder is operating at the same mixture ratio, to use our method described below the data needs to be normalized so that each cylinder is at the same mixture ratio point.

Most publications show how the CHT changes as the fuel/air mixture is leaned or richened. Also of interest is why some cylinders may experience a greater or lesser temperature change than others. Extending the idea further, why might one engine produce a greater change in CHT than the same engine in a different airplane?
"A cylinder which is shielded from the air blast or improperly air cooled will have a much greater rise in temperature due to leaning of the mixture than one which is properly cooled." An old-old study N.A.C.A. Technical Note No. 388.

Take a rich running engine for example; fuel is used to cool the cylinder. We also have the airflow (and some oil cooling) that also cools the cylinder. But, as we lean the engine we take away fuel's role in cooling the cylinder. We are left with air and oil cooling. Changes in air flow has a greater impact on CHT temperatures on a lean engine.

Within some rich to lean range those cylinders (or engines) that are not as well air-cooled will experience a greater CHT change. For the first time we now have recording instruments in many cockpits that can capture this data. Can we deduce cowling or baffling problems by comparing the amount of change?

How Antiseize Changes Bolt Torque

2012-08-02T16:09:00.003-07:00

Antiseize changes the amount of clamping force produced by a set amount of applied torque.

AN6 (3/8-24) bolt tightened to 75% of bolt proof strength. Camping force: 5,788 pounds. Required torque:

Cadmium plated and dry 28.9 lb Ft.
Moly anti-seize 19.9 Lb Ft.

Caution should be exercised if you use thread lubricants other than what the manufacture specifies. Thread lubricants change the friction between the threads, and between the nut and bolt face and the flange. This results in the bolt being over-tightened if thread friction is lowered, and under-tightened if thread friction is increased. The chart shows how different types of thread lubricants change torque requirements.

In an investigation of a helicopter accident it was found that by applying anti-seize to the entire fastener instead of just to the threaded portion (as required) resulted in a 70% increase in fastener stress from 24,440 to 43,262 lbf. with the same torque applied! The increased stress resulted in the bolt breaking. It was also found that cleaning the bolt with MEK (methyl ethyl ketone) increased the stress from 24,440 to 25,706 lbf. A good reason to consult the manufacturer's torque recommendations.

source:
"Failure of bolts in helicopter main rotor drive plate assembly due to improper application of lubricant" by N. Eliaz, G. Gheorghiu, H. Sheinkopf, O. Levi, G. Shemesh, A. Mordecai, H. Artzi,
Published in Engineering-Failure-Analysis #10, 443-451, www.sciencedirect.com

Further Reading:
How to use Thread Compound

Aircraft Screw Countersink Angle

2012-08-02T15:59:00.001-07:00

Industrial 82 degree -- Aircraft 100 degree

Most aircraft countersunk screws have a 100 degree countersink angle, whereas in the US most commercial grade countersunk screws have an 82 degree angle. England uses a 90 degree countersink angle. MIL-STD-1515 "Fastener Systems for Aerospace" requires 100 degree countersink.

The 100 degree angle is preferred when fastening into soft materials as distributes the pressure over a larger area than the 82 degree screw.

Another consideration is head height. If you notice the two screws above, the 100 degree has a thinner head than the 82 degree. If your sheet thickness is thinner than the head height of the flat-head screw, then the head will penetrate past the bottom of the sheet. This will require that you also make a small countersink in the tapped part that you are attaching to. It also means that the resulting clearance hole in the sheet will be larger than the recommended one. The 100 degree can be used in thinner sheet without the head penetrating through the skin. Note, there is such a thing as a "undercut" 82 degree flat head screw, although not common, it has a shorter head for use in thin materials.

Typically countersunk fasteners are designed so that they do not fracture at the head, so they should support the full force generated in the shank. The larger considerations are the mating parts and how they take the shear and compression stresses.

In a multi-fastener group, if there is any real out-of-position error in the alignment of bolts to the countersink, some of the fastener heads will see an added bending stress (the misaligned head is being forced to the center of the countersinksink). Contrast this to a more standard head, where mis-alignment of the bolt and hole can be accommodated, with no added bending moment.

When used with a nut, there is some float and alignment is less of an issue. When used in a tapped hole in a mating part, misalignment will add to the bending stress on the head.

Reusing Locknuts

2012-08-02T15:29:00.001-07:00

MS21044 locknut comes in two styles

MS21044 Spring Beam Nut (left):

This type of nut has thin slots cut down through the top few threads with the resulting fingers bent slightly inward. At installation, the bolt springs the fingers out and the fingers grip the bolt with a prevailing torque.

MS21044 Distorted Thread Nut (right):

This type of nut has been slightly crushed at the top. (notice the marks on the side of the nut). When the round bolt reaches the oval portion of the nut it springs the nut back round. This spring action grips the bolt and adds friction that prevents loosening.

A review of research papers tends to conclude that of the two; the elastic lock nut is more re-usable than the all-metal lock nut.

Two problems with the all-metal design:

The all-metal locking mechanism rubs the threads and removes the protective plating and damages the threads. This makes the bolt more prone to corrosion and galling damage. Replacement with a new nut does not restore the damaged bolt threads.
The wearing away of metal tends to loosen the grip of the all-metal lock nut.

These problems are best illustrated by the loss of 17 lives in a Sikorsky S-92A where all-metal self-locking nuts were used in areas that required frequent removal. In this accident a titanium stud was used with a silver plated locknut. Titanium is prone to galling and the silver plating on the nut was used to prevent galling. However, as described in the accident report:

"Examination of a new stud and nut showed that galling damage developed after the first installation and that the damage became progressively more severe with repeated installation/removal cycles. Testing of the occurrence and exemplar studs and nuts showed that after 13 to 17 assembly cycles, the nut self-locking feature was significantly damaged and fragments were separating from the crests of the threads" Report A09A0016

The elastic lock nut is more friendly to the threads, locks out moisture and prevents corrosion; and the nylon deforms rather than gouges. Except in areas of high-temperature, the elastic lock nut is generally preferred.

Recommendations for use per FAA AC43.13-1B Acceptable Methods and Practices:

Elastic lock nuts are not to be installed in areas exceeding 250 degrees F.
Do not reuse elastic lock nuts if the nut cannot meet the minimum prevailing torque values shown in the chart
Do not use self-locking nuts on parts subject to rotation
Do not use self-locking nuts where the loose nut, bolt, or washer may fall or be drawn into the engine air intake scoop.
Do not use self-locking nuts to attach access panels, doors, or any parts that are routinely disassembled before or after each flight.

Several aircraft accidents (Canadian TSB Report A97O0055) and aircraft control problems (C-130 aileron) and the loss of Bell-206-L-1 helicopter, N2138Y NYC01LA088, and Sundance Helicopter caused by lock nuts coming loose have challenged the idea that any style lock nut can be reused.

The idea of inspecting a lock-nut for minimum prevailing torque sounds good on paper, but as illustrated in the Canadian accident, it may not be a good practice to use in the field. In this accident the engine manufacturer's maintenance manual (ROTAX), stated: "self-locking nuts must be replaced with new items after removal in the event the friction torque has diminished."

When the Transportation Safety Board did their own tests on the M8 lock nut PN 942-035 used on the exhaust, they found that friction torque diminishes each time the M8 locknut is installed and removed. In this accident the loose lock nuts created a situation that led to an accident.

The usual method a mechanic would use to check for "diminished friction torque" is to see if the nut would unscrew by hand. If it does, then it is no good and is replaced. If it cannot be unscrewed by hand then it is OK. THIS METHOD DOESN'T WORK as they found out in the loss of Bell 206 N2138Y.

The "finger test" of the re usability of a locknut is not valid. Informal tests done by the author show that it takes 4 to 5 inch-pounds of torque to prevent a mechanic from screwing a 3/8-16 AN nut down by hand. For a 3/8-16 elastic stop nut, the minimum breakaway torque is 12 inch-pounds and a new nut produces 80 inch-pounds according to the Acceptable Minimum Limits for self-locking nuts per MIL-N-25027 (in actual tests 102 inch-pounds were required to free run the nut).

Notice that a new nut in our example produces 80 inch pounds of prevailing (friction) torque when new but can degrade down to 12 inch pounds before rejection! Could it be that in some high-vibration applications (read helicopter) that 12 inch pounds is not sufficient to hold the nut in place?

Maintenance Flow Issues in re-using and inspecting locknuts:
In the Sundance Eurocopter AS350 Accident, the loss of one locknut caused the loss of control of the helicopter. Instead of using a new nut, both the FAA, NTSB believe that the nut should only be replaced if it fails inspection. Why inspect a 50 cent nut if the helicopter and 5 people's life depends on it? Especially if these "nut inspections" have resulted in previous accidents (Bell 206 N2138Y).

Nut inspection introduces not only an extra inspection requirement but a bottleneck to workflow. Most likely a proper inspection is more expensive than the nut itself. If the nut fails inspection then you have the cost and time of the inspection and still have to procure a locknut. The tendency in human nature is to get the job done and the nut is probably not all that bad. However, if it was in the mechanic's mind that critical locknuts are not re-used then a new locknut is made available as a standard item for replacement. Given the critical nature and the saving of an inspection step -- why would anyone suggest that the mechanic should inspect the darn thing?

The NTSB's "Safety issues" regarding Sundance Eurocopter AS350 with 5 lost lives is:

Reuse of degraded self-locking nuts
Maintenance personnel fatigue
Lack of work cards with delineated steps
Lack of human factors training for maintenance personnel

I have a better safety suggestion; "Don't reuse critical locknuts" Simple and to-the-point.

Air Force T.O. 1-1A-8 now states: "New self-locking nuts shall be used each time components are installed in critical areas throughout the aerospace vehicle." This seems a more practical policy given the low cost of a lock-nut. And from the UK: Fasteners with a fibre or nylon friction element should only be used once, and must not be used in locations where all-metal stiffnuts are specified. All-metal stiffnuts should not be re-used in locations vital to aircraft safety (e.g. control runs) but may be re-used in other locations providing the locking quality remains satisfactory. CAP 562

If the NTSB and FAA had made this a requirement after the loss of N2138Y, then the Sundance AS350 accident need not have happened. Shall we add another Safety Issue:

The failure of the NTSB and FAA to correct known industry safety issues.

NAS679 Nut

NAS 679 Low Height Lock Nut

In 1969, Cessna issued Service Letter SE69-28 after finding that NAS679 nuts, size 10-32 were cracking due to "heat treatment embrittlement". SE69-28 indicated that Cessna had discontinued their use and called for their replacement in certain critical applications.

NAS679 nuts, other than size 10-32 were used extensively by Cessna until 1980 when they were replaced by MS21042 nuts. While the company amended its parts catalog for the affected aircraft to reflect the change, Cessna did not specifically indicate that the NAS679 nuts had been superceded and there was no service information requiring replacement of NAS679 nuts, other than those called out in SE69-28.

There has been other reports of these nuts cracking in the Cessna Wing Carry Through Spar Assemblies. For this reason, were allowed and applicable, MS21042 may be a better choice than NAS679
----
Service Difficulty Reports (SDR) published stated that the 16 nuts were MS21042L6, although when investigated by the Transportation Safety Board of Canada it was found that the cracked nuts were actually NAS679. Reference AAC6-54 dated 2/95

Does the bolt loosen slightly after you torque the bolt?

2012-08-02T15:13:00.000-07:00

25 miles out at sea and cylinder blows off but engine runs long enough to get him to the beach!

Yes, by a process called "embedment relaxation", fastener preload is reduced. As each surface is pressed together, the high spots are crushed and deformed to form a surface capable of supporting the load.

Surface Finish:There are multiple surfaces in the joint:

1. between the head of the bolt and the washer,
2. between the bolt washer and faying surface (faying surface is the surface of the object being fastened together),
3. between the two faying surfaces,
4. between the faying surface and nut washer, and
5. between the nut washer and the nut face.

Each of these surfaces squeeze together. Any paint, sealants, nicks, or alignment errors are gradually crushed down to support the load. Bolt threads also embed. Threads are pulled in shear, slightly increasing the thread pitch. Nut threads are compressed and lose a little pitch. As embedment occurs, the surfaces press further together and reduce the bolt's clamping force (preload). Note thate relaxation occurs without any off-rotation of the nut.

Creep:Any gaskets, sealant, paint, or lubricants, between the surfaces will extrude from between the surfaces. One reason not to use high poission's ratio (0.5) products, such as rubbers, in tension joints.

How much preload is lost?
Under optimum joint conditions in a lab one can expect between 1 and 11 percent. Obviously then, in the field, more clamping force can be lost. Lockheed did a lab test, Report No. LR 25049 where they tested 1 inch by 12 UNJF thread size L-1101 engine pylon bolts. They used lab conditions, with hardened steel bushings. In a static joint with no load fluctuations, between 1 and 11% preload was lost. The greatest loss occurred in the first eight hours after installation.

How this effects the mechanic?It's up to engineering to included embedment relaxation in the joint analysis and torque recommendation. The mechanic need not compensate for embedment relaxation but he does need to follow the torque recommendation (conditions of torque) carefully. Surfaces need to be clean and undamaged.

It's also considered good technique to torque slowly to allow the surfaces to adjust to one another. Years ago when assembling engines we used a torque and hold technique where the full torque was applied and then held for 20 seconds to allow the joint to stabilize. One potential problem with snap style torque wrenches is that torque is immediately released at the torque point. If the joint hasn't stabilized and is still moving together then it hasn't been fully tightened; another reason to apply torque slowly.

What about multiple clicks of the torque wrench?This technique is widely used by mechanics but does introduce some questions about its effectiveness. Once the nut stops rotating then it might take a lot more torque to break the nut free and get it moving again. The nut may not rotate not because the joint is fully loaded but because there is too much friction to break the nut loose.

Just the opposite has been shown to occur where there was plenty of lubricant applied to the thread and nut surfaces so the nut was free to rotate. Subsequent application of click torque progressively tightened the joint far past the proper amount. This occurs because when you apply torque you apply a axial tension to the bolt - you stretch it, and you apply torsional tension to the bolt - you twist it. When you're wrench snaps and releases the twisting force, the bolt untwists. This untwisting occurs in the portion of the bolt where the torsional stress is highest - in the threads and causes a slight tightening of the nut onto the bolt. In effect, twisting stress is turned into axial strain.

Preventing vibration induced bolt loosening

2012-08-02T15:07:00.000-07:00

MS21045 Lock Nut comes in two styles -- spring beam (left) or elliptical beam (right)

Items that increase loosening resistance:
1. Tighter hole tolerances.
2. Increased bolt preload (tension); increasing bolt diameter, yield strength, preload.
3. Smaller thread helix angle. Fine thread bolts. Fine thread lock nut
4. Minimize joint relaxation
5. Maintain large friction forces - high preload, don't lubricate threads and mating surface.
6. Increased surface contact.
7. Increase thread contact area - Class 3 (aerospace fasteners) fit instead of Class 2 (commercial) fasteners.
8. Galling susceptible materials such as uncoated stainless steel.
9. Corrosion susceptible materials such as uncoated and unlubricated steel
10. Minimize transverse movement of faying surfaces - increase joint friction by not lubricating joint surfaces.
11. Increased contact between mating surfaces - smooth surfaces.
12. Nylon lock nut instead of all-metal lock nut to dampen resonant vibrations.
13. Aircraft lock nut instead of commercial grade lock nut.

References:
Experimental Analysis of Thread Movement in Bolted Connections Due to Vibrations, Auburn University, Prepared for George C. Marshall Space Flight Center Research Project NAS8-39131, March 1995

Aircraft Hardware compared to Industrial Hardware

2012-08-02T12:06:00.000-07:00

Hi John,I have your Mechanics Toolbox and found it very good and helpful. But I have been asked about what the SAE equivalents for Aircraft bolts and can't find the answer in your book or on the internet. I have a friend that wants to save money on prop bolts. I told him that aircraft bolts are manufactured to a stricter standard than SAE bolts. I also asked him, "Why take a chance?" However it would be nice to have a more specific answer. Do you have any tables showing what is equivalent? Also any reasons for using standard bolts or not would be appreciated.

-------------reply----------------

There is no equivalency between standards organizations as the products are made with different objectives in mind. A quick example, much commercial hardware is installed in static structures and has a static load. Aircraft hardware is installed in dynamic structures and is subject to dynamic loads - An aircraft bolt, "propeller" bolts" for example, will feel millions of stress cycles whereas a bolt used on a building will have constant stress and no load reversals. An important design consideration for aircraft hardware is fatigue strength - or strength when subject to millions of stress reversals. Pick fasteners for strength AND toughness.

Increased fatigue strength
Bolt Thread. One method of achieving fatigue strength is to round out the thread root so there is less of a notch. This is called the "J" thread and is used on almost all modern aircraft hardware (UNJF for example). Commercial grade bolt thread form (UNF or more technically UNRF).

The J thread increases fatigue strength by 40%.
The J thread increases shear strength by 10% (thread tensile stress area of 110.765 compared to 103.20 mm2 for 1/2-20 thread.

J thread has a more rounded root to increase fatigue strength.

If you just looked at "bolt strength" (usually ultimate or yield strength) it is unchanged. Using our propeller bolt for example, we are very interested in fatigue strength. The J thread is a rolled thread. Rolling the thread increases thread strength. Commercial hardware may use rolled or cut threads.More modern NAS hardware has other features to increase fatigue strength such as greater radius between the shank and head - which commercial or SAE hardware will not have.

More thread engagement for added strength
Most aircraft hardware has a class 3 thread engagement fit (UNF-3B) while most commercial hardware has a class 2 thread engagement fit (UNF-2B).

Better corrosion performance
Corrosion control and material compatibility is extremely important in aircraft hardware. Aircraft hardware most not be galvanic to aluminum. This is why lots of cadmium plated steel is used and almost no stainless steel bolts in aircraft. Along with corrosion protection, aircraft fasteners are often oven baked at longer temperatures to prevent hydrogen embrittlement. High strength bolts are particularly sensitive to corrosion and hydrogen embrittlement.

Different Torque values
Cadmium plating has excellent lubricity (low K factor) so that for a given amount of torque from your torque wrench, more tension is produced than would be if the plating were zinc. If you were to substitute a different bolt with different plating one would need to change the torque wrench setting. How would one know what torque value to use?

Increased thread shear strength
Bolt shear strength at the threads is greater. Another consideration is that much hardware store hardware is manufactured to a class 2 thread fit whereas aircraft is manufactured to a class 3 thread fit. Class 3 provides increased thread shear strength (approximately 10% increase in shear area).

Better alloy steel
Steel alloy's come in different grades - Generally, aircraft hardware uses "aircraft grade" alloy whereas commercial hardware uses "commercial grade" alloy. Not to say that "Aircraft" is superior to "SAE" as each is designed for a specific industries' requirements.

Why Metal Gaskets Leak

2012-07-30T07:47:00.001-07:00

A gasketed joint leaks when imperfections in the joint are beyond the capabilities of the gasket, in which case replacing the gasket may not fix the problem. Joint problems are often revealed in the gasket.

Gaskets tell a story about sealing surfaces, but all too often they are discarded without giving them a chance to tell their story. Whenever joints come together the contact (faying) surfaces conform and the joint functions properly; or joints can leak, move about, loosen, wear, and cause problems. Gaskets, and washers acting as gaskets, seal by embeding into imperfections in the faying surfaces -- to a point. You can prevent problems by inspecting gaskets and surfaces to see how they are doing.

Metal gaskets provide clues to the condition of the sealing surface

Spark Plug Gasket -- leakage at upper right corner

The spark plug gasket above was leaking at the upper right hand corner. I would suspect that the sealing surface on the cylinder is damaged and leakage may continue even with a new gasket. Close monitoring for leakage after a test run and for the first few operations would be prudent. A leaking spark plug gasket interrupts the heat path causing the spark plug to operate hotter and possibly cause pre-ignition. Gas leakage also erodes the boss resulting in even more leakage if not corrected.

Gaskets tell you if they are overtightened

Spark Plug Gasket - Deformed from excessive torque.

Spark plug gasket cupping from overtightening. Place gasket on flat surface and push on edge to see if it is flat or warped. Axial displacements around the inside edge

Crushed washer -- too much torque or washer material is too weak

Typical Joint Imperfections:

Silicone rubber used on exhaust gasket to embed into imperfections

Improper surface finish -- too rough. Lapping surfaces might help
Damaged surfaces -- scrapes, gouges, erosion, pitting, etc.
Faying surfaces not pulled tight leaving gap.
Faying surfaces not parallel -- torque cannot pull surfaces together leaving a gap
Foreign material between surfaces

Trouble Shooting - The Question to Ask Before you Begin

2012-07-24T08:26:00.000-07:00

Is the equipment being operated in accordance with its design limitations? Is the equipment being operated in ignorance to, or in disregard to the manner in which the manufacturer intended? Equipment cannot be expected to operate normally when operated abnormally.

Manufacturer's establish operating limitations based on design limitations, testing, and experience. Modifications to the equipment either by physical changes or operation means that the alteration wasn't anticipated or accounted for in the original design and in the operational limitations of the equipment. Whether the equipment will operate properly or safely while still retaining an acceptable safety factor is a complex engineering question outside of the expertise of the operator and mechanic.

The hypothesis that the modification is good can be confirmed time and time again -- but it will never be proven. You may not realize what the possible weakness in the modification is until conditions occur in which the weakness is revealed. Too often modifications are based on trying it out and seeing how it works. Whether it's a new propeller, an engine modification, or airframe modification, the criteria for sucess is often the flight experience of others. But, successful flight experience doesn't guarantee a safe or problem free design. Tacoma-Narrows-bridge, walkways of the Kansas-City-Hyatt-Regency Hotel, Mianus-River-bridge on I-95, British-Comet airplane, Japan Airlines Flight 123, China-Airlines-flight-CI-611 all provided problem-free service for decades until sudden and catastrophic failure due to serious design or maintenance flaws.

Solutions to operation and/or design induced problems are outside of the mechanic's charter.

How to Diagnose and Fix Rocker Shaft and Bushing Wear

2012-07-23T13:28:00.000-07:00

A good procedure to follow when analyzing failures is to question everything you are told

In a well engineered system ("you have a problem but know-body else does") focus on alignment, abrasives in the system, and shaft surface finish.

Lycoming Rocker Arm - Notice bushing is thinner at top than at bottom

Lubrication Primer

When the rocker arm is stationary, lubricant is squished out from between the shaft and bushing. Metal-to-metal contact occurs along a line where the shaft contacts the bushing.^1.This is called "boundary lubrication".

As the rocker arm moves, oil is forced between the shaft and bushing separating the surfaces.This fluid film prevents metal-to-metal contact. You can see the progression from boundary lubricant to fluid film (hydrodynamic lubrication) in the bushing wear pattern. At first the contact is full but progresses to a feathered edge. For the lowest wear, we want to maximize fluid film lubrication and minimize boundary lubrication.

The gap or clearance between the shaft and bushing allows space for the fluid film. As will be discussed below, if the shaft wears into the bushing the effective clearance decreases. Without clearance, the oil cannot penetrate between the shaft and bushing even though plenty of oil might be covering the shaft and rocker arm. Without an oil film the boundary lubrication region increases resulting in increased contact between shaft and bushing. This increases wear.

Principles:

In a boundary lubrication system where there is direct contact between the shaft and bushing, shaft surface finish is critical. Any shaft roughness cuts into the bushing. Replace any rough shafts regardless of dimensional limits.
Locking down the shaft prevents it from rotating in the boss (early engineers knew better) if the friction between the shaft and rocker arm are excessive. A floating shaft rotates where the friction is lowest, which can be at the boss if for some reason (friction, abrasives, seizing) occurs between the rocker arm bushing and shaft. Friction and wear are reduced in a floating shaft configuration.
When the shaft wears into the bushing, clearance is reduced as the effective bushing diameter conforms to the shaft diameter (shaft falls into the divot), contact area between the shaft and bushing increases. This increases the area of boundary lubricant and reduces fluid film lubrication. It also prevents heat transfer and washing away of abrasive particles.The lubricant film is further reduced.

Galling on shaft

Shaft Wear - Mostly

The hardened shaft can wear without the bushing wearing. This is typically caused by abrasive particles embedding themselves into the bushing and then scoring the shaft. This is called "3-body wear" as a 3rd body, the particle is required. Particles can be dirt or contamination or burrs broken from a rough shaft.
Incompatible materials. Example is the use of Aluminum Bronze bushing with rough surface finish. 2-body wear.
Adhesive wear from high-spots creating enough friction and heat to cause micro-welding. Relates to surface finish and oil supply.

Bushing Wear - Mostly

Excessive shaft pressure - alignment and fit.
Shaft surface finish too rough - file cutting
Incompatible materials - entire population of bushings affected. Substitution of aluminum bronze for leaded bronze caused fuel pump galling, shaft breakage and the loss of all aboard Cessna-404 VH-ANV.
Too little clearance between shaft and bushing - binding and seizing.
No lubricant.
Adhesive wear from high-spots creating enough friction and heat to cause micro-welding. Relates to surface finish and oil supply.

Course of Action

Fix the damage. Replace both shaft and bushing.
Correct the problem. Inspect for contamination, inspect alignments, fits, and clearances, do a manufacturer literature research and contact the manufacturer service department for suggestions. Might be a service bulletin correction. Check for proper parts and proper repair process.
Monitor wear for the next few hours.

1. The theoretical contact between one cylinder (shaft) nested inside another cylinder (bushing) is a line but in practice the applied load causes elastic deformation of the bushing so that the contact area is rectangular. This creates shear forces in the bearing material. Two methods of decreasing applied loads and shear forces is to increase the contact area by reducing shaft clearance and by reducing friction by using a smoother shaft finish. Reducing clearance is not without problems as this reduces the oil film and consideration of thermal expansion might result in shaft binding.